Ovel Design of Charging Infrastructure For Electrified Bus Transit Network Systems

NOVEL DESIGN OF CHARGING
INFRASTRUCTURE FOR ELECTRIFIED BUS
TRANSIT NETWORK SYSTEMS
Adedayo Ayodeji Asaolu
Institute of Energy and Environment
Department of Electronic and Electrical Engineering
University of Strathclyde
This thesis is submitted for the degree of Doctor of Philosophy
August 2022
To my Beloved parents Afolabi and Victoria Asaolu
To my wife Mary and my children Enoch, Nathaniel and Joana
i
“Climate change is a global problem, and transport systems' electrification is
considered a critical component of global climate solutions.”
ii
DECLARATION
This thesis is the outcome of the author's original research work. It has not previously
been submitted for review to a university or other institution to obtain a degree. Under
the terms of the United Kingdom Copyright Acts, as specified by the University of
Strathclyde Regulation 3.50, the author owns the copyright to this study. Always give
appropriate credit for the use of any information found in or extracted from this thesis.
Signed:
Date: August 02, 2022
iii
Novel design of charging infrastructure for electrified bus transit network systems-Adedayo Asaolu -06/2022
ABSTRACT
Developed countries globally are designing and implementing strategies to reduce
long-term greenhouse gas emissions and increase resilience to climate change, with
electric vehicle (EV) adoption as a priority. In developing countries, the electrification of
public transport and fleet vehicles will likely progress ahead of privately owned ones as
these can be more readily regulated to encourage adoption. However, it is likely that a
mixture of public and private charging infrastructure would be deployed but that 'hub-
based solutions supported by renewable energy mini-grids present an attractive option.
This work develops a mathematical modelling toolkit to model the BRT system
electrification modelling. The mathematical model that manages the routing of BEBs is
extended to a longitudinal dynamic model that describes transit buses' energy
consumption rate. To support this, a mathematical optimisation process for determining
the critical design variables of allocation of the conductive and inductive chargers is also
presented. The proposed mathematical analysis is based on designing and modelling
multi-terminal (hub) charging infrastructure to cater to the bus transit network system's
charging demand.
This novel charging optimisation tool is developed to demonstrate how multiple

charging strategies can be employed for a fleet of BEBs to study essential factors of the
transit system: such as the trade-offs between alternative charger designs, charger
locations, battery sizes, and cost.
In this scenario, a series of contextualised simulation studies have been undertaken that
focus primarily on the current situation in Lagos, Nigeria. The simulation results for the
average power consumption of BEBs show that traffic conditions and route congestion
increase the BEB energy consumption. Also, the BEB charging study shows that this
proposed transit charging strategy saves costs and could be a key for energy management
for the BEB transit network. Furthermore, this work estimates the possible GHG emission
saving figure based on the case study.
iv
ACKNOWLEDGEMENTS
Over the last three and a half years, developing this PhD thesis has been one of the most
demanding yet satisfying experiences. This research would not have been possible
without the assistance and encouragement of numerous individuals, to whom I wish to
express my heartfelt gratitude:
First, I would like to express my gratitude to Professor Stuart Galloway, my academic

supervisor, for his invaluable guidance, unwavering encouragement, and regular
objective meetings that help shaped this research work. I will be eternally grateful to him
for the opportunity to begin, grow, and complete this PhD under his supervision.
Additionally, I would like to express my profound gratitude to the Nigerian Petroleum

Technology Development Fund (PTDF) for sponsoring this research work.
Sincere appreciation is also extended to former and current colleagues in the Distributed
Energy Research (DER) and Aircraft Electrical Systems (AES) groups for making this
much more enjoyable than it would have been otherwise. I would like to express my
gratitude to Dr Rory Telford, Dr Bruce Stephen, Damien Frame, Kyle Smith, Calum
Edmunds, Madalisto Chikumbanje, Allison Strachan, and Lewis Hunter for their
assistance with this work at various times.
It is difficult to find the appropriate words to articulate my heartfelt appreciation for my

wife, children, and parents. None of my accomplishments would have been possible
without their love and kindness. Similarly, I am indebted to my uncle, Professor S S
Asaolu, and academic mentors, Dr I B Oluwafemi, for their unflinching support from
thousands of miles away that has sustained me throughout these years.
Finally, I would like to express my gratitude to all of the friends who helped make these
years in Glasgow so enjoyable and unforgettable and who assisted me in balancing the
sometimes-difficult academic life. They all contributed, perhaps unknowingly, to the
completion of this study in some way. I want to express my heartfelt appreciation to the
v
group of Nigerian students who attended Strathclyde University between 2017 and 2021.
Specifically, Jamiu Aladejana welcomed me upon my arrival in Glasgow and provided
me with my first accommodation. Additionally, I want to express my gratitude to Ajayi
Taiwo and Dr Fyail Jibji-Bukar from the Institute of Energy and Environment for their
friendship over the years. They were sharing the challenges of the PhD journey, and the
daily lunch made this experience extremely enjoyable.
vi
CONTENTS
1 INTRODUCTION ........................................................................................................ 1
1.1 THE NEED FOR THE ELECTRIC TRANSIT SYSTEM ........................................................ 4
1.2 PROBLEM DEFINITION ............................................................................................... 8
1.3 RESEARCH OBJECTIVES .......................................................................................... 11
1.4 SUMMARY OF CONTRIBUTIONS ............................................................................... 12
1.5 THESIS OUTLINE ..................................................................................................... 14
1.6 PUBLICATIONS ARISING FROM THE WORK OF THIS THESIS ....................................... 15
2 LITERATURE REVIEW.......................................................................................... 17
2.1 THE TRANSIT SYSTEM ............................................................................................ 18
2.1.1 The bus rapid transit system .......................................................................... 19
2.1.2 The sustainable BRT system ........................................................................... 23
2.1.3 Features of BRTs ............................................................................................ 25
2.2 ELECTRIC BUSES ..................................................................................................... 31
2.3 OPTIONS FOR BEBS CHARGING .............................................................................. 36
2.4 CHARGING STRATEGIES .......................................................................................... 38
2.4.1 Overnight Charging ....................................................................................... 38
2.4.2 Opportunity Charging .................................................................................... 38
2.4.3 In-Motion Charging ....................................................................................... 39
2.5 CHARGING TECHNOLOGY AND CHARGING LOCATION FOR TRANSIT ELECTRIC BUS 40
2.5.1 Charging locations ......................................................................................... 41
2.5.2 Charging technologies ................................................................................... 44
2.6 BATTERY TECHNOLOGY ......................................................................................... 50
2.7 BATTERY ELECTRIC BUS ENERGY CONSUMPTION MODELLING ................................ 51
2.8 STOCHASTIC EV CHARGING IMPACT AND INTEGRATION OF THE BEB CHARGING
DEMAND INTO THE ELECTRICAL DISTRIBUTION SYSTEM .............................................. 57
2.9 ELECTRIC VEHICLES CHARGING DEMONSTRATION WITH MICROGRIDS AND OFF-

GRID SYSTEMS ............................................................................................................. 60
2.10 CHAPTER SUMMARY ............................................................................................. 63
3 BEB TRANSIT FLEET DESIGN AND MATHEMATICAL MODELLING ..... 66
3.1 TRANSIT FLEET MATHEMATICAL MODEL ................................................................ 68
3.2 BEB ENERGY CONSUMPTION MODEL .................................................................... 71
3.2.1 The longitudinal dynamic model for the BEB energy consumption ............... 73
3.3 MODELS OF CHARGING INFRASTRUCTURES ............................................ 77
vii
3.3.1 NOVEL DESIGN OF CHARGING INFRASTRUCTURE FOR BUS RAPID
TRANSIT SYSTEM .................................................................................................. 77
3.3.2 Optimisation problem for the BEB transit model's priority-charging
infrastructure allocation. ........................................................................................ 83
3.3.3 Optimisation problem for the allocation of the priority-charging infrastructure
in a multi-terminal transit model ............................................................................ 89
3.3.4 Optimisation problem for the allocation of the priority integrated charging
model ....................................................................................................................... 90
3.4 DESIGN OF INDUCTIVE CHARGING SYSTEM FOR BUS RAPID TRANSIT NETWORK... 94
3.4.1 The BRT inductive charging system overview ............................................... 96
3.4.2 Wireless Charging System ............................................................................. 97
3.4.3 Mathematical optimisation model.................................................................. 98
3.5 CHAPTER SUMMARY ............................................................................................. 102
4 BEB TRANSIT FLEET ELECTRIFICATION NUMERICAL CASE STUDIES
104
4.1 INTRODUCTION ..................................................................................................... 104
4.2 OVERVIEW OF LAGOS BUS RAPID TRANSIT-CASE STUDY .................................... 106
4.3 BEB ENERGY CONSUMPTION SIMULATION .......................................................... 108
4.3.1 The Advanced Vehicle Simulator (ADVISOR) ............................................. 108
4.3.2 The Advanced Vehicle Simulator (ADVISOR) structure and capabilities ... 110
4.3.3 ADVISOR BEB energy consumption simulation.......................................... 113
4.3.4 Simulation results......................................................................................... 119
4.3.5 Discussion and Summary ............................................................................. 121
4.4 SIMULATION BEB TRANSIT CHARGING INFRASTRUCTURES ................................. 122
4.4.1 PSO Modeling .............................................................................................. 123
4.4.2 Transit fleet electrification simulation studies ............................................. 127
4.4.3 Charging infrastructure simulation studies ................................................. 132
4.4.4 Transit charging demand profile ................................................................. 145
4.4.5 Summary and discussion .............................................................................. 146
4.5 CHAPTER SUMMARY ............................................................................................ 147
5 BEB TRANSIT FLEET ELECTRIFICATION NUMERICAL CASE STUDIES
(2)-INDUCTIVE CHARGING AND CARBON EMISSIONS SAVINGS ............ 148
5.1 INTRODUCTION ..................................................................................................... 148
5.2 CASE STUDY OF INDUCTIVE CHARGING SIMULATION............................................ 149
5.2.1 Introduction.................................................................................................. 149
viii
5.2.2 Data and input parameters .......................................................................... 151
5.2.3 Inductive charging simulation studies ......................................................... 152
5.3 A CASE STUDY OF CARBON EMISSIONS SAVINGS OF ELECTRIFIED BRT SYSTEM ... 159
5.4 CHAPTER SUMMARY ............................................................................................ 164
6 CONCLUSIONS AND FUTURE WORK ............................................................. 165
6.1 THESIS CONCLUSIONS .......................................................................................... 165
6.2 FUTURE WORK ..................................................................................................... 173
7 REFERENCES ......................................................................................................... 177
8 APPENDICES .......................................................................................................... 213
8.1 APPENDIX A: PSO ALGORITHM ........................................................................... 213
8.1.1 PSO Basic Modelling Concepts ................................................................... 215
8.2 APPENDIX B: SUPPLEMENTARY DATA FOR CASE STUDY ..................................... 220
8.2.1 Average Lagos BRT driving (minutes) ......................................................... 220
8.2.2 Lagos BRT weekdays operational Schedule ................................................ 222
8.2.3 Lagos BRT Saturday operational Schedule ................................................. 222
8.2.4 Lagos BRT Sunday operational Schedule .................................................... 223
8.2.5 Lagos BRT Corridor Lanes .......................................................................... 224
ix
LIST OF TABLES
TABLE 2-1 KEY BENEFITS OF THE BUS RAPID TRANSIT SYSTEM (ADAPTED FROM [96]) .... 20
TABLE 2-2 CHARACTERISTICS OF ELECTRIC TRANSIT BUSES .......................................... 33
TABLE 2-3 COMPARATIVE EVALUATION OF HEB, FCEB AND DB ( [12]) ........................ 34
TABLE 2-4 SUMMARY OF BEB CHARGING POSSIBILITIES (ADOPTED FROM [184]) ........... 37
TABLE 2-5 BATTERY ELECTRIC BUS CHARGING STRATEGIES (ADOPTED FROM [12]) ....... 39
TABLE 2-6 CHARACTERISTICS OF LEVEL 1, LEVEL 2 AND LEVEL 3 CHARGERS [219] [220],
[221] ........................................................................................................................ 47
TABLE 4-1 BEB PARAMETER (SOURCE: BYD ADL ENVIRO200EV) ............................ 114
TABLE 4-2 AVERAGE BEBS ENERGY CONSUMPTION BASED ON TRAFFIC FLOW

CLASSIFICATION ..................................................................................................... 121
TABLE 4-3 ROUTE DISTANCE AND RUNTIME (SOURCE: LAMATA) ............................... 129
TABLE 4-4 MODELLING AND SIMULATION PARAMETER (ADAPTED FROM [245], [340],
[100])..................................................................................................................... 131
TABLE 4-5 SIMULATION RESULT FOR THE SCENARIO I ................................................... 133
TABLE 4-6 SIMULATION RESULT FOR PRIORITY CHARGING (SCENARIO II) ..................... 136
TABLE 4-7 SIMULATION RESULT FOR THE INTEGRATED MODEL (SCENARIO III) ............. 141
TABLE 4-8 SIMULATION RESULT FOR THE INDUCTIVE CHARGING MODEL (SCENARIO IV)
............................................................................................................................... 143
TABLE 5-1 BEB PARAMETER (ADOPTED FROM [101]) ................................................... 151
TABLE 5-2 RESULT OF ALLOCATION OF INDUCTIVE SIMULATION ................................... 152
TABLE 5-3 SENSITIVITY ANALYSIS OF THE MULTI-TERMINAL CONFIGURATION OF
INDUCTIVE CHARGING TRANSIT DESIGN ................................................................. 157
TABLE 5-4 COMPARISON OF DIESEL BUS AND BEB GHG EMISSION (GCO2/KM) ........... 160
TABLE 5-5 BEBS GHG EMISSION SAVING FIGURES AT DIFFERENT TRAFFIC CONDITIONS
............................................................................................................................... 161
x
LIST OF FIGURES
FIGURE 1-1 GHG EMISSIONS FROM DIFFERENT MODES OF TRANSPORT (ADOPTED FROM
[49])........................................................................................................................... 6
FIGURE 2-1 STRATEGIES FOR SUSTAINABLE BRTS [137] ................................................. 24
FIGURE 2-2 EL METROPOLITANO'S TRUNK CORRIDOR [148] ........................................... 26
FIGURE 2-3 BUS TERMINAL IN OSHODI, LAGOS [151] ...................................................... 27
FIGURE 2-4 LAGOS BRT BUS STOP [154] ......................................................................... 28
FIGURE 2-5 LAGOS BRT BUS DEPOT [156] ...................................................................... 28
FIGURE 2-6 A TYPICAL BOARDING PLATFORM FOR EASY ENTRY/EXIT [157]..................... 29
FIGURE 2-7 TRANSMILENIO PREPAID BOARDING AND PASSENGER INFORMATION IN BOGOTA

(ADOPTED FROM [158]) ........................................................................................... 30
FIGURE 2-8 THE DEFINING COMPONENTS OF DIFFERENT ELECTRIC BUSES (SOURCE: [12]).
................................................................................................................................. 32
FIGURE 2-9 BATTERY ELECTRIC BUS CONFIGURATION (ADOPTED FROM [12])............... 33
FIGURE 2-10 BEB CHARGING TECHNOLOGIES ................................................................. 44
FIGURE 2-11 TRANSIT INDUCTIVE SYSTEM (ADOPTED FROM [101])................................. 46
FIGURE 2-12 PANTOGRAPH SYSTEM (SOURCE: [227]) ..................................................... 48
FIGURE 3-1: THE STEPWISE MODEL OF THE BRT ELECTRIFICATION PROCESS ................... 67
FIGURE 3-2 TRANSIT SCHEDULE FOR ASSIGNMENTS D ...................................................... 70
FIGURE 3-3 BEB POWERTRAIN SYSTEM CONFIGURATION (ADAPTED FROM [289]) ......... 72
FIGURE 3-4 FREE BODY DIAGRAM OF A BEB IN MOTION (ADAPTED FROM [291]) ............ 74
FIGURE 3-5 PROPOSED ALLOCATION OF CHARGING INFRASTRUCTURE FOR THE BUS TRANSIT
SYSTEM .................................................................................................................... 79
FIGURE 3-6 FLOW CHART FOR THE TERMINAL-BASED CHARGING MODEL ......................... 82
FIGURE 3-7 MULTI-TERMINAL ROUTE BASED CHARGING MODEL ..................................... 87
FIGURE 3-8 MODEL OF TWO-TRANSIT TERMINAL ............................................................. 92
FIGURE 3-9 ALLOCATION OF INDUCTIVE CABLE LENGTH ................................................. 94
FIGURE 3-10 BRT ROUTE WITH BUSES ............................................................................. 96
xi
FIGURE 3-11 DESIGN OF THE BEB INDUCTIVE TRANSIT SYSTEM ...................................... 97
FIGURE 3-12 PROPOSED BRT INDUCTIVE CHARGING SYSTEM .......................................... 98
FIGURE 3-13 MATHEMATICAL MODEL TRANSIT INDUCTIVE SYSTEM ................................ 99
FIGURE 3-14 ALLOCATION OF INDUCTIVE CABLE ........................................................... 101
FIGURE 3-15 SUMMARY OF BEB TRANSIT FLEET DESIGN AND MATHEMATICAL MODELLING
............................................................................................................................... 103
FIGURE 4-1 MODELLING ELEMENTS AND SIMULATION OVERVIEW ................................. 105
FIGURE 4-2 LAGOS BRT ROUTE WITH THE LOCATION OF BUS STOPS (SOURCE: PRIMERO)
............................................................................................................................... 107
FIGURE 4-3 ADVISOR VEHICLE INPUT WINDOW ........................................................... 112
FIGURE 4-4 ADVISOR SIMULATION SETUP WINDOW .................................................... 112
FIGURE 4-5 ADVISOR RESULTS WINDOW ..................................................................... 113
FIGURE 4-6: FREE TRAFFIC (AVERAGE SPEED: 33KM/H) ................................................. 116
FIGURE 4-7 MOVE TRAFFIC (AVERAGE SPEED: 21KM/H) ................................................ 117
FIGURE 4-8: APPROPRIATE TRAFFIC (AVERAGE SPEED: 18KM/H).................................... 117
FIGURE 4-9: INAPPROPRIATE TRAFFIC (AVERAGE SPEED: 12KM/H)................................. 118
FIGURE 4-10 CRITICAL TRAFFIC (AVERAGE SPEED: 7KM/H) ........................................... 118
FIGURE 4-11 THE ROUTE ELEVATION PROFILE [332] ...................................................... 119
FIGURE 4-12 RESULT (FREE TRAFFIC) ............................................................................ 120
FIGURE 4-13 PSEUDOCODE OF THE PSO ALGORITHM ..................................................... 124
FIGURE 4-14 GRAPHICAL REPRESENTATION EXPLAINS THE CONSTRAINT-HANDLING
MECHANISM INCORPORATED INTO OUR PSO ALGORITHM (ADAPTED FROM [339]).126
FIGURE 4-15 LAGOS TRANSIT NETWORK SHOWING BUS TERMINALS (SOURCE: LAMATA)
............................................................................................................................... 128
FIGURE 4-16 SCENARIO I, TRADE-OFF ANALYSIS ........................................................... 134
FIGURE 4-17 CHARGING DURATION (MIN) AND SOC (%) OF BEBS PRIORITY AND NON-
PRIORITY (50KW) SCENARIOS ................................................................................ 137
xii
PRIORITY (200KW) SCENARIOS .............................................................................. 139
PRIORITY (400KW) SCENARIOS .............................................................................. 140
FIGURE 4-20 OPTIMAL DESIGN LAGOS BRT NETWORK ELECTRIFICATION ..................... 142
FIGURE 4-21 TRADE-OFF ANALYSIS OF SCENARIO I, II, III & IV .................................... 144
FIGURE 4-22 BEB CHARGING DEMAND PROFILE ............................................................ 145
FIGURE 5-1 REDESIGNED TRANSIT SYSTEM (SCENARIO II) ............................................. 150
FIGURE 5-2 REDESIGNED TRANSIT SYSTEM (SCENARIO III) ........................................... 151
FIGURE 5-3 SCENARIO I- STATE OF CHARGE ANALYSIS .................................................. 154
FIGURE 5-4 SCENARIO II- STATE OF CHARGE ANALYSIS ................................................. 154
FIGURE 5-5 SCENARIO III- STATE OF CHARGE ANALYSIS ............................................... 155
FIGURE 5-6 SENSITIVITY ANALYSIS AT DIFFERENT NUMBER OF TERMINALS ................... 156
FIGURE 5-7 ALTERNATIVE CONFIGURATIONS OF INDUCTIVE CHARGERS FOR LAGOS BRT

NETWORK ............................................................................................................... 157
FIGURE 5-8 COMPARISON OF DIESEL BUS AND BEBS GHG EMISSION AT DIFFERENT
TRAFFIC CONDITION .............................................................................................. 161
FIGURE 5-9 COMPARISON OF (100 TO 1000 BEBS) DIESEL BUS AND BEBS GHG EMISSION
(TCO2 PER YEAR) AT DIFFERENT TRAFFIC CONDITIONS.......................................... 162
FIGURE 6-1 PROPOSED TERMINAL BASED TRANSIT ENERGY SYSTEM ........................... 174
FIGURE 6-2 PROPOSED MULTI-TERMINAL BASED TRANSIT ENERGY SYSTEM ................ 175
FIGURE 8-1 IN PSO, EACH PARTICLE CONSIDERS ITS PRIOR VELOCITY, PERSONAL BEST, AND
GLOBAL BEST TO DEFINE ITS CURRENT VELOCITY AND POSITION (ADOPTED FROM
[386])..................................................................................................................... 216
FIGURE 8-2 THE INERTIAL WEIGHT'S (W) EFFECT ON THE VELOCITY VECTOR IN PSO
(ADOPTED FROM [386]). ........................................................................................ 218
FIGURE 8-3 WHEN W=0, C1 =1 AND C2 =1, EXPLOITATION AND LOCAL SEARCH
PERFORMANCE IN PSO (ADOPTED FROM [386]). .................................................... 218
xiii
FIGURE 8-4 EXPLORATION AND GLOBAL SEARCH RISE PROPORTIONATELY TO W, C1 AND C2
VALUES (ADOPTED FROM [386]). ........................................................................... 219
xiv
LIST OF ABBREVIATIONS AND ACRONYMS
A. Abbreviations
ADVISOR Advanced Vehicle Simulator
BEB Battery Electric Bus
DER Distributed Energy Resources
DSM Demand Side Management
ESS Energy storage systems
EB Electric Bus
EV Electric Vehicles
GHG Greenhouse Gases
HVAC Heat, Ventilation, and Air Conditioning
ICE Internal Combustion Engine
LoS Level of Service
PBT Public Bus Transit
SoC State of Charge
PSO Particle swarm optimisation
BRT Bus Rapid Transit
VKT Vehicle-kilometres Travelled
LAMATA Lagos Metropolitan Area Transport Authority
HEB Hybrid Electric Bus
EM Electric Motor
FCEB Fuel Cell Electric Bus
BSS Battery Swapping Station
MG Micro-grid
xv
B. PARAMETERS
𝐴 BEB cross section area (𝑚2 )
𝑎𝑡 BEB acceleration at time t (m/𝑠 2 )
𝑎𝑡,𝑗,𝑑 BEB acceleration at time t, trip j, and distribution d (m/𝑠 2 )
𝐶𝑡𝐷 , 𝐶𝑟 Drag, rolling resistance coefficient
𝐸𝑏𝑐𝑎𝑝 BEB battery energy capacity

𝑐𝑜𝑛𝑠
𝐸𝑡,𝑑 BEB energy consumption rate
𝐸𝑡 BEB energy status at any instance

𝑡𝑟𝑖𝑝
𝐸𝑙𝑟 Route Energy demand in kWh/km
𝑔
𝐹𝑡𝑎 , 𝐹𝑡 , 𝐹𝑡𝑟 , 𝐹𝑡𝑝 Aerodynamic, grade, rolling resistance and traction forces
𝑝
𝐹𝑡,𝑗,𝑑 BEB traction force at time t, trip j, and distribution d (N)
𝑏
𝐹𝑡,𝑗,𝑑 BEB brake force at time t, trip j, and distribution d (N)
g Gravitational force (m/𝑠 2 )
𝑙𝑟 Length of the route
𝑀𝑡 , 𝑀𝑡,𝑗,𝑑 Mass of BEB at any instance
𝑃𝑎𝑢𝑥 BEB auxiliary power demand

𝑐𝑜𝑛𝑠
𝑃𝑡,𝑑 Power consumption of the BEB
𝑅𝑤ℎ𝑒𝑒𝑙 Wheel radius
𝑆 Split ratio
𝑆𝑂𝐶(𝑡) State of charge at the time (t)
𝑆𝑂𝐶𝑏𝑚𝑎𝑥 , 𝑆𝑂𝐶𝑏𝑚𝑖𝑛 Max, min state of charge of the BEB
𝑇𝑐𝑎𝑝 Transmitter capacity

𝑚 𝑤
𝑇𝑡,𝑗,𝑑 , 𝑇𝑡,𝑗,𝑑 Motor and wheel torque
𝑡𝑑𝑒𝑛𝑑 End time for distribution d (mins).
xvi
𝑡𝑑𝑠 Start time for distribution d (mins).
𝑡𝑑𝑤𝑎𝑖𝑡 Waiting time after each trip within distribution d (mins).

𝑐𝑦𝑐𝑙𝑒
𝑡𝑑 Cycle time for distribution d (mins).
𝑡𝑑𝑡𝑟𝑖𝑝 Trip time for distribution d (mins).
𝑣𝑡,𝑗,𝑑 BEB speed at time t, trip j, and distribution d (m/s).

𝑚 𝑤
𝑤𝑡,𝑗,𝑑 , 𝑤𝑡,𝑗,𝑑 Motor and wheel rotational speed
𝜌 Air density (kg/𝑚3 )
∅𝑡,𝑗,𝑑 Road slip angle at time t, trip j, and distribution d (deg.)
𝜂𝑐𝑜𝑛 , 𝜂𝑚 , 𝜂𝑇 Conversion, motor and transmission efficiency
𝐸𝑏𝑐𝑎𝑝 BEB 𝑏 battery energy capacity (kWh).
𝐸𝑏𝑚𝑎𝑥 Maximum energy threshold in BEB 𝑏 (kWh).
𝐸𝑏𝑚𝑖𝑛 Minimum energy threshold in BEB 𝑏 (kWh).

𝑎𝑟𝑟
𝐸𝑏,𝑗,𝑑 Energy capacity of the BEB 𝑏 after trip 𝑗 in assignment 𝑑 (kWh).
𝑑𝑒𝑝
𝐸𝑏,𝑗,𝑑 Energy capacity of the BEB 𝑏 at the start of trip 𝑗 in assignment 𝑑 (kWh).
𝑐ℎ,𝑙
𝐸𝑏,𝑗,𝑑 Energy gained by BEB 𝑏 via charging with low-priority charger during trip 𝑗 in
assignment 𝑑 (kWh).
𝑐ℎ,𝑚
𝐸𝑏,𝑗,𝑑 Energy gained by BEB 𝑏 via charging with medium-priority charger during trip
𝑗 in assignment 𝑑 (kWh).
𝑐ℎ,ℎ
𝐸𝑏,𝑗,𝑑 Energy gained by BEB 𝑏 via charging with high-priority charger during trip 𝑗 in
assignment 𝑑 (kWh).
𝐸𝑏𝑖,𝑖+𝑛 Energy required by BEB 𝑏 to travel from terminal i to i+n (kWh).
𝐶𝑏𝑏𝑎𝑡 Cost of BEB 𝑏 on-board battery (£/kWh).

𝑙
𝐶𝑐ℎ Cost of charging with low-priority charger (£).
𝑚
𝐶𝑐ℎ Cost of charging with medium-priority charger (£).
ℎ
𝐶𝑐ℎ Cost of charging with high-priority charger (£).
𝐶𝑓𝑙 Fixed cost charge for using low-priority charger (£).
xvii
𝐶𝑓𝑚 Fixed cost charge for using medium-priority charger (£)
𝐶𝑓ℎ Fixed cost charge for using high-priority charger (£).
𝐶 Cost of energy per kWh (£).

𝑙
𝐶𝑖𝑛𝑠𝑡 Low-priority charger installation cost (£).
𝑚
𝐶𝑖𝑛𝑠𝑡 Medium-priority charger installation cost (£).
ℎ
𝐶𝑖𝑛𝑠𝑡 High-priority charger installation cost (£).
𝑖𝑛𝑑
𝐶𝑖𝑛𝑠𝑡 Inductive charger installation cost (£).
𝐶𝑊𝑃𝑇 Wireless power transmitter expenditure cost (£).
𝐶𝑐𝑎𝑏𝑙𝑒 Inductive cable expenditure cost (£).
𝐶ℎ 𝑆𝑒𝑙𝑒𝑐𝑡 Charging priority selector.

𝑖𝑛𝑑
𝑃𝑐ℎ Power capacity of the inductive charger (kW).
𝑖𝑛𝑑
𝑃𝑢𝑠𝑒(𝑏,𝑗,𝑑) Power used by BEB 𝑏 when moving over inductive cable in trip 𝑗 within
assignment 𝑑 (kW).
𝑎𝑟𝑟,𝑖
𝑡𝑏,𝑗,𝑑 Arrival time of the BEB 𝑏 at the bus terminal 𝑖 after trip 𝑗 in assignment 𝑑
(minutes).
𝑐ℎ,𝑖
𝑡𝑏,𝑗,𝑑 BEB 𝑏 conductive charging duration at the bus terminal 𝑖 after trip 𝑗
inassignment 𝑑 (minutes).
𝑐ℎ,𝑙
𝑡𝑏,𝑗,𝑑 BEB 𝑏 low-priority charging time for trip 𝑗 in assignment 𝑑 (minutes).
𝑐ℎ,𝑚
𝑡𝑏,𝑗,𝑑 BEB 𝑏 medium-priority charging time for trip 𝑗 in assignment 𝑑 (minutes).
𝑐ℎ,ℎ
𝑡𝑏,𝑗,𝑑 BEB 𝑏 high-priority charging time for trip 𝑗 in assignment 𝑑 (minutes).
𝑐ℎ
𝑡𝑖𝑛𝑑(𝑏,𝑗,𝑑) BEB 𝑏 induction charging in trip 𝑗 within assignment 𝑑 (minutes).
𝑠,𝑖
𝑡𝑏,𝑗,𝑑 BEB 𝑏 scheduled departure time conductive at the terminal 𝑖 for trip 𝑗 in
assignment 𝑑 (minutes).
𝑡𝑟𝑎𝑣
𝑡𝑖𝑛𝑑(𝑏,𝑗,𝑑) BEB 𝑏 travelling time over induction cable in trip 𝑗 within assignment 𝑑
(minutes).
𝑇𝑐𝑎𝑝 Transmitter capacity (kWh).
xviii
𝜂𝑐ℎ Charger efficiency (%).
𝜄 𝑚 ℎ
𝑃(𝑚𝑎𝑥) , 𝑃(𝑚𝑎𝑥) , 𝑃(𝑚𝑎𝑥) Maximum power capacity low, medium and high priority
conductive charger respectively (kW).
𝑑𝑖𝑆 , 𝑑𝑖𝑒 Start and end of inductive cable (m).

𝑡𝑟𝑎𝑣
𝑑𝑖𝑛𝑑(𝑏,𝑗,𝑑) BEB 𝑏 travelling distance over induction cable in trip 𝑗 within assignment 𝑑
(minutes).
𝑁𝑏 Number of BEB 𝑏.
𝑖𝑛𝑑
𝑁𝑐ℎ Number of inductive chargers.
𝚤 𝑚 ℎ
𝑁𝑐ℎ , 𝑁𝑐ℎ , 𝑁𝑐ℎ Number of low, medium and high priority.
𝑁𝑊𝑃𝑇 Number of wireless power transmitters.
𝑛(𝑇 𝑙 ) Number of routes distance that requires a low-priority chargers.
𝑛(𝑇 𝑚 ) Number of routes distance that requires medium-priority chargers.
𝑛(𝑇 ℎ ) Number of routes distance that requires a high-priority chargers.
𝑛(𝑀) Total of routes combination in the transit network.
𝑛(𝐷) Number of diagonal elements in the matrix .
𝑇 Matrix representation of the routes combination in the transit system.

𝑐𝑜𝑛𝑠
𝑃𝑡,𝑗,𝑑 Power consumption of the BEB 𝑏 in time 𝑡 used for trip 𝑗 in assignment 𝑑 (kW).
𝜄 𝑚 ℎ
𝑃(𝑡) , 𝑃(𝑡) , 𝑃(𝑡) Power used by low, medium and high-priority conductive charger over time
interval 𝑡 (kW).
𝑡𝑜𝑡𝑎𝑙
𝑃(𝑡) Power used by all the chargers in the transit network over a time interval (t) (kW).
∝𝑙 , ∝𝑚 , ∝ℎ Probability for low, medium and high priority charged BEBs, respectively.
𝛽 Binary variable for inductive charging
𝐵𝑐𝑎𝑝 BEB battery capacity (kWh)
𝐶𝐵𝐸𝐵 , 𝐶𝑊𝑃𝑇 𝐶𝑊𝑃𝑅 , 𝐶𝑏𝑎𝑡 , 𝐶𝑐𝑎𝑏𝑙𝑒 Costs of BEB, WPT, WPR, battery and cable (£)
𝑑𝑖𝑒 , 𝑑𝑖𝑒 Start and endpoint of inductive cable (m)
𝐸𝑡 BEB energy status at any instance (kWh)

𝑁𝐵𝐸𝐵 , 𝑁𝑊𝑃𝑇 Number of BEB, WPT
xix
𝑃𝑔𝑎𝑖𝑛(𝑡) , 𝑃𝑢𝑠𝑒(𝑡) Power gain, use by the BEB (kWh)
𝑇𝑐𝑎𝑝 Transmitter capacity (kWh)
𝑐1, 𝑐2 Acceleration coefficients
𝑛𝑉𝑎𝑟 Number of variables
𝐺 Best previous positions of all particles
𝑡 Current iteration of the algorithm.
𝑡𝑚𝑎𝑥 Maximum number of iterations
𝑟1, 𝑟2 Random variables
𝑢𝑏, 𝑙𝑏 Upper and lower bound
𝑋𝑖 , 𝑉𝑖 Position and velocity of the particle
𝑤, 𝑤𝑚𝑖𝑛 , 𝑤𝑚𝑎𝑥 Inertia weight, min. and max. weight
𝑃𝑖 Best previous positions of i-th particles

𝐺𝐵𝐸𝑆𝑇, 𝑃𝐵𝐸𝑆𝑇 Global and personal best particle position
C. INDICES
𝑏 Indices for transit network battery electric buses (BEBs)
𝑟 Indices for transit network routes
𝑖 Indices for the terminal in the transit networks.
𝑗 Indices for BEB's trips.
𝑡 Indices for time steps.
𝑏 Indices for transit network BEBs.
𝑑 Indices for BEB's assignments.
D. SETS
ℬ Set of transit buses (BEBs)
ℛ Set of transit network routes
𝑆𝑑 Set of transit schedule trips with distributions d
𝐷 Set of transit network assignments.
xx
𝐽𝑑 Set of trips within the assignment 𝑑.
𝜏 Set of optimisation time steps.
xxi
Chapter 1: Introduction
1 INTRODUCTION
Air is an integral part of numerous indispensable cycles on Earth; the air is needed to live
by most species, including human beings. The component of air in Earth's atmosphere is
about 78% nitrogen and 21% oxygen. Air also has insignificant amounts of other gases,
such as neon, hydrogen, and carbon dioxide. Air also contains lots of microscopic
particles, and these tiny particles in the air are called aerosols. Aerosol, like dust, is picked
up naturally when the wind blows. Air can also carry soot, smoke, and other particles
from vehicle emissions, power plants, and industrial production, which are the main
contributors to air pollution [1]. Air becomes polluted when it contains harmful
concentrations of mixed particles and gases such as soot, smoke, methane, and carbon
dioxide. Famous among the air pollutants are particulate matter, black carbon, ground-
level ozone (O3), nitrogen dioxide (NO2), nitrous oxide (N2O), sulphur dioxide (SO2) and
carbon monoxide (CO).
The world health organisation indicates that nine out of ten people have to breathe in
air with a high level of pollutants [2]. Air pollutants' health consequences include
respiratory infections, asthma, chronic obstructive pulmonary disease, and lung cancer,
even in combination with stroke and heart diseases [3]. Air pollution is considered to
account for the death of about 7 million people annually [4]. In 2017, air pollution
contributed to about 9% of global deaths [4]. Air pollution has been considered one of the
world's leading risk factors for mortality [5]. The cities have been considered to account
Adedayo Asaolu - August 2022 1

for 64% of global energy use and 70% of carbon emissions [6] [7]. The increase in the
population living in the cities is likely to increase in these proportions unless the cities
lead towards sustainable energy transition and decarbonisation.
Developed countries are making an effort to improve the air quality in their cities, with
about 51% reported in 2018 to meet the WHO air quality guideline [8][9]. However, less
than 3% of the cities in developing countries could meet the WHO air quality standard
(WHO 2018C) [8]. In 2019, the transport sector emissions accounted for 24% of global
emissions and road transport emissions are responsible for 75% of the total emissions in
the transport sector (IEA, 2020) cited in [10]. In addition, CO2 emissions from the global
transportation sector fell by more than 10% in 2020, owing to global lockdown measures
imposed by the Covid-19 [10]. The 2019 road transport emission figures compared to
1990 have seen road transport emissions have increased by 68% [11]. Decarbonising road
transport has urged the public to seek a sustainable means of transportation. The
international energy agency (IEA) has determined that walking, cycling, and public
transport as the city modes of transport could reduce significant emissions and save about
one trillion dollars by 2050 [7].
With cities' transport decarbonisation in mind, zero-emission Battery-Electric Buses

(BEBs) are recently receiving increasing attention globally. Their adoption can be
generally attributed to emissions reduction (UITP, 2017) [12]. Also, the battery-electric
bus can reduce emissions and costs and transport more people simultaneously than an
individual passage vehicle. In [13], the authors estimate that it takes about 100 electric
cars to accomplish the same environmental relief as can be gained from a single 18 m
battery-electric bus. In addition to the potential emissions reduction when electricity
originates from renewable sources, electric buses offer significant additional benefits
such as zero-emission, quiet operation, high efficiency, reliability, and better acceleration
than traditional buses [14].
Compared to trolley bus technology, they eliminate the need for constant grid
connection, and the transit route can be modified without infrastructure change (e.g.
pantographs). The cost of ownership is lower compared to diesel buses [14]. In [15], the
comparative on-road evaluation of diesel buses with battery-electric buses BEBs is
examined. In this work, the BEBs cuts petroleum use by 85–87% compared to a diesel

bus and attain a 32–46% decrease in fossil fuel use and 19–35% in CO2 emissions while
evaluating from a life-cycle perspective. However, other types of fuels can cut transport
emissions (e.g. hydrogen); none of them includes such an attractive set of benefits for
addressing the challenges of modern urbanisation as electricity does (the overall grid to
wheels efficiency of BEVs is about 50% and 60% for fast and slow charging respectively,
while hydrogen has an overall grid to wheels efficiency is about 30% [16],[17])).
Generally, developed countries are planning and implementing schemes for reducing
greenhouse gas emissions, and the critical focus is the electrification of the transport
system. Examples include the European roadmap to road transport's electrification [18]
and the UK transport energy infrastructure roadmap to 2050 [19]. These documents (ref.
[18] and [19]) detail the plan on how the UK and EU are planning to electrify the vehicular
transport system by the year 2050 [19]. However, the electrification of public transport
and transit networks is likely to evolve ahead of the individual owned vehicle because it
can be controlled by government policy and deliberate investment in public transport
infrastructures [20]. For example, most public transit buses in Shenzhen, China, are
electrified using an overnight charging strategy [21]. Notably, the power infrastructure
(i.e. energy need and the energy capacity of charging infrastructure) for public transport
fleets is different from that for individual passenger vehicles. The electric public transport
system's charging infrastructures are more of a hub base (i.e., terminals, depots, and bus
stops) that can be powered easily with renewable energy mini-grids [20]. A similar
example is Daimler central charging station for electric buses in Mannheim, Germany.
The charging station uses both plugin and pantograph chargers [22]. Another example is
a high-power electric bus charging station introduced in America by Proterra. This
Proterra charging station uses energy from RES (solar) and the grid. The electric bus
charging station can simultaneously charge up to 100 buses with the 125 kW capacity
chargers. This charging station also has the capacity for vehicle-to-grid (V2G) [23].
In developing countries (most especially in sub-Saharan Africa), investments in urban

transport infrastructure have been considered a tool to engage in reducing congestion,
time, energy consumption, safety, improvement of environmental outcomes, and
enhancing the economic development of the contemporary cities [24], [25], [26], [27].
Studies have shown that changes in the economic fortunes of developing countries' city
inhabitants significantly affect transportation systems, particularly automobile

dependency. This expulsion of vehicles on urban routes has correspondingly increased

road traffic congestion and GHG emission provision [28]. In many developing cities, the
public transportation system is characterised by disorganised and largely unregulated
demand that outstrips supply. Consequently, many citizens are often left out of productive
employment and other urban life benefits [29]. Enhancing these cities' functionality
depends on improvements to transport networks and providing reliable, affordable, and
safe access to urban opportunities. Firms and workers can collectively generate scale and
specialisation with better urban mobility and enhance productivity growth.
A typical example is Lagos (Nigeria), which has a population of about a 21million. The
characteristics of the Lagos transport system include the unavoidable daily experience of
the worst traffic situation due to road congestion and GHG emissions predominantly from
automobiles [30]. Lagos has a diesel-based bus rapid transit (BRT) system that connects
people from the suburbs to the central business district [31]. The greening of this public
asset, in combination with the favourable greenhouse gas (GHG) solutions, is an attractive
option for policymakers to eliminate 'drive-way congestion and support the adoption of
EVs [32] [33].
1.1 The need for the electric transit system

Private car ownership has soared significantly over the last few decades globally, and this
trend will continue. For example, between 2008 and 2020, the number of licensed drivers
in the USA increased by about a 19million [34]. In Great Britain, this figure increased by
4million [35]. In Lagos, Nigeria, it is estimated that more than 5 million cars and 200
thousand commercial vehicles are on the road, with a daily average of 227 vehicles per
road kilometre [36]. However, there will always be many people who will not have access
to a car and therefore be reliant on some form of public transport if their mobility demands
are to be met (in the UK, about 31% of people are without cars) [37]. Also, it is generally
accepted that lifestyle changes have increased dependence on mobility to access
opportunities for jobs, shopping, education, health, leisure and recreation. It implies that
there is an ever-growing demand for transportation. Generally, mobility is a means to an
end, not an end in itself. People move about to carry out planned activities at various
locations where those activities can take place [38] [39].

Walking, cycling, motorcycle, private cars, and public transport are the conventional
passenger transport modes in urban areas. Walking and cycling are for short trips, mainly
the former [38] [32]. For example, In Great Britain, the average distance a person walked
in 2016 was 198 miles, and the average distance people cycled was 53 miles [40]. Suppose
motorcycle activity is considered as providing for a comparatively small fraction of trips.
In that case, the significant competition between modes in urban areas lies between
private cars and public transport. In the idealistic scenario, it is expected that car owners
would be prepared to avoid using their car and shift to public transport. Nonetheless, since
most people who have a car will use it, transfer to public transport will only happen if
there are sensed advantages over individual vehicles. This shift will probably occur if
travel by car has become difficult, costly and unreliable [41]. Therefore, it is vital to
examine the existing urban travel situation, particularly during peak travel periods.
As car ownership increases, more journeying is made by car and fewer by public
transport. Limited travel by public transportation means inadequate revenue or limited-
service frequency, or fare increase by the transport operators [37] [38]. Fare and limited
service further discouraged the use of public transport and increased individual passenger
vehicle use. Increased private vehicle travel worsens road congestion, making travel by
bus also delayed and unreliable, thus encouraging personal car travel [42]. Therefore, the
downward spiral of bus public transport usage seems inevitable. However, increasing car
travel has created many severe detrimental consequences, of which the three most
important that are enumerated in [38] are:
1. Traffic congestion
Traffic congestion is an abnormality common in most cities and is increasingly happening

outside the regular commuter peak periods of travel. The effects of traffic congestion
include visual pollution, increased noise and lost economic output. Traffic congestion
also causes traffic diversion onto unsuitable roads as drivers attempt to find alternative
routes to avoid congestion, increasing journey times, making the journey unreliable, and
unacceptable driver stress and frustration. Traffic congestion impinges adversely on the
level of service that can be provided by public bus transport (but paradoxically can make
bus transit (with busway)-based public transport more attractive) [43] [44] [45].
2. Energy consumption and air pollution

Although manufacturers are continually developing more efficient engines, cars use
finite petroleum products. Increasing public transport use relative to the car will conserve
this essential resource [46]. Increasing car travel increases air pollution [47] [48]. On a
global, 88% of CO2 emissions, 48% of NOx and 37% of volatile organic compounds
originate from road vehicles [38]. Word-wide transport emission control is considered a
critical issue requiring urgent attention to salvage human health and the planet. Figure 1-1
shows the GHG emissions from different modes of transport. Apart from walking and
cycling, the GHG favourable means of transportation is the use of public transport.
Figure 1-1 GHG emissions from different modes of transport (Adopted from [49])
3. Safety
Vulnerable road users such as motorcyclists, cyclists and pedestrians are prone to a higher
risk of accidents with increased travelling vehicles [50], [51]. Hence, the improved use
of public transport would have significant safety advantages.
Nonetheless, individual vehicles have many benefits to city residents. However, it has
gotten to a level that using personal vehicles results in traffic congestion that extends
beyond the regular morning and evening peak periods and results in ever-increasing and
unreliable travel times [38],[52]. Therefore, it is likely that if a satisfactory and reliable

public transport service were to be introduced for many journeys, a shift from car to public
transport could be envisioned [53], [54].
The regular bus services use the same road as other vehicles and consequently are
subjected to traffic congestion. In order to mitigate traffic congestion, bus lanes with flow
or contra-flow can be introduced to enable buses to avoid traffic [29]. Also, roads may be
constructed for the exclusive use of dedicated buses (busways). For example, buses are
usually subjected to road traffic except for the BRTs with a dedicated busway [29]. The
former is cheap, but it creates a barrier to other movements (including pedestrians) that
need to cross because buses are given signalled priority over other traffic junctions,
increasing a traffic delay [38]. Bus services on busways avoid delays to services operating
on the ordinary road network [55], [56]. They can be readily linked with residential,
industrial, and shopping areas with regular bus operations.
In summary, increased public transportation usage is necessary, especially in urban

areas, which significantly benefits environmental protection and transportation mobility.
Adopting transit systems (rapid bus transit with busways, light rail, and subways) is
considered a possible form of public transport capable of solving the urban transport
problem. It is most likely that bus transit will move ahead of other transit means in
developing countries. Because it is affordable, safe, reduces congestion, minimises
transport energy consumption, and reduces GHG emissions [57] (especially with
electrified bus transits systems [58]).
Apart from an electrified transit system's emission benefits, the electric powertrain's
added advantage is its electric drive motors' greater energy conversion efficiencies than
the Internal Combustion Engine (ICE). The energy conversion efficiency for the electric
drive motors is about 76% (i.e., from the on-board battery to the wheels), which is about
five times higher compared to the typical efficiency of the ICEs of 16% (i.e., from on-
board fuel storage to drive the wheels) [59]. Therefore, the electrification of the bus fleet
can be used as a tool to make a considerable shift from individual vehicles to public
transport that will provide a valuable benefit in eliminating road congestion and reducing
GHG emissions [39], [60]–[63].

1.2 Problem definition

Scientific research, international agencies, public attention, and political interest in
problems associated with transportation, energy and the environment have motivated the
adoption of electric vehicles (EV) and renewable energy generation. Developed countries
globally are designing and implementing strategies for reducing long-term GHGs and
increasing resilience to climate change impact with EV adoption as a priority. Examples
include adopting battery electric vehicles in Denmark, where the goal is to have one
million EVs out of 2.5 million cars by 2030 [64]. In Sweden, the EV has a massive 49.4%
share in 2020, four times more than the 11% share in 2019 [58] [65]. In Norway, 64.5%
of all vehicles sold in 2021 were electric-only vehicles [66].
There is a growing interest in low carbon solutions in developing countries, including

investment in renewable energy resources and EV adoption. An example is Nigeria's low
carbon development plan to cut GHG emissions by 20% by 2030 [67]. Also, the Rwandan
government has committed to reducing emissions by 38% by 2030, which equates to 4.6
million tonnes of CO2 [68]. Kenya's government has set a target of 5% of newly registered
vehicles being electrified by 2025 [69]. Morocco's government intends to save 23% on
energy in the transportation industry by 2030 [70]. Egypt is developing a national e-
mobility strategy to create the first 100 vehicles in August 2022 and build 3,000 charging
stations [69]. Hence, integrating EV charging demand into the power system is a global
challenge.
Several studies in the literature focused on mitigating the challenges of adopting

personal BEVs. This literature explores the incorporation of the energy storage system
with the BEVs parking lot to reduce charging demand pressure on the grid (i.e., enhances
peak shaving and valley filling), reduce the capital investment on the construction of
BEVs charging infrastructure, and minimise operation expenditure [71],[72]. The studies
on the design of BEV charging stations vary in the literature [72]–[78]. The main
objectives of the design of charging stations are to (i) determine the vehicles energy
demand, (ii) ensure that the allocation of charging infrastructure satisfies the BEVs
charging schedule, (iii) minimising the investment in the charging station infrastructures,
(iv) satisfying the BEVs SoC and power distribution constraints, and (v) exploring the
opportunity of the EVs charging stations to deliver ancillary services.

Besides the majority, research efforts are devoted to promoting personal BEVs. The
electrification of transit buses has been considered the most promising approach to fast-
track EV adoption and reduce GHG emissions [79]. Transit bus fleets operate on fixed
dedicated routes to carry passengers using a predetermined daily operating schedule,
making them a possible target for full electrification [80]. Also, the electrified transit bus
system can take full advantage of renewable power generation [81]. In this regard, BEB
technologies with onboard batteries and chargers are a prominent solution for moving
toward zero-emission public bus transit systems [82].
However, in developing countries, EV adoption has been confronted with more

complex challenges because the accessibility to electricity is considered insufficient (for
example, the average access rate to electricity in Africa is 43% [83]). Therefore,
accommodating EVs in the electricity system is a technical, social, and economic
problem. The electricity system will likely have to cope with increases in demand from
EVs and the electrification of other sectors, such as consumers, agriculture, cooling and
industry needs. Moreover, the interest in low carbon transport is beyond just tokenism, as
some countries are building EV assembly plants [84]. Nevertheless, the transition from
ICE internal combustion engines to EVs is not smooth. There are similar barriers to
Europe, such as lack of charging infrastructures, cost and flawed policies [85]. However,
to improve the EV's growing uptake in developing countries, it is vital to provide an
alternative energy source to support the EV charging infrastructures without dependence
on the existing power system such as solar renewable mini-grid.
In developing countries, the electrification of public transport may progress ahead of

individual passenger vehicles because it can be driven by regulations that will encourage
early adoption. Notably, the power infrastructure for public transport fleets is different
for consumers because the charging infrastructures are more nodal [86] (i.e. terminals,
deports and bus stops) that can be supported by co-located renewable energy technology
mini-grids and other DERs. For example, in Lagos, Nigeria, which has about 21 million
inhabitants, there exists the worst traffic congestion and GHG emission predominantly
from automobiles [30], even with an existing bus rapid transit BRT [87]. The greening of
this public transport that combines a favourable GHG solution and does not impact the
incumbent road could be attractive to promote policies that will eliminate high-way
congestion and promote EV adoption in developing countries.

There is already a considerable body of research work addressing the issues of EV

charging, and a common theme is the need to apply demand-side management to control
the timing of these additional power system loads [88]–[91]. Some work has considered
how EV charging loads could be dynamically controlled to contribute to frequency
response services (often via aggregators) [92]–[94]. While offering an important
downstream low voltage solution, this area has not been studied extensively in developing
countries. However, some work has looked into applying off-grid systems to electric
vehicle charging in developing countries. Specifically, in [95], the authors looked into the
possibility of using an electric Tuk-tuk (a three-wheeled vehicle used for public transport
in Asia and Africa) battery charging station in the rural areas of the Democratic Republic
of Congo (DRC).
Moreover, in [96], the authors considered the demonstration of a low-cost electric

vehicle charging station for developing countries using four solar panels of 255 watts
each, batteries, a charge controller, and an inverter. In [75] and [76], the authors
recognised the importance of low-cost renewable energy source charging infrastructures
in developing countries. However, the models are only applicable to small and low-range
electric vehicles, and typically they are challenged in their ability to scale by limited grid
capacity or shortages in a (centralised) generation.
The work of this thesis will be to assess the infrastructure needs and estimate the energy
demand to achieve an eco-friendly city and reduce traffic congestion through the adoption
of electrified bus-based passenger transport solutions. A case study for Lagos in Nigeria
that focuses on the bus rapid transit BRT located there will be a common point of
reference and focus for this work. The transiting of the Lagos BRT system into an
electrified system is critical for improving mass passenger transport in the city while
advancing low emission transport solutions. However, the Nigerian grids are weak, with
the voltage being sensitive to the connection on the new load. The incumbent generating
capacity is not sufficient for the domestic and industrial demand [97]. The overall
available capacity as of 2019 is 3.7GW; however, peak demand is 8.25GW [98].
According to a recently published report by the National Bureau of Statistics, Nigerians
only get 6.8 hours of daily electricity from the national grid [99]. In this scenario, it is
vital to address charging infrastructures that will support the integration of electric vehicle
electrification with little to no dependence on the existing power system.

Previous work on the analysis of the interactions between the transit operation and
energy demand of the BEBs transit system, [29], [70], [78], [101], [102] sheds light on
several challenges, such as but not limited to; (i) the impact of charging strategies and
charging technologies on the transit demand (ii) the effect of charger capacity upon the
operational demand profile and the charging duration of buses; (iii) the impact of the route
traffic situation and other BEB requirements to preserve the operational schedule; and
(iv) the consideration for renewable generation mix to meet the transit demand; (v) the
impact of BEBs transit energy demand on the utility grid (i.e., voltage, transformer's,
feeder's capacity, power balance and security of supply). Hence, due to the lack of security
of power supply in developing countries and the variation in bus transit operation and
utility grid profile, it is essential to carry out a comparative evaluation of various charging
strategies and charging technologies, and investigate how to optimise trade-off of transit
operation, utility grid demand profile, and BEB configuration (battery size, charger
power, and the number of required chargers) in orde to enhance transport electrification,
reduce emission and problematic congestion in the developing countries.
The work of this thesis will outline the performance requirements for a practical system
and determine a suitable charging infrastructure for transit bus operation in developing
countries. The context of this work suggested that large and expensive onboard batteries
are not necessary for city-scale applications, and recharging times are conserved to a
minimum. This will ensure that battery-electric buses can be used for similar duty cycles
as conventional vehicles without compromising the transit operational schedule—the
recommendations are intended to be technically realistic and financially viable in the
developing countries' circumstances.
1.3 Research objectives

With the increasing population of cities, transport demand, individual vehicle usage,
traffic congestion, and pollution, there comes the need to consider the shift from the
existing means of transportation in the cities to a more reliable, healthy and sustainable
means. This thesis will provide supportive research to facilitate the adoption of electrified
bus transit systems and their alignment with the power system in the context of
developing countries by offsetting the present research gaps that challenge the realisation
of electrified bus transit networks as a means to mitigate GHG emission and traffic

congestion in developing countries cities. The main hindrances to adopting onboard

battery-based electric vehicles in developing countries include weak grid, limited range,
and economic viability. Therefore, the development of novel charging strategy, planning
and transit operation models must maintain the state of charge of the BEB on-board
battery, charging infrastructure and power availability constraints while optimising their
infrastructure economic metrics. In light of the above, this thesis will address the
following objectives:
1. To investigate and analyse the impact of traffic congestion on BEB energy

consumption and GHG emissions. This analysis can support transit bus operators to
precisely evaluate the energy consumption characteristics of BEBs in diverse
operating environments and enhance the accurate schedule of the BEB transit fleets.
2. To investigate and model the alternative means and strategy of charging the battery-
electric bus for challenging grids (common in developing countries) that do not
typically support the overnight charging strategy of large battery-sized buses
commonly used in Europe and China.
3. To develop and analyse the techno-economic model of the multi-terminal-based
inductive charging system for a bus transit network. This will primarily be concerned
with capital costs to optimise the length of inductive cable, power transmitters'
location, power transmitter's capacity, onboard battery capacity, and the number of
power transmitters.
4. To estimate the daily BEBs charging power demand for the designed transit-based
charging model.
The electrification of the public transit system that can meet these set objectives can be
based on the array of different charging technologies. These charging technologies can
include a combination of conductive and inductive charging technology, minimisation of
energy usage, reduced electric buses' onboard battery and minimisation of the overall
investment cost.
1.4 Summary of contributions

This thesis makes the following key contributions:

1. This thesis presents a wide-ranging review of the electrification of bus transit systems.
This review ascertains the developments in the research area and finds gaps for future
research works.
2. The development of a mathematical modelling toolkit is developed for BRT system

electrification modelling: The bus transit electrification mathematical formulation
presented in this thesis is distinct from what has been presented in previous similar
research work. This mathematical model considers the critical decision variables such
as location of the charging infrastructure, onboard battery capacity and cost for
allocating charging infrastructure to determine the transit system charging demand
and satisfy the transit optimisation objective. This type of mathematical model
improves the accuracy of the dynamic transit system analysis. Another important
mathematical model is for multi-terminal priority charging presented in this work.
This mathematical model is the basis for the priority-charging model, described as the
aggregate numbers of battery-electric buses that charge at the BEB terminal with the
high, medium, and low charging priority. The advantage of this type of model is the
aggregation of transit energy demand to avoid peaks and valley issues and reduce the
energy use by the transit system.
3. The modelling capability of combined inductive and conductive charging

infrastructure: this thesis presents the use of both conductive and inductive charging
technology for the bus transit system. The system is designed to minimise onboard
battery and reduce bus dwelling time due to the long charging duration. Such a system
is developed to manage the system energy demand and ensure that the operational
service is reliable by optimising the charging station infrastructure: charger operating
capacity and chargers' number. Integrating a conductive and inductive charging
system that can reduce on-board BEB battery is considered an advantage on already
challenged grids
4. Energy consumption modelling related to GHGs for Lagos: This thesis presents a
method for calculating the average energy consumption rate for bus transit fleets
while considering route traffic conditions. This model overcomes the disadvantage of

using the manufacturer's nominal BEBs energy consumption value. Furthermore,

based on the Lagos BRT case studies, a case study of carbon emissions savings of
electrified BRT systems is computed by comparing the consumption of diesel and
electric buses in various traffic scenarios with the associated GHG emission figure.
5. The design and economic modelling of inductive charging only infrastructure system
that provides some information on the costs for the Lagos BRT: The terminal-based
inductive charging infrastructure's design and economic modelling reduce the size of
the electric bus's onboard battery while reducing the capital cost of the transit system.
The inductive charging infrastructure model determines the size of the batteries, the
location of the transmitter, and the length of the inductive cable. This study
demonstrates that financial savings are attainable if the inductive chargers are located
a nearly equal distance apart.
1.5 Thesis outline

The remainder of this thesis is structured as follows:
Chapter 2 introduces bus transit systems, types of transit buses and their characteristics,
transitways, types of chargers and chargers' technologies, and bus schedule system,
referred to in the remaining part of this thesis. This review chapter also examines the
battery-electric buses' energy consumption, electric transit system and charging
strategies, resource modelling for electric vehicle charging systems, and the impact of
low carbon transport systems. The review indicates a considerable body of research work
addressing the issues of EV charging, and a common theme is the need to apply demand-
side management to control the timing of these additional power system loads. This area
has been studied extensively but not in developing countries. However, research into
providing an alternative energy source such as solar renewable mini-grid to support the
EV charging infrastructures with no dependence on the existing power system in the
developing countries (especially where there is a lack of security of power supply) needs
more comprehensive study.

Chapter 3 provides a thorough discussion of the various mathematical modelling stages

as well as the objective equation and constraints for the implementation optimisation
algorithm. The design of the transit fleet routing model, the BEB energy consumption
model, the BRT transit fleet charging infrastructure model, and the estimation of transit
energy demand are all covered in this chapter. The mathematical formulation for
conductive charging, inductive charging, multi-terminal charging, and priority charging
with respect to the transit system is also presented in this chapter. This will form the
backbone of the modelling methodology for studying the bus system electrification
considered in this thesis. A contextualised case study based on the Lagos bus rapid transit
(BRT) is introduced in Chapter 4, which will provide the technical and operational
features for applying the developed methodology. In addition, this chapter describes the
Advanced Vehicle Simulator (ADVISOR) modelling tool and presents data based on
Lagos BRT's current operations. The information defines the route speed, route length,
terminal locations, transit schedules, route elevation, and route traffic system. The BEB
energy usage per km is estimated using the ADVISOR and transit data. A PSO algorithm
is introduced to provide optimised solutions to the transit network's conductive, inductive,
and hybrid (conductive and inductive) charging systems. The number of chargers,
charging point locations, charger power capacity, and onboard battery capacity are
determined. The system's energy demand is also assessed. The considerable CO2 emission
reductions that can be achieved are also examined in this chapter. The findings of the
study in this chapter can be used to promote the decarbonisation of transit networks in
(mega) cities across the world and provide a basis for planners, transportation experts,
and policymakers to assess the effect of switching to a low-carbon electrified BRT
system.
Finally, Chapter 5 draws together the overall conclusions of the thesis and provides
some recommendations for future research work.
1.6 Publications arising from the work of this thesis

1. Asaolu, A., and Galloway, S., 2020, August. Optimal design of conductive and
inductive charging system for bus rapid transit network. In 2020 IEEE PES/IAS
PowerAfrica (pp. 1-5). IEEE.

2. Asaolu, A., Galloway, S. and Edmunds, C., Allocation of the inductive charging
system for bus rapid transit network. In 2020 IEEE International Smart Cities
Conference (ISC2) (pp. 1-8). IEEE.
3. Optimal design of conductive and inductive charging system for bus rapid transit
network (Abstract publication and presentation conference). http://www.ieee-
manchester.org.uk/wp-content/uploads/2019/11/Session-2-2-Adedayo-
Asaolu.pdf

Chapter 2: literature review
2 LITERATURE REVIEW
The increasing population of cities globally leads to traffic congestion, increasing

vehicular emission, and causes many hazards to human health. To this end, developed
nations are transforming and growing investments in public transit systems that include
upgrading public rail systems, trams, subways, and public transit buses. In developing
countries, the increasing number of urban dwellers is expected to increase public and
freight transportation demand levels in the nearest future [102]. In contrast, many
developing countries cannot prioritise investment in public transit solutions such as the
metro and rail systems despite facing the same challenges. While it varies from country
to country, in such cases, the incumbent systems are already challenged (e.g. frequency
of electricity blackouts, ageing infrastructure), and this, together with a lack of
investment, means that they cannot transition at the same rate as developed nations in this
regard. Hence, with less investment in heavy infrastructure, light solutions are likely to
be more readily available in developing countries. On this basis, the bus rapid transit
system's adoption is expected to proliferate the public transit system in these low
economic countries [103].
Without change, it is also expected that the transportation sector's GHG emissions will
increase to higher emission rates [104]. This increase in GHG emissions will, in turn,
necessitate the need for deploying innovative approaches that can meet the growing
transportation demands while addressing their environmental concerns [102] [103] [104].
The adoption of electrified transportation systems is anticipated as a promising solution
that could considerably contribute to the decrease in environmental pollution [105] [33].
In this regard, this chapter presents a background on the bus transit system and a literature
survey on the challenges that face the seamless adoption of electrified transport networks.
The remainder of this chapter is organised as follows: the first section covers transit
systems in detail, including substantial information on the BRT system, the sustainable
BRT system, and BRT components. The next sections go further into electric bus

configurations, BEB charging strategies, charging technologies, and battery technologies.

This chapter includes a review of the literature on battery-electric bus energy
consumption modelling and the impact of EV charging on the power distribution system,
which will motivate and position this thesis's work. Additionally, some interesting
literature review on off-grid charging electric vehicles is included in this thesis because
the developed model also estimates the transit system energy demand. This transit energy
demand profile can be used as the starting point for resource modelling for the bus transit
systems, as suggested for future research work. And the final section summarises the
identified gaps in the work of previous authors.
2.1 The Transit System

A transit system is defined as coordinating buses and trains that run on predetermined
routes and are open to the general public [106]. Unlike individual passenger
transportation, public transportation is normally run on a schedule and specific routes.
Trolleybuses, fast transit (metro, subway, underground, and buses), trams, trains, and
urban buses are well known public transportation systems. Most public transportation
systems operate on a set schedule along fixed routes with separated lanes. The urban
environment is deemed favourable for transit because it provides necessary conditions for
its effectiveness, such as high density and significant short-distance travel demands.
Because transportation is a shared service, it might theoretically profit from economies
of scale connected to high mobility demands and high densities [107]. The increased
demand that makes public transportation services viable and profitable is a critical benefit
of public transportation in the urban setting.
In [108], the authors carried out a cost assessment to demonstrate the transit system's
advantages (i.e. metro, light rail transit LRT, or bus rapid transit BRT). The evaluation
carried out in this work shows a significant reduction in passengers’ expenditures, travel
time, per capita fuel usage, and CO2 emission per passenger trip. A considerable shift
from individual passage vehicles to transit mode requires the system to be flexible and
relatively low cost while offering significant environmental benefits, which can only be
achieved using a bus rapid transit (BRT) system [109]. The BRT provides opportunities
for cities seeking efficient and cheap transit solutions that meet increasing urban growth

in developing countries. While many developing countries have made plans and proposals
for rail development due to its high speed capacity, these plans have not been
implemented due to the high implicit costs [110] [103]. Recent studies have shown a vast
increase in the adoption of a BRT system in cities across the globe because it has an
efficiency level that mimics the LRT and can be implemented within a short period to
reduce cost [113]–[115]. The next section presents details on the application and
advantages of the bus rapid transit system.
2.1.1 The bus rapid transit system

The common transport challenge confronting most metropolitan areas can be summarised
as traffic jams, pollution, delivery delays, and inadequate pedestrian facilities. A large
body of research has identified the BRT system as a cheap and quick means of solving
this problem (for example, Mexico City's Metrobús [116]). The BRT system has been
determined to compete with expensive, sophisticated alternatives (metro, light rail, and
subway), and it is both fast, frequent, and cheap [117]. The BRT often has dedicated bus
lanes to help reduce congestion and offer more efficient vehicular transport solutions than
private car ownership. Dedicated bus lanes can double, even triple the bus speeds and
move more than four times as many passengers per hour as regular-lane traffic [111].
In [120], the authors highlight that "the BRT is part of the response to continued rapid
urbanisation effects ongoing in most countries, particularly the larger cities in the
developing world." Also, the study in [121] states that the BRT can deliver an excellent
mass public transportation service that costs "between 4 and 20 times less than an LRT
(Light Rail Transit) system and between 10 and 100 times less than an underground type
system". The BRT is a transportation mode that incorporates buses and rail transit speed
flexibility in terms of operation. The working system's idea is straightforward: the buses'
traffic is made through a dedicated lane – this strategy makes possible a quicker, safer,
and better bus service. Moreover, the BRT incorporates features mainly used in
underground systems, such as off-board fare collection, platform-level boarding, and
articulated vehicles.

In [109], the BRT is described as an innovative transit system that integrates the quality
effectiveness of the metros system with the flexibility and relatively low cost of buses
while offering significant environmental benefits. In an exceptional circumstance, the
BRT systems can achieve marked levels of efficiency in terms of speed, capacity,
passenger comfort, and convenience comparable to rail-based systems and, remarkably,
can be built to reduce cost and construction time. Certainly, BRT can provide
opportunities for cities in developing countries, where there is a growing demand for
affordable transit solutions that meet the needs of increasing urban population growth and
mobility demand.
In summary, the BRT is considered a viable option for alleviating traffic congestion in
developing countries' cities because the capital and operating costs are relatively lower
than the rail-based transit system. Another significant advantage is the interconnectivity
and ability to operate on typical streets with inexpensive bus stops. The implementation
period is less in comparison to rail-based systems [122].
Table 2-1 Key benefits of the bus rapid transit system (Adapted from [96])
BRT Impact Strategies Empirical evidence

1. TransMilenio BRT's
GHG and local air 1. Reduce vehicle-kilometres
implementation in Bogota,
pollutant emissions travelled (VKT) by shifting
combined with new fuel quality
reductions. passengers to high-capacity
regulations, saves about 1
BRT buses
million tCO2 per year [123].
2. Replace/scrap older, more
2. Mexico City's Metrobús Line 1
polluting traditional
achieved substantial carbon
vehicles.
monoxide reductions, benzene,
3. Introduce newer technology
and particulate matter (PM2.5)
BRT buses.
[116].
4. Improved driver training
results in improved driving
cycles, leading to reduce

emissions and fuel

consumption.
Reduced exposure 1. Improve vehicle 1. In Bogota, after implementing

to air pollutants. technologies, cleaner fuels, TransMilenio (BRT), the SO2
and reduce ambient air emissions reduce by 43%, NOx
pollution citywide or inside declined by 18%, and a 12%
the BRT buses. decline in the particulate matter
2. Reduce passengers' [123].
exposure to air pollution at 2. Reduction in local air pollutants,
bus stations or inside the specifically particulate matter
bus by reducing travel for Metrobús Line 1 in Mexico
times. City, is expected to eliminate
about 6,000 days of lost work
and 12 new chronic bronchitis
cases, and three deaths per year,
saving an estimated USD 3
million per year [124].
1. In Johannesburg, the BRT users
Travel time savings 1. Segregated busways that
save an average of 13 minutes
detached BRT buses from
each way [125].
mixed traffic;
2. In Istanbul, a typical Metrobüs
2. Prepaid level boarding and
(BTR) passenger saves 52
high-capacity buses speed
minutes per day [126].
passenger boarding.
3. Traffic signal management
and high-frequency bus
service minimise waiting
times.
1. Bogota's TransMilenio (BRT)
Improve road safety 1. Ease pedestrian crossings •
reduces on-road crashes and
– reductions in High-capacity BRT buses
injuries along the main corridors
fatalities and reduce VKT.
[127].
crashes

2. Segregated bus routes 2. On average, in Latin America,

decrease interaction with the BRTs have reduced fatalities
other vehicles (i.e. mixed and injuries by over 40% along
traffic). their routes.
3. The BRT reduces on-the-
road competition, which is
expected to change drivers'
behaviours by improving
training.
Increased physical 1. The spacing of BRT stations 1. The users of Mexico City's
Metrobús increase their walking
activity tends to require longer
duration by about 2.75 minutes
walking distances than all
per day.
other motorised modes
2. Users of the Beijing BRT walk
except Metro.
additional 8.5 minutes daily.
2. Higher operation speeds
motivate BRT users to walk
to stations.
Table 2-1 gives a summary of the benefits of BRT. It shows that implementing BRT
systems in the cities has provided a range of benefits, including reducing travel time,
reducing congestion, creating a positive environment in the cities, and improving safety.
Other benefits include shifting from individual passage vehicles to public transit and
improving human efficiency (gain in a working hour), which is a key consideration for
this thesis's work. The Metro bus (BRT) in Mexico City reduces CO2 by 45%, Benzene
by 69%, PM 2.5 by 3%, and road accidents reduce by 80%. The BRT's introduction in
Istanbul, Turkey, saved each passenger 316 hours per year, removed 80,000 individual
passenger vehicles from the road, reduced CO2 emissions by 623 tons per day, and cut
the accidents rate by 64% [128]. In Johannesburg, BRT passengers save an average of 13
minutes per trip [125], and similarly, in Istanbul, BRT passengers save an average of 52
minutes per trip [126] [128].
The BRT passengers in Mexico City have a walking increase of about 2.75 minutes
more per day on physical health activities. Also, in Beijing, BRT passengers have added

8.5 minutes of daily walking time. Furthermore, in terms of physical activities, the BRT
users in Mexico City walk an average of 2.75 minutes additional per day than before
implementing its BRT system. Similarly, as a result of the BRT system, commuters in
Beijing have added 8.5 minutes of daily walking time [129]. According to the World
Health Organisation, it is beneficial that adults aged 18-64 years should engage in a
minimum of 150 minutes of moderate-intensity aerobic physical activity throughout the
week. Alternatively, do at least 75 minutes of vigorous-intensity aerobic physical exercise
throughout the week or a combination of moderate and vigorous-intensity activity [130].
It means that commuters' short distance walks to and from BRT bus stops contribute to
their health.
Considering the above and with many of the same challenges identified, implementing
sustainable BRT systems in developing countries' cities will form a specific focus for this
thesis's work.
2.1.2 The sustainable BRT system

Transportation is the backbone of growth in developing economies [131]. However, along
with this comes a significant source of environmental problems, as its share of pollutant
emissions, energy consumption, and greenhouse gas emissions have grown [132].
Problems are exacerbated by vehicle travel levels rapidly outstripping existing
infrastructure systems' capabilities, leading to traffic congestion and even more fuel use
and air pollution than would otherwise occur. The problems are particularly acute in the
developing world's largest cities (such as Lagos, Nigeria [133], Delhi and Mumbai in
India [134], Nairobi in Kenya and Rio de Janeiro in Brazil [135]). Growing populations
and high traffic densities with vehicles of all types mean significant congestion, slow
travel speeds, human exposure to much-polluted air, and high rates of morbidity and
mortality from traffic accidents [136]—the increasing desire for reforms to the current
levels of car usage in most global developing countries cities needs to change. Any
intervention should reflect the capacity to reduce commuter travel time, reduce the
proliferation of cars and motorised trips, and reform existing often unreliable public
transport to attract choice passengers from car usage to transit usage. This shift and
recalibration of mode share towards efficient transit-dependent mobility patterns are
considered ingredients of sustainable transportation in cities [137].

Existing historical transportation approaches are unsustainable in the long term because
of:
1. the limited nature of petroleum reserves,

2. Volatility in the fossil fuel markets causes fuel/energy insecurity,
3. the negative impacts of petroleum-based emissions on air quality, and
4. levels of traffic congestion in many cities.
While there are other (equally important) issues, including secondary emissions and
traffic deaths, these points above are the most relevant to this thesis's work [138].
The World Bank also defines what it refers to as the three pillars of sustainable
transport as (1) the economic and financial sustainability (ensuring that transportation is
cost-effective and continuously responsive to changing demands); (2) environmental
sustainability (emphasis on better planning of land use and stricter management of
demand, including the use of pollution and congestion charges to incentivise public
transport); and (3) the social sustainability (designing transport strategies to provide
social inclusivity with improved accessibility to employment, education, and health
services) [139]; and (4) the mobility equity-not everyone can afford to buy a car or learn
to drive . It is difficult to disagree with these points and the advantages of a sustainable
BRT system continue to be important. Making the BRT sustainable requires deliberate
effort from transit operators and regulating agencies to address the outlined problems.
Figure 2-1 Strategies for Sustainable BRTs [137]

Figure 2-1 illustrates the strategies for transforming and implementing a sustainable BRT
system. These are grouped into three categories (vehicle technologies, road, and human)
based on the component of the transit system infrastructures. Integrating electric drives
and renewable energy technologies into the transit system is expected to curtail the
existing diesel buses' adverse environmental effects [140]–[142]. In the second category,
having dedicated busways [143], [144], a smart transit scheduling system [145], and the
electric bus re-charging technology that eliminates charging downtime will make the
transit system perform beyond environmental sustainability but address the issues with
road congestion and financial sustainability. In the third category, creating emission zones
(that only permit EV drivers to use problematic congestion areas), and creating charging
zones (pay to use) will reduce congestion and motivate a shift from an individual
passenger vehicle to the transit mode. Also, increasing fuel tax costs and reducing transit
fleet costs can reduce personal passenger vehicles' use and increase BRT's attractiveness.
This work of this thesis touches on all three of these categories. However, the primary
focus is on the BRT system's electrification to support the environmental sustainability
of public transportation systems. These sustainability strategies would encourage public
transport usage and help manage the common congestion in the cities of many developing
countries.
2.1.3 Features of BRTs

The BRT system's main components include segregated busways, bus stops, bus
terminals, high-capacity buses (rolling stock), a passenger boarding platform, a real-time
passenger information system, a bus route map, and the control centre (intelligent unit)
branding, and bike-share facility. These identified BRTs components are further
discussed below:
2.1.3.1 Segregated busway

In BRT corridors, segregated busways are meant to allow commuters to transit more
quickly [146]. The busway differs from the bus lane in that the bus space is physically
separated from the traffic (see Figure 2-2). A raised block/pillar separates the busway
from the traffic, while a painted line separates the bus lane. Because the bus lane does not

restrict traffic from entering the system, it can impair free flow and lengthen bus travel
times [147].
Figure 2-2 El Metropolitano's Trunk Corridor [148]
The BRT lane is separated from other traffic using blocks in the Peru El Metropolitan's
Trunk Corridor, as shown in Figure 2-2. This segregation type is provided with a
dedicated and continuous lane for buses to travel against the road's block edge. The bus
stops are appropriately built to provide simple access to the neighbouring streets, but as
demonstrated, they limit the opportunities for bus overtaking.
2.1.3.2 Bus terminal and bus station

A bus terminal, sometimes known as a terminus, is the starting or ending point of a bus
route and the location where buses stop, turn, or reverse and wait before continuing their
trip back. At bus terminals, passengers board and depart from vehicles. It is also an ideal
location for managing transit services [149]. The terminus is the starting point for a
timetable. Terminals can be in bus stations, interchanges, bus garages, and bus stop,
among other places. The terminals can begin and end at the same location or possibly in
multiple locations depending on the route. A bus transit terminal may be located near
other vital facilities like universities, retail malls, or hospitals. It may be part of a
transportation hub or 'interchange.' A bus stop or loop on a residential street can be
considered a minor terminal [150].

Figure 2-3 Bus Terminal in Oshodi, Lagos [151]
Although the terms "bus terminal" and "bus station" are sometimes interchanged, the
latter is usually more accurate because some routes pass by the station without stopping.
The word "bus station" usually refers to a location off the main path with only the most
basic passenger amenities. A terminal might be a fully equipped bus station or a simple
road intersection. In many cities, most passengers begin and conclude their journeys at
bus stations, which may account for a significant amount of the money generated by
operators [152].
2.1.3.3 The bus stop

A bus stop is a designated location where buses stop to allow passengers to board and
disembark. Bus stops are typically built to reflect utilisation, with shelters, seating, and
possibly electronic passenger information systems in high-traffic areas and a basic pole
and flag in low-traffic areas. In some areas, bus stops are crowded together to form
transportation hubs, allowing for interchange between routes from surrounding stops and
other modes of public transportation to maximise convenience [153].

Figure 2-4 Lagos BRT Bus stop [154]
For operational purposes, there are three types of stops: scheduled stops, where the bus
should stop regardless of demand; request stops (or flag stops), where the vehicle will
only stop on request; and hail and ride stops, where the vehicle will stop anywhere along
the designated section of road on demand [153]." Discharge/set-down only" or "pick-up
only" may be the only restrictions at some stops. Some stops may be designated as "timing
spots," where the vehicle will wait ahead of schedule to guarantee proper synchronisation
with the schedule. Skip-stops are sometimes used to boost efficiency and decrease bus
stops in dense metropolitan areas with high bus volumes. Specific stops in distance or
zone-based fare collection systems can also be used to establish fare stages.
2.1.3.4 Bus depots

A depot is a transportation system's operating hub. There includes parking, bus servicing
and maintenance, an administrative role, and employee facilities. A fully enclosed depot
is referred to as a "garage." An operator may have one or many depots, depending on the
size of its fleet and regional reach [155].
Figure 2-5 Lagos BRT Bus Depot [156]

Buses are assigned to routes, and crews are assigned to each job at the depot, dispatching
buses on time and processing cash payments from drivers. These facilities also contain a
signing-on office where crews report for duty and a cash office where drivers pay in
revenue earned while on the job. A canteen and medical facilities for employees, off-duty
personnel, backup bus drivers, and overnight lodging for crews at larger depots are
probably appropriate [155]. Figure 2-5 shows a typical example of a BRT bus depot in
Lagos, Nigeria.
2.1.3.5 Boarding platform

Station platforms should be level with the bus floor for quick and easy boarding, allowing
wheelchairs, disabled passengers, and baby carriages to board with minimal delays. These
BRT stops are distinct from street-level bus stops since stopping outside designated
platforms or having standard buses stop at high-level platforms is challenging for high-
floored buses [155].
Figure 2-6 A typical boarding platform for easy entry/exit [157]
As shown in Figure 2-6, the boarding platform should allow for smooth passenger entry
and exit and easy entry and exit for the elderly and physically disabled.
2.1.3.6 Fare payment system

The BRT system uses a contactless off-board fare payment system like that of rail
systems. Passengers enter the bus terminal by swiping their pre-loaded BRT smart card

over the turnstile at the entrance, and they board the appropriate bus as soon as possible,
as illustrated in Figure 2-7. Hence, using this fare payment method reduces boarding time
while reducing trip time.
Figure 2-7 Transmilenio prepaid boarding and passenger information in Bogota

(Adopted from [158])
2.1.3.7 Rolling stock

The rolling stock of a BRT system is primarily articulated or bi-articulated buses with a
passenger capacity of 80 to 250 people. These large vehicles can replace informal
minibuses, reduce regular vehicle mileage, and reduce per capita fuel consumption and
pollution associated with city commutes. To keep the system's emissions low, the buses
in the BRT system must adhere to emission criteria. For compliance with the standard
BRT diesel vehicles, particulate matter traps, ultra-low-sulphur diesel fuel, and selective
catalytic reduction are required [159]. Particulate matter emissions from Euro IV and V
authorised vehicles are lowered by twice as much, according to the efficiency level of
decreasing particulate matter emissions from introducing Euro VI buses and USA 2010
emission requirements [160]. By institutionalising and maintaining compliance with these
stated transportation operators' criteria, particulate matter (PM), nitrogen oxide (NOx),
CO2 emissions, air pollution, and global warming-related externalities can all be reduced.
The recent adoption of electric buses into the transit system is projected to affect the
transportation system by reducing pollution significantly. The advantages of adopting
battery-electric buses (BEBs) in a transportation system go even further in terms of air
quality, noise reduction, and energy efficiency [161]. With sustainable BRTs in mind,

greening this form of public transportation while incorporating a low-GHG solution may
be an appealing choice; thus, this study investigates the advantages of electric buses and
the component of electric bus transit.
2.2 Electric buses

An electric bus "can operate by different degrees of electrification that depend on the
configuration of the propulsion system" [14]. Some can be continuously fed by external
sources – for instance, a trolleybus powered by overhead wires. Other ones can store the
electricity onboard, typically in batteries. Several electric buses fit into this category;
examples include:
 Hybrid Electric Bus (HEB): The Electric Motor (EM), as well as an Internal
Combustion Engine (ICE), provide the traction power of this technology. In some
cases, the HEB battery can also be charged using plugin technology to connect
the grid [14].
 Fuel Cell Electric Bus (FCEB): Uses hydrogen fuel cells to generate electricity
onboard during operation [162].
 Full Battery Electric Bus (BEB): This vehicle uses the energy stored in the bus's
onboard battery to provide the electric motor propulsion [162]. Power is
transferred to the battery-electric bus through electric charging systems, while
regenerative braking is used to recover kinetic energy during the vehicle operation
[12].
Environmental gains (no exhaust pipe emissions), better air quality, and lower total cost
of ownership (TCO) are the key advantages of electric buses over diesel alternatives
[163]. The lower TCO is attributed to the lower cost per kilometre of operating on
electricity as compared to petrol, while the battery cost currently results in a higher
average acquisition price. Because of these advantages, policymakers are setting goals to
promote acceptance, such as the Netherlands' requirement that all zero-emission vehicle
(ZEV) public transportation buses be sold by 2025, with a 100% ZEV fleet by 2030 [164].
Bus makers and transportation organisations signed the European Clean Bus Deployment
Initiative [131] to formalise a joint commitment to promote the implementation of clean

buses. Compared to internal combustion engines, hybrid electric buses, and fuel cell
electric buses. Until recharging the onboard battery, the BEB's driving range is limited.
Figure 2-8 The defining components of different electric buses (Source: [12]).
Nonetheless, they are more efficient and environmentally friendly. Figure 2-8 shows the
classification of Battery Electric Buses. A hybrid bus combines two power sources: a
conventional diesel engine and an electric motor. Fuel cell electric buses are hybrid
electric buses that are powered by fuel cells and batteries (s). An electric motor is used in
a battery-electric bus powered by an onboard battery charged by mains electricity.
Electric buses emit no tailpipe emissions, earning them the designation of "zero-emission
capable." Hence, integrating BEBs into the public transit system is considered a game-
changer for polluted cities in developing countries.
The configuration of the battery-electric bus includes auxiliary devices (AUX), an

electric motor (EM), a transmission system (TX), and a final drive (FD), as shown in
Figure 2-9. This indicates where these different elements typically reside to enable the
energy for a BEB to be stored in a battery and then provided to the electric motor. Electric
charging systems transmit energy to the vehicle, and regenerative braking is utilised to
recover kinetic energy during operation. As indicated in Figure 2-9, the energy for a BEB
is stored in a battery and then delivered to the electric motor. Lithium iron phosphate,
lithium-titanate, and nickel cobalt manganese lithium-ion batteries are the three most
prevalent types of batteries used in BEBs [165].

Figure 2-9 Battery Electric Bus Configuration (Adopted from [12])
Table 2.2 summarises the characteristics of popular types of BEB. The common brand
BYD has produced various models. BYD manufactured most of the onboard battery-
electric buses used globally in the public transit system [157]. In Europe, the most
commercialised EBs are Volvo, Solaris BYD and Alexander Dennis (BYD ADL) [166].
Table 2-2 Characteristics of Electric Transit Buses
Model Type Battery Range (km)

capacity (kWh)
BYD ADL Enviro200EV [167] Single-deck bus 348 250
BYD ADL Enviro400EV [167] Double-deck bus 382 250
Enviro500EV [167] Double-deck bus 660 400
BYD 23' Coach [168] Single-deck bus 121 200
Solaris Urbino 12 [171] Single-deck bus 145 100
Volvo 7900 [172] Single-deck bus 76 200
Electric buses are becoming more common in public transportation fleets around the
world. China was the first country to adopt electric buses, and it took a few years for other
countries to follow suit. Approximately 98% of the world's electric buses are currently in

operation in Chinese cities [166]. In comparison to 2017, the European electric bus
industry grew by 48% in 2018. Battery-electric buses and plugin hybrids, trolleybus IMC,
and fuel cell buses were among the 4,000 electric buses operating in Europe in 2019 [166].
The electric bus market in India (70,000 buses sold in 2017) has much potential. India
will account for more than 10% of worldwide yearly demand for electric buses by 2025,
according to Interact Analysis, which is more than Europe and North America combined.
Electric buses accounted for roughly 0.5% of the overall public transit bus market in the
United States at the end of 2017. Electric buses were in service in about 9% of all
transportation agencies. The US government is also supporting and funding research into
public transportation electrification. [173].
Table 2-3 Comparative evaluation of HEB, FCEB and DB ( [12])
HEB FCEB BEB
Purchase Price +50% +200% +100%
Maintenance Cost More Much More Less
Operation Cost Less Much More Much Less
Infrastructure More More More
Range Less Less Much less
Weight More More More
Refuel Time Less More More
Emissions -20% -75% -85%
Energy Efficiency Better than DBs +150% +450%
Electric buses are already becoming the bus of choice for many cities and public transport
providers worldwide. Table 2-3 provides a comparison between diesel buses and electric
buses across a range of categories. The BEBs present several advantages: the operation
and maintenance costs are low, the energy efficiency is very high, and the low emissions.

Therefore, the BEBs appear as the best and most viable option for electrifying public
transportation. Nonetheless, the technology still faces some challenges regarding the
range, weight, and refuel time factors. In counterbalancing the limited BEB driving range,
three operation conceptions have been proposed: battery swapping, wireless charging,
and onboard battery charging [81] [174] [33]. In addition, the concept of charging the
onboard battery of a battery-electric bus is presently under wide-ranging research and
development as a necessary means for the extensive–scale adoption of electric bus fleets
[175]. In this regard, two types of BEB charging–based systems have been proposed:
overnight and opportunity. The difference between the two types is the trade-off in the
BEB battery capacity, i.e. driving range, and the required rated power of chargers, i.e.,
charging time [81] [82].
BEB systems that rely solely on overnight charging require chargers that have
relatively smaller power ratings. Overnight BEB chargers are installed at the bus
depot/garage to provide a long time charging for a large battery capacity, compared to
opportunity–based BEB. However, previous studies showed that overnight–charging
BEB systems might require an increase in the BEB fleet size relative to the current diesel-
based fleet to maintain the predefined operation-scheduling timetable of transit networks
[82]. In contrast, opportunity–charging BEB systems use on–route high–powered
chargers to provide regular and short time charging for BEBs equipped with relatively
smaller battery capacity [176]. On–route charging systems use mainly automated
overhead pantographic arms to charge the BEBs in the transit hub(s) (i.e., bus terminal(s))
or along the transit routes [177]. Several worldwide demonstration projects showed that
on–traffic charging technologies could sufficiently charge BEBs within 4–6 minutes
[177].
In contrast to other research works that focus on the adoption of individual passenger
EVs. The bus transit network's electrification (especially BRTs) is considered a key
component for the cities in developing countries from transportation challenges because
it can reduce traffic congestion and GHG emissions, and it is often within the control of
local authorities and policymakers to influence this change. Hence, the research work of
this thesis will focus on the challenges enumerating the various charging options,
technologies, and strategies to enhance the BEBs' driving range in the developing country

context. To date, these have not been fully addressed despite the progress made
internationally.
2.3 Options for BEBs charging

When dealing with BEBs, there are three main types of charging options: static,
stationary, and dynamic [178] [179] [180]. These are as follows:
 Static strategy: this is usually used when the BEB will be parked for an extended
period, such as overnight at a depot. The vehicles are charged during this period.
This method enables a longer re-charging time. Since the chargers are designed to
have charging powers in the range of 40 – 120 kW, the charging infrastructure
costs are lower than other charging strategies [181]. Those vehicles normally have
a regular driving range of around 300 kilometres which allows for day-to-day
service like diesel buses without the need for re-charging. On the other hand, the
battery's expense – due to its high capacity – significantly raises the price of such
buses [182]. Since such vehicles are heavier than regular buses, they can only
accommodate a small number of passengers.
 Stationary strategy: when the vehicle is stationary, the BEB is charged for a
limited period. As a result, charging must be done quickly – in the range of 3 to
10 minutes [183]. As a result, charging powers range from 150 to 600 kW, and
high infrastructure costs. Nonetheless, this technique allows for smaller batteries
to be used in the buses and as a result, they are less costly and lighter. Furthermore,
the driving range becomes nearly limitless when charging stations are strategically
located along the road.
 Dynamic Charging Strategy: this charging approach allows the vehicle to be
charged when it is in motion. This can be accomplished by inductive power
transfer (IPT) [57] or catenary-based fast-charging systems. Although the
associated infrastructure costs are substantial, the technology reduces the BEBs'
onboard battery.

Table 2-4 Summary of BEB charging possibilities (Adopted from [184])
Strategy Static Stationary Dynamic

Mid-day opportunity Mid-day opportunity
Typical Overnight charging at
charging due to charging due to
application the depot
facilities in the route charging lanes
30 up to 150 kW (for
Charger 150/300/450 kW (fast
buses with high 25-150+ kW
type charging)
range)
Charging Mostly pantograph
Mostly plugin WPT
technology Plug-in (less common)
Typical
100 – 300 km/day 200 – 500 km/day 100 – 200 km/day
Range
Refuel
3 – 8h 3 – 10 min 10 – 15 min
Time
1 Low battery costs 1 Low battery costs
Needs dedicated 2 High charging 2 Very high
Costs
infrastructure (e.g. infrastructure costs infrastructure costs
drivers
depot or car park) 3 Medium 3 Very low
maintenance costs maintenance costs
The features of the different BEB charging options are summarised in Table 2-4. The
Static approach is widely used at the depot (overnight charging) and is the most used in
developed cities with stable grid supply. The BEBs, in this case, have a large battery used
during the regular operating schedule and parked at the depot overnight to charge in this
situation. The possibility exists to provide co-located renewable energy (especially solar)
during the daily operational schedule; stationary and static methods can be applied to
developing countries and cities with power supply security challenges. The co-located
renewables can help offset domestic electricity costs, improve resilience to power "black-
outs," and boost green credentials because some of the power originates from low-carbon
sources.

The work in this thesis will investigate the possible trade-off between charging options
based on cost, BEBs battery sizes, charger location and the charger's power capacities.
These features are relevant to a range of stakeholders in this setting.
2.4 Charging strategies

Electric buses typically have a shorter range than diesel buses, and a good question that
municipalities, transport operators, and manufacturers are still facing is where and when
to charge them. Electric buses can be recharged in three different operational scenarios
presented in this section: overnight charging, Opportunity Charging, and In-Motion
Charging. Each of these categories is described in the following:
2.4.1 Overnight Charging

Buses are parked at the depot after the regular schedule is completed. The bus, for
instance, departs at 4 a.m. and returns at 8 p.m. The depot time is used for maintenance,
cleaning, and preparation, and an electric bus can use this period to recharge its batteries.
The time to fully recharge depends on the battery capacity and the power output of the
transformer. Therefore, an Overnight Charging strategy uses this time in the depot to
recharge the bus for operation the following day. For this strategy to work, these vehicles
would need large batteries. For example, the 2018 EVOBUS eCitaro is a 12 m long
Battery Electric Bus (BEB) with lithium-ion batteries with a capacity of 243 kWh that
lasts in a worst-case summer scenario for 150 km [185].
2.4.2 Opportunity Charging

Opportunity charging describes operational cases in which buses are recharged at the
depot and designated charging stations throughout the network. Buses do not have to
drive back to the depot to recharge. Time and energy can potentially be saved. As a result,
smaller batteries are needed for buses [186]. However, it is vital to schedule the buses to
have sufficient energy to meet the route demand. Backup solutions for different scenarios
must ensure stable operation in case of incidents, construction works, or detours. Also,
recharging stations can be equipped with inductive charging pads to save time. As soon
as the bus occupies the station, it is detected, and recharging is initiated.

In Eindhoven, the Netherlands, Heliox fast charging equipment was strategically

delivered at the bus depot to create the most efficient and cost-effective option. Both the
opportunity and overnight chargers connected with the buses automatically via a roof
mounted pantograph, resulting in a robust and time-efficient charging system. Heliox
provided 12 Heliox FAST DC 2x30 kW chargers with 24 separate charging stations for
overnight charging. As a cost-cutting measure, the ten Heliox 300kW opportunity
chargers were automatically dropped at night. Heliox's latest innovation, the SMART
Switch, was used to maximise charging intervals. The SMART Switch allows charging
two buses with a single charger while automatically switching from the fullest to the
emptiest bus [187].
2.4.3 In-Motion Charging

In-Motion Charging enables electric buses to recharge while moving. This strategy has
the advantage that no extra time for recharging must be considered during scheduled
operations. When set up smart, the electric bus has access to In-Motion Charging that run
perpetually (for example, the online electric vehicle OLEV presented in [188]). As with
opportunity charging, the flexibility is restricted due to the fixed lines that must be called
regularly to ensure that the state of charge (SOC) is always high enough to reach the next
power lane. This strategy is usually combined with an overhead catenary power supply
(a typical example of a hybrid trolleybus proposed for Berlin [189]). Some test projects
worldwide equip roads with copper coils to wirelessly charge the batteries while the bus
moves [178].
Table 2-5 Battery electric bus charging Strategies (Adopted from [12])
Strategy Overnight Opportunity

Chargers capacity typically, 30-150kw superchargers 150+ Kw
Location depot terminals, on-road
Charging technologies plug-in, pantograph pantograph, induction
Range 100-250km/day 200-500km/day
Battery capacity usually huge small

Bus weight higher bus weight and low weight and reduce the
high energy consumption energy consumption rate
rate
Cost higher battery cost lower battery cost
lower charging higher charging infrastructure
infrastructure cost cost
As shown in Table 2-5, there is no standardisation for the charging strategy.

Nevertheless, the recent literature suggests that opportunity charging can play a decisive
role in improving the BEB's feasibility systems [190], [191]. Besides, in [163], the authors
conclude that "for longer routes, opportunity charging saves 10-20% (of the Total Cost
of Ownership (TCO)), as it enables a significantly smaller battery". The financial
advantages combined with the enhancement of driving range and fast charging make this
charging strategy the most promising. As a result, this thesis' research considers
opportunity charging to ensure that battery-electric buses can be used for similar duty
cycles as conventional vehicles without jeopardising the transit operational schedule.
2.5 Charging technology and charging location for Transit

Electric Bus
The primary barrier to adopting and developing an electrified bus transit system
globally is the lack of supporting infrastructure [192], [193]. Especially in the developing
countries where the accessibility to electricity is considered limited [194] and due to cost,
their circumstances do not typically support the overnight charging of large battery-sized
buses commonly used in Europe and China. Generally, the cost of building onboard large
capacity battery buses for overnight and depot charging is exceptionally huge. Also, there
is a lack of a clear plan to deploy dedicated charging infrastructure (both conductive and
inductive) at the bus terminals, which requires a considerable investment cost [195]. The
trade-off between the cost of operation and charging demands is a key consideration in
providing a reliable and affordable bus transit operation.

In [196], the authors present two concepts for charging the battery-electric bus; standard
and fast charging. The standard charging is described as using the bus depot overnight
with 'modest' charging power. The electric buses in this category are usually equipped
with a high-capacity battery, which increases the bus's weight and increases bus energy
consumption. These buses are scheduled to work during the bus operation hours and then
retire to the bus depot for charging. The fast-charging infrastructures can be inductive
(wireless) or conductive (connected). The buses are usually scheduled to charge during
the daily schedule operation using inductive chargers 'on-the-fly' as they traverse along
bus routes and the conductive chargers at the bus terminal during scheduled downtime
[33]. The fast charging strategy supports a small onboard battery size and significantly
reduces the bus's weight [194].
Using overnight depot charging provides more charging opportunities where the grid
supply is reliable [185]. As mentioned previously, the grid is already challenged in
developing countries, and the addition of scale loading could present further challenges
to supply quality. As part of this thesis's work, on-the-fly inductive charging combined
with multi-terminal conductive charging can provide easy access and effectively manage
the onboard battery state of charge while reducing the burden on the incumbent grid [33].
"When determining which strategy to take, the consideration of energy usage, charging
times and possibilities, operation safety, and battery lifetime should be inevitable [196]".
Detailed below are the possible charging locations and charging technologies for the
electric bus transit system:
2.5.1 Charging locations

An outline of the different types of charging locations is provided as background in this
section. The different features are highlighted and linked to the relevant research literature
in the area where appropriate.
2.5.1.1 Depot Charging

The popular charging location for electric buses is the depot. At the end of the transit
buses' daily operational schedule, the buses are usually packed at the depot, usually at
night. This location and time are considered the best place and duration for charging the

electric buses. These overnight depots' dwell times chargers are usually configured to
have low capacities [197].
A perfect environment for the depot charging of an electric vehicle is usually

comparatively anticipated and follows a daily schedule with overall high mileage. These
basic requirements make conveyance buses ideal candidates. A city bus covers a mean of
220 km per day [198]. Consistent with manufacturers, a battery's capacity can now last
for a distance of 100 km to 300 km between charges [199]. Therefore, even today and
positively soon, a city bus could manage short routes without requiring any charging
infrastructure on its route. This depot charging reduces the investment required for
infrastructure and minimises the operational complexity; if the vehicle must charge too
often, efficient scheduling could also be difficult to plan. However, it is the most time-
consuming charging strategy for electric buses. This is because charging usually happens
when transit buses finish their scheduled routes or stay within the depot during the
shift. The complete charge process for depot charging takes about 3-8 hours, counting
on the charging infrastructure's capacity [181].
Considering the cost, charging at the depot requires the least infrastructure, as no other
equipment is needed except for the depot charger. However, a big onboard battery
capacity is excessively concerned because electric buses must complete their daily trip
without charging. Additional increases in the battery capacity may not be so helpful
because the bus becomes inevitably heavier (with a doubling of vehicle mass, energy
consumption rises by 40 to 60% [200]). The overall energy consumption increases and
reduces the maximum payload of the bus.
2.5.1.2 Terminal and Multi-Terminal Stop Charging

In this case, chargers are usually positioned at the start or end stops in a bus route in
the terminal charging system. The chargers can be located at each terminal for a very long
route with multiple terminals. The terminal and multi-terminal chargers are configured to
take advantage of the stationary time between bus scheduled services [191] [201]. In the
terminal charging system, the electric bus battery capacity should be sufficient to allow
at least journeying between the terminals to circumvent the bus running out of energy due
to external factors such as traffic conditions.

Compared with depot charging, the electric bus has a small onboard battery, and the
charging is usually more frequent. The charger capacity has a higher power (mostly
pantographs). Also, the battery costs are lowered due to the lower battery capacity, while
infrastructure costs are higher due to several fast chargers' needs.
It is essential to ensure that the bus has enough time to recharge the onboard battery
during the terminal charging because the charging duration may impact the operational
bus timetable [202] [203]. Although terminal stop charging involves a considerable dwell
time (typically for customer boarding), operational bus driving can be adjusted to the
scheduled time during the charging procedure to accommodate possible charging delays
[191].
From the grid network aspect, terminal charging avoids peak hour charging. In [181],
the authors present various optimisation algorithms that can manage electric bus charging
power. They make use of a range of different optimisation algorithms to explore this:
Genetic algorithm [204], dynamic programming [205], exponential smoothing model
[206], and locally optimal scheduling [207]. The techniques employed have been
primarily applied to solve scheduling problems linked with the terminal charging of
electric buses. In this work, a PSO algorithm is used to allocate charging infrastructures
to electric bus transit systems in a terminal-based scenario.
2.5.1.3 Along-the-Route Charging

The along-the-route charging system means having charging infrastructure at one or more
points along the bus route to charge the electric bus's onboard battery. When considering
the along-the-route charging system for an electric bus transit system, a trade-off analysis
of the onboard battery capacity, location of charging infrastructure, and the number of
charging infrastructures is important to optimally allocate the network infrastructure
[208] [33], [101]. Compared to the depot and terminal charging methods, the along-the-
route method saves the onboard battery's significant expenditure cost because the battery
size is relatively small. The investment in charging is the highest because numerous high
capacity chargers are usually required along the route to meet the daily operational
charging demand [209]. The main advantage of this charging location is the possibility to
operate full-electric buses without disrupting the daily operating schedule (electric buses

can be used as diesel buses); the onboard batteries will be re-charged at every charging
point along the route [209], [201].
2.5.2 Charging technologies

The availability of charging infrastructures is vital for the reliability and adoption of BEBs
in the public transportation system. Charging public transit BEBs are confronting higher
demand for electrical energy for BEB charging and the time requirements. Considering
the trade-off of these challenges to meet the transit demand, three main types of charging
emerge: Plug-in charging, Pantograph charging, battery swapping, and wireless charging.
Fig.2.10 shows the summary of the key charging technologies, and the detail of these
charging technologies are covered in turn below.
Figure 2-10 BEB Charging Technologies
2.5.2.1 Wireless charging

A wireless power transfer (WPT) system generally consists of a power supply,
transmitter (primary coil), receiver (secondary coil), oscillator circuit, and matching
circuit. The AC supplied by a power source is changed into low-frequency AC using an
oscillator. The oscillator's output is fed into the push-pull circuit that supplies the
transmitter coil. This transmitter coil is also referred to as the primary coil. This primary
coil transmits this power to the receiver coil separated by a certain distance; the energy
received by the secondary coil is then rectified and regulated before output to the BEB

battery charger. The inductive charging system is environmentally suitable and cheap to
maintain because it is not affected by water, ice, dirt, and chemicals [101]. A more
detailed description of the system's power transfer mechanism and hardware
configuration can be found in [210] [211] [212].
Compared to the conductive means of charging electric buses, the wireless charging
system eliminates the impact of environmental conditions (such as; rain, snow, and
extreme temperatures) that can cause discomfort for users when connecting an electric
bus manually at the charging station. The wireless charging system also eliminates issues
with the failure of power cords that can cause safety issues (risk of electrical sparking and
electrical shock) [213]. The other advantages of wireless charging systems include
providing fast and frequent charges while operating the BEB. This option allows for
battery downscaling with the resulting benefits of lighter weight BEBs, reduction in the
cost of the BEBs, energy savings resulting from the decrease in mass, and reduction of
CO2 emissions.
In [214], the authors compare the plugin charging system's energy use with the wireless
charging system. The wireless charging system consumes 0.3% less energy and emits
0.5% fewer greenhouse gases. The wireless charging system is becoming increasingly
common in providing an alternative charging system for public transit [213].
The three main wireless power transfer classifications commonly used for public buses
are capacitive, inductive, and resonant inductive power transfer [215] [216]. The most
common use is the inductive charging system. The inductive charging system is an
integrated on-road charging system that charges the onboard battery remotely while the
bus moves over the installed inductive cable area. The length of the inductive cable, the
size of BEB's battery, and the number of transmitters installed on the road directly affect
the transit system's overall performance and cost. Figure 2-11 illustrates a typical design
of the inductive charging system. The wireless power receiver device is installed at the
lower part of the electric bus. This wireless power receiver device remotely collects
electricity from the transmitter. The wireless receiver delivers power to the BEB motor,
the battery, or both, depending on the motor's power requirement and the battery state of
charge [217]. When the BEB moves on a busway where no power transmitter unit is
installed, the BEB motor uses the battery's power. To eliminate energy waste, the BEBs

that charge via inductive means is considered to have an infra-red sensor that turns 'ON'
the supply that activates the energy exchange between the transmitter and the receiver
circuit. When the BEB moves away from the inductive cable, the infra-red sensor signals
to turn 'OFF' the power supply [211] [101].
Figure 2-11 Transit inductive system (Adopted from [101])
This technology's demerits include low energy transfer efficiency due to losses during
coil-to-coil transfer; high installation costs, which are considerably higher than plugin
charging [216]. This technology is considered the most suitable for dynamic charging
possibility and works best for "along-the-route" charging locations because the onboard
battery receives charge while operating the electric bus in an inductive transit route.
2.5.2.2 Plug-in Charging

'Plug-in' is a conductive charging method commonly adopted for generally charging
electric vehicles and is primarily used to charge small vehicles (e.g. cars). These plug-in
chargers are classified based on the supported output voltage and the rate at which they
will charge a battery. Level 1 charging is the slowest and can be done with 120 volts and
15 amps AC in most wall outlets. Level 2 charging is faster and requires 240 volts and 60
amps of AC for a maximum of 22 kW. Level 3 charging is the fastest and produces more
than 50 kW of power [218]. Level 1 and 2 charging can be carried out at home with the
correct power adapter, and Level 2 and 3 charging are commonly carried out at fixed
public charging stations [219] [220], [221]. Table 2-6 compares the characteristic of Level
1, Level 2, and Level 3 chargers.

Table 2-6 Characteristics of Level 1, Level 2 and Level 3 chargers [219] [220], [221]
Power level types Level 1 Level 2 Level 3

120VAC(US), 240VAC(US), 208-600VAC
Voltage
230VAC(EU) 400VAC(EU) or VDC
Power range <3.7kW 3.7-22kW >50kW
Charging duration 11-36 hours 1-6 hours 0.2-1 hours
Charger topology On-board On-board Off-board
Source supply type 1-phase 1-phase or 3-phase 3-phase
Charging type Slow charge Medium charge Fast charge
Battery capacity 15-50kW 15-50kW 15-50kW
Slow charging rates are not suitable for electric buses because the battery (approx. 400
kWh) is huge, and charging could take very long, limiting trip schedules [222]. Fast
charging tends to be the most convenient choice for both EBs and ETs. However, fast
chargers can have some negative impacts on the BEBs battery that include heating up
during charging, outsized conductors, and high voltage AC/DC converters are required.
As a result of the high-power systems and short charging times, special grid operation
conditions must be considered. [223] and [222] are example of research work that focuses
on the grid impact of fast charging. A typical alternatives include storing energy through
distributed generation in a fast-charging station to help the grid cope with the peak loads.
2.5.2.3 Pantograph charging

Conductive charging is based on energy transfer to electric vehicles by direct electric
contact. The most common solution for conductive charging of electric buses is based on
the usage of pantographs. A pantograph system is a type conductive charging solution
that allows the electric buses to charge quickly at the terminals or stops. The pantograph
system has an automatic connecting system, DC-conductive charging supply equipment,
conductive rails fixed to the vehicle's roof, conductive poles, and communication systems
[224]. The Automatic Connecting System controls and monitors the connection device
for the conductive charger attached to the infrastructure above the vehicle [224]. The

Automatic Connecting System includes an Automatic Connecting Device, which

connects or disconnects the conductive charger to the electric bus-charging interface.
These pantographs are like the ones that have been used for decades on trains and
trams. There are two types of conductive chargers in electric buses, depending on where
the pantograph is located: onboard and off-board [225]. In the onboard conductive
charging solution, the pantograph is situated directly on the bus's roof. This roof-mounted
equipment will contact the overhead supply system located at the charging station. In the
case of off-board conductive charging, the pantograph is installed in the charger's
overhead mast, so only a contact system is required onto the bus, reducing infrastructure
cost and the bus's overall weight. Buses are charged when they arrive at the charging
station, and the pantograph is extended to make electrical contact; this process is usually
automated and takes several seconds. The possibility of installing onboard or off-board
systems provides a high degree of flexibility. Conductive systems can transfer high
power, with capacities ranging from 100 kW to 450 kW; tests are already being made
with 600 kW chargers [226]. Hence, such systems make it possible to fully charge an
electric bus in 4-10 minutes [218] [219].
Figure 2-12 Pantograph System (Source: [227])
Conductive chargers are offered by various companies worldwide, such as Siemens,

ABB, and Opbrid, and have already been tested in many cities, among which Stockholm
can be found [225] [226]. Some of the advantages of conductive charging are acceptable
efficiency levels and lower exposure to electromagnetic fields than inductive charging

systems. The main disadvantages of conductive charging technology are high

maintenance costs due to the contact between the conductor and the collector, visual
impact, exposure to the external environment, and vandalism.
Fig.2.12 illustrates a typical Pantograph system. The automatic pantograph charger

connected to the High Voltage source contacts charging rails (Top Down: Pantograph
mounted to charging station and Bottom-Up: Pantograph mounted to the electric bus).
The wireless position sensors detect when the electric bus is located at the charging station
and connect. High power at the rate of up to 600 kW is transferred to the onboard battery.
The charging is typically about 4 to 6 minutes, making charging during layovers at route
terminals and endpoints possible with the pantograph system.
2.5.2.4 The Battery Swapping Station (BSS)

A Battery Swapping Station (BSS) is where EV users can swap their discharged battery
with a charged battery [228]. Hence, a large battery stockpile is needed to meet the
customers' demands. It is usually assumed that the batteries are owned by the BSS and
rented to customers [229], [230]. The key benefit of BSS for customers is that they can
immediately have a charged battery, like petrol stations. Moreover, the BSS generally
benefits EV batteries since their lives are not affected when they are not charged at fast
charging levels. This is possible when the number of batteries owned by a BSS is many,
and they have time flexibility to charge them. The BSS usually includes a vehicle
platform, lift, alignment, equipment rollers, battery lifts, conveyor shuttles, storage racks,
rails, and electrical connection alignments [231].
The main electrical components of the BSS include a distribution transformer, AC/DC
chargers, battery packs, and a battery energy control module [231]. The transformer
converts the grid's high voltage to a lower voltage suitable for the batteries. Since the
batteries require DC energy, the AC/DC converts the AC energy from the transformer.
The control module allows charging at different power levels depending on grid
requirements. One of the key advantages of BSSs is that a third party might own the
batteries and be responsible for swapping them with fully charged ones, monitoring their
health, and retiring the batteries when they are no longer suitable. Furthermore, the BSS
offers time benefits to users, like typical petrol stations, avoiding long wait times. The

main limitations of BSS include standardisation of EV battery packs, acceptance of the

BSS model, and the reliable estimation of battery state-of-health.
The Better Place business model, which requires customers to purchase a vehicle but
leases the Better Place battery pack [230], is an example of a BSS business plan.
Customers must pay a predetermined cost for up to a specified amount of miles driven
per year and re-charge their cars at Better Place-provided charge stations and a network
of public charging stations. Consumers would drive to the battery's limits for longer
journeys and be routed to the next battery-swapping station, where robotic technology
would install a new one in around 5 minutes. Better Place's idea is to separate the battery
from the vehicle physically. Better Place pioneered automatic battery-switching stations,
which use robots to replace a drained battery with a fully charged battery in roughly five
minutes. Battery-swapping stations are preferable to chargers because they work faster;
they have installed roughly 37 battery-swapping stations in Israel. Better Place can lower
the price of an electric automobile and upgrade the battery as technology advances by
keeping ownership of the battery. Better Place lowers the car's price, similar to how
wireless phone providers discount their hardware and make money by selling minutes. It
charges vehicle owners a monthly price for the battery and electricity, which is calculated
based on the number of miles they travel [232], [233].
2.6 Battery Technology

The battery is the most expensive component of most battery electric vehicles [234].
Besides, they are accountable for the heavyweights of the battery-electric buses. These
huge weights are due to the high energy density required to power electric vehicles. The
electric bus batteries are primarily based on lithium-ion technology. The capacity of the
battery is not constant throughout its lifetime. The capacity decreases progressively with
the battery age. The common factors responsible for decreased capacity include charging-
discharging cycles, temperature, and high Deep of Discharge (DoD) [235].
As fast charging is becoming increasingly common, slowing down the battery ageing
at a high State of Charge (SoC) requires the charging current to be decreased to avoid
exceeding the battery's upper limit voltage [191]. Therefore, an SoC threshold must be

implemented to keep the maximum and minimum charging values in an optimal range to
enhance the battery's lifetime. The maximum and minimum SoC threshold usually ranges
between 20% to 80% of the battery's capacity.
Another essential factor that must be highlighted is the interaction between re-charging
infrastructures and the onboard batteries' requirements. For instance, the overnight
charging strategy reduces the charging frequency. On the other hand, the opportunity
charging requires frequent onboard battery charging. The energy demand is usually
during the daily peak, impacting the grid if the charging is not controlled. The overnight
charging strategy is more suitable for developed countries where grid supply is reliable.
The opportunity charging method ensures that charging occurs when and where grid
supply is present and available or with the usage of solar renewable energy sources and
is seen to be appropriate for developing countries with poor grid reliability or supply
security. The opportunity charging method can also be designed to lower the cost of a
battery-electric bus by accommodating a reduced onboard battery size. However, a high-
capacity charger must be used more frequently and quickly to recharge the buses.
It should be noted that the battery in an EV is its most expensive component,

accounting for around 50% of its entire cost [236]; consequently, the affordability of EVs
is directly related to the affordability of a battery. As a result, as battery costs fall, the
charging infrastructure required for "small onboard batteries" and the opportunity
charging model may alter.
2.7 Battery electric bus energy consumption modelling
There is a considerable body of research work addressing the analysis of energy

consumption of battery electric bus fleets in the city transit networks; some of these works
include [191], [237]–[239]. The transit bus's energy usage is used to forecast transit
service costs. The total energy consumed by an electric bus is usually the aggregate of
three loads: (1) energy consumed by the traction system to drive the vehicle, (2) energy
consumed by the heating, ventilation, and air conditioning HVAC system, and (3) energy
consumed by the rest of the electrical system, such as lighting or control systems. [240].
With accurate estimation under real-world traﬃc conditions, the schedule and the
charging demand of the BEBs can be accurately defined [44]. In [239], the authors

calculate the energy consumption of battery-electric buses by adding the kinetic,

potential, and rotational components of the vehicle's propulsion. Consideration is given
to energy loss due to frictional forces, rolling resistance, and HVAC. The combined
energy output from the vehicle's onboard battery and energy obtained by regenerative
braking equals the energy consumed for constant propulsion and the constant BEB
regeneration efficiency. The model considered in this work did not consider the variable
efficiency of vehicle propulsion and regeneration as a function of torque and rotational
speed. The authors of [238] present a backward modelling method for calculating the
energy consumption of a battery-electric bus. The authors investigated a particular driving
cycle for city buses and the longitudinal vehicle dynamics as a function of velocity in this
work. The cumulative force acting on the bus is determined by adding the aerodynamic,
rolling, and climbing resistance forces and the HVAC loads.
The BEB's energy consumption models are usually defined based on the relationship
between the energy consumption rate and their impact factors [241]. This energy
consumption rate is mostly expressed in kWh/km when considering the modelling of
electric bus transit fleets. In [242], the energy consumption estimated for 10m BEB varies
from 1.6-3.2 kWh/km and 1.7-4.1 kWh/km for 12m BEB. In [243], the authors present a
single-deck and double-deck BEB energy consumption average of 1.6 and 2.5,
respectively. In [242], the authors quantified the double-deck intercity BEB energy
consumption rate range between 2.4-4.5 kWh/km. In [244], the BEB energy consumption
is given as 0.8–1.2 kWh/km, and in [195], it is estimated to be1.5 kWh/km.
Calculating the electric bus energy consumption without reflecting the high
unpredictability of real-world situations is usually inaccurate [195], [243]. Moreover,
overlooking uncertainty in the BEB energy consumption and actual travel time could
result in over-sizing or under-sizing the BEB battery, leading to an infeasible plan for the
transit system [243]. Therefore, it is essential to evaluate the BEB energy consumption
based on the transit route characteristic that incorporates the uncertainty in speed profiles.
Numerous works of literature have investigated the factors that influence the energy
consumption of the battery-electric bus. The factors that typically affect bus fleets' energy
consumption include road topography, battery weight, the weather, and load variation
[239], [244]. Besides, the energy consumption of BEBs can be can also be influenced by

external factors such as route traffic congestion [44]. In [245], the authors present a model
of BEB that uses wireless power transfer technology. The authors estimate the energy
demand between each of the stops using predefined velocity profiles in this work.
Nonetheless, the authors fail to account for unpredictability in speed and uncertainty in
energy demand.
In [246], the authors considered the grey relational analysis method to examine the
impact of external factors on the battery-electric bus's energy consumption. The factors
considered include the day of the week (weekday or weekend), weather conditions
(temperature, fog, or rain), route length, the BEB HVAC, and route traffic situation. These
outlined factors are modelled as numerical values using fuzzy rules. A wavelet neural
network was also adopted to train these external factors and the BEB energy consumption
data to develop a prediction model. The developed prediction model uses actual survey
data that reflects the numerous external factors data to train the wavelet neural network.
Nonetheless, most transit bus corporations do not have sufficient data to develop and
operationalise the neural network model. [247] and [248] used the vehicle dynamics
equation to model BEB with actual driving data for a specific route. The historical data
examined are the bus drive cycles that do not consider uncertainties due to route traffic
congestion.
In [244], the authors present a BEB energy consumption simulation model that
integrates different operating conditions using measured data from the existing bus transit
network. The auxiliary power was considered 6 kW in mild weather conditions, 14 kW
in cold or hot conditions, and 22 kW in extremely cold conditions. In this work, the author
estimates the gross weight of the BEB as 3750kg by assuming the number of onboard
passengers to be 50. The authors in [241] considered the BEB energy consumption in a
real-world traffic congestion scenario. The model considered in this work is a data-driven
model that computes the route energy consumptions based on both the positive kinetic
energy and the regenerative braking of the BEB. In [249], the authors present a multi-
objective stochastic predictive control model for evaluating BEB energy consumption. A
Markov-chain-based stochastic driver model was developed in this work to determine the
BEB demand power in various speed ranges. The authors, however, did not take into
account real-world driving cycles. In [250], the author computes the BEB energy

consumption based on the distance between stops and the BEB average speed. This work
represents the probability of traffic light stops with a binomial distribution function. The
route elevation profile is used to determine the average slopes between stops. In [251],
the authors use the vehicle dynamics model and support vector machine method to predict
BEB driving cycle. In this work, the road slope is assumed to have the upper and lower
limit of 3% and -3%, respectively. Also, the authors in [252] use the vehicle longitudinal
dynamics model to estimate the power consumption of BEB. In this work, the HVAC is
assumed to be constant power of 6 kW. In [195], the authors develop a model to optimise
charging infrastructure for BEBs in urban settings. The authors presumed in this work
that the downward sailing along the route compensates for any energy consumption
increases caused by the uphill direction.
The data published by the BEBs manufacturer are usually considered as the nominal
value because this data does not reflect the real-world factors that can affect the BEB
energy consumption value. Field trial demonstration and simulation software with data
tailored toward the specific case studies have been used to estimate the real-world value
of BEB energy consumption figures for a particular BEB’s transit route. In [238], the
authors compared the simulated energy consumption with the energy consumption
measured by Proterra using the central business district CBD driving cycle. The authors
in [237] validate their model using the Standardised on-road test cycles (SORT)
developed by the International Association of Public Transport.
In [253], the authors developed an algorithm that generates arbitrary driving cycles
based on the original cycle characteristics of transit buses. In comparison to the original
vehicle patterns, a random driving cycle is created that produces an equal speed
distribution and power spectrum. This is promising in terms of providing the basis for
improving the realism of driving patterns, but as the modelling approach requires an
accurate driving cycle to start with might be unavailable without a real-life demonstration
or at least sufficient measured data. SUMO is an open-source traffic simulation platform
developed by the Institute of Transportation Systems at the German Aerospace Center
(Simulation of Urban Mobility). To simulate an effective driving cycle, the SUMO needs
only a few input parameters (such as the bus route and vehicle type). [254]. The SUMO
is a microscopic model that allows the simulation of each vehicle as a separate entity.

Transit data, topography and GPS data and vehicle characteristics can be supplied as the
model input data. The simulation output is the velocity of the targeted vehicles at any
time. The authors [247] generate a driving cycle for plugin hybrid cars using a 2D
simplistic SUMO version. This version of SUMO does not account for the height
variations of the road. In [255], the authors introduce a 3D SUMO that incorporates
topographic data as part of the input parameters. The 3D feature enhances the accuracy
of the generated driving cycle. In [238], the authors examine the possibility of combining
testing cycles for modelling purposes to meet up with some specific road conditions. A
typical example is the Manhattan Bus Cycle (MBC) combination, and the City Suburban
Cycle (CSC) is used to represent a suburban driving pattern.
Models of vehicle energy consumption can be grouped into forwarding and backward
models. Backward models measure the tractive contribution required at the wheels and
"function backwards" towards the engine. Alternatively, "forward models" begin with the
engine and function with transmitted and reflected torque [256]. The Advanced Vehicle
Simulator (ADVISOR) will simulate in both directions simultaneously [257]. The energy
and signal flow in the forward-facing models are identical to the actual flows in the
powertrain systems of real vehicles. The calculation starts with the energy source and
progresses to the vehicle's propulsion system [256]. The models that face backwards
operate in the opposite direction of the real tractive energy flow. The downstream
component's energy requirements determine the energy consumed by the upstream
component. The driving cycle establishes the initial energy requirement for the part
immediately adjacent to the input. The simulation then continues component by
component until the energy source is reached [258], [259]. The ADVISOR was first
established at the National Renewable Energy Laboratory in November 1994. It was
developed as an analysis tool to assist the US Department of Energy (DOE) in developing
hybrid electric vehicle (HEV) technologies through the Hybrid Electric Vehicle
Propulsion System contracts with Ford, General Motors, and DaimlerChrysler [257]. Its
primary function is to demonstrate the system-level interactions between hybrid and
electric vehicle components and their impact on vehicle efficiency and fuel economy. The
authors of [234] used the ADVISOR to assess solar reflective car shells. The authors in
[100] calculated the BEB's kWh/km energy consumption along particular routes using

the ADVISOR model. The authors used the BEB speed profile, route topography, and
auxiliary load rating in this work to improve the accuracy of the BEB energy consumption
value. Nonetheless, the author neglects to account for the effect of weather-related
uncertainty on the BEB energy consumption value.
Milton Keynes's demonstration project in the United Kingdom successfully converted

from a diesel fleet to an electric bus transit system in 2014 [260]. There are eight buses
in the fleet. The transit route is 24 kilometres long, and buses run 17 hours a day. Inductive
chargers were installed to improve the range of buses, with some at the depot and others
along the route for opportunity charging. When evaluating the Milton Keynes project, a
comparison was made between the projected average energy usage of the buses and the
five-monthly recorded energy consumption values. Extreme weather events result in a
change in the auxiliary load, which increases BEB energy consumption.
Additionally, the experiment considered the effect of the driver's performance on the
bus's average energy consumption. The driving demonstration demonstrates that the
average energy consumption of the bus ranges between one and two kWh/mile.
Additionally, it was discovered that the path topography accounted for 1-2% of the transit
bus's energy consumption. Additionally, this work explores the reliability of the power
charger by comparing the power drawn from the grid to the power supplied to the battery.
The average performance of a charger is 78%. This study of the Milton Keynes project
demonstrates that the buses' actual output is consistent with the simulated results.
The literature reviewed in this section illustrates that factors affecting bus fleets'
energy consumption typically include road topography, battery weight, temperature, and
load variance [239], [244]. Additionally, external factors such as route traffic congestion,
which is a characteristic of a transit bus speed profile, may affect the energy consumption
of BEBs [44]. As a result, it is critical to assess BEB energy consumption using transit
route characteristics that account for speed profile uncertainty. As a result, this thesis
develops a longitudinal dynamic model of BEBs transit and performs backward
simulations using the ADVISOR software.

2.8 Stochastic EV Charging Impact and Integration of the BEB

Charging demand into the Electrical Distribution System
The widespread adoption of EVs will impact incumbent power systems if left
unmanaged. There would be an increase in the grid's load, resulting in an increased
likelihood of overvoltage and overcurrent events, increased harmonic and reactive
current, voltage imbalances, and interruptions [261]. There is already a considerable body
of research work addressing these issues, and a common theme is the need to apply
demand-side management to control the timing of these additional power system loads.
Examples include various factors for evaluating PHEVs' impact on the distribution
network, such as driving patterns, charging characteristics, charge timing, and vehicle
penetration [262]. In [263], the authors explore stochastic modelling and simulation
techniques to evaluate the effect of electric vehicle charging demands on the distribution
network to alleviate power security problems that may emerge as a result of widespread
EV adoption. Stochastic modelling was used to assess over-current and under-voltage on
a three-phase delivery load flow analysis by examining driving patterns, charging
characteristics, charging timing, and vehicle penetration. The Roulette wheel selection
principle and Monte Carlo simulations are used in this work by introducing different
uncertainties into the simulation. The author's numerical results indicate that smart
charging may help mitigate power security concerns that may arise due to EV adoption.
The authors, however, neglect to account for regular and hourly vehicle delivery, which
would increase their model's accuracy. The authors in [264] and [265] considered the
impact of uncoordinated EV charging on the grid and how coordinated charging of EVs
can reduce power losses. In [265], the authors specifically look into deterministic and
stochastic analyses of the potential load impacts on actual distribution circuits to
investigate if controlled charging of the PEVs can potentially reduce the loading impacts.
In this work, the stochastic results show that the temporal and spatial diversity of PEVs
charging on the system mitigates mass overloads of any particular asset class for
penetration levels in the range of 2-8%.
In [266], the stochastic approach based on a Monte Carlo simulation is applied to

investigate the impact of EV charging on the load distribution system. The system
considered is where both the standard charging at home and quick charge service at a

public charging station are considered. In this work, the authors assumed the number of
EVs to be charged in a local facility is based on the sum of local EVs owned by
households in the studied area and EVs arriving from external areas. In this paper, the
simulation approach is based on Monte Carlo indicates that inadequate load capacity
occurs during peak hours of the day and early night-time. However, the authors fail to
account for the possibility of a strategic valley filling to manage the peak hours' electric
load. In [267], the authors compare the possible effect of charging electric vehicles at
home and fast-charging stations on transformer loading and system bus voltage. The
system considered is where the impact of the EV fast-charging stations on the distribution
grid is evaluated. In this work, the stochastic charging model is applied. The result
indicated that PEV charging at home has minimum impact on the distribution grid even
at 100% penetration level and fast charging stations affect transformer loading and system
bus voltage. This paper suggested that local energy storage and voltage conditioning
devices, such as SVC (Static Var Compensator), should be used at a fast-charging station
to improve system reliability.
In [268], the authors coordinate the charging of multiple plug-in hybrid electric
vehicles in residential distribution grids to minimise the power losses and maximise the
main grid load factor. The coordinated charging considered is one where stochastic
programming is applied for 24 hours daily profile. The PHEV is taken to have 30%
penetration in the case study considered, while the power losses and maximum voltage
deviation for the charging period were measured. In this work, the comparative evaluation
of a stochastic modelling approach and a deterministic approach reveals that a stochastic
modelling approach minimises power losses and voltage deviation. In [269], the authors
examine the optimisation strategy of controlled electric vehicle charging by considering
the demand-side response and regional wind and photovoltaic to optimise the grid's peak-
valley difference. The strategic EV-controlled charging considered is where the
probability model of wind and PV power output is developed. In this work, a formal
optimisation based on genetic algorithms is used to good effect to efficiently determine
the start and end time of the valley price and the peak-valley price.
In [270], the authors investigate the future electric vehicle charging demand to propose
an algorithm that can shift vehicle charging to absorb surplus wind generation to mitigate
the increase in peak load, low voltage, and substation overloading. The model considered

is when the domestic car pattern is studied by analysing the United Kingdom time of use
data. A Monte Carlo simulation approach based on car departure and arrival is used to
determine the car location in this work. The authors provide the basis to estimate the
future impact of electric vehicle charging on the distribution transformer, especially in a
scenario where the householders start charging as soon as they arrive home. However,
the authors fail to account for the impact of commercial vehicles on the distribution grid.
In [271], the authors propose optimal charging of electric vehicles to mitigate the
distribution constraint (due to the growing uptake of EVs) by moving the charging to off-
peak periods. The optimal charging considered is when electric vehicle charging is
regarded as a linear optimisation problem that considers both the present and the
anticipated constraints in the distribution network over a ﬁnite charging horizon. In this
paper, a linear optimisation based on charging demand and vehicle arrival and departure
is used with two objectives: maximising all vehicles' overall charging and minimising
charging cost. The authors provide essential cost and time optimisation. The authors
estimated that uncontrolled electric vehicle charging could lead to network failures at
only 10 to 15 % uptake rates. The system should accommodate more than 80 % of the
proposed EVs load with optimal load control.
Scientific research, international agencies, public attention, and political interest in

problems associated with transportation, energy, and the environment motivate the bus
transit system's electrification [101]. The power grid for public transportation is distinct
from consumers', as the charging infrastructure for electric public transportation is more
centralised (i.e., terminals, depots, and bus stops) [33]. As a result, charging technologies
are expected to create an entirely new transportation–energy paradigm that will be
incomprehensible to transit and utility operators and other stakeholders [102].
Additionally, transit bus operators have rigid bus schedules and operating logistics that
must be maintained throughout the service of BEB transit systems [100]. According to
[272], BEBs may stick to their scheduled operational schedules if an acceptable charging
strategy is implemented.
On the other hand, the distribution network operators would have right to be concerned
about the potential adverse effects of electrified bus transit networks on power grids. The
authors of [174] note that charging scheduled BEBs can result in power losses and
compromise the power system's stability. According to [82], incentive charging of BEBs

results in voltage control problems in power distribution networks. According to [238],

the BEB waiting time for charging reduces the frequency of transit services. Recent
research on the smooth adoption of electrified bus transit systems includes [273], which
provides a scenario-based BEB transit service scheduling method for routing BEB fleets.
However, it does not consider the transit system's adherence to the BEB's operating
schedule. The authors of [203] analyse a charging schedule for transit BEBs to minimise
the BEB's charging cost. Nonetheless, a fixed charging and queue duration was set. The
authors of [274] develop a BEB transit design algorithm to minimise charging
infrastructure construction costs. The authors presume that the number of chargers is
equal to the number of BEBs in this work. The authors of [81] describe how to design
electrified bus transit systems using a mixed-integer linear algorithm. The size of the
BEBs' onboard battery capacity has been predetermined in this work.
As part of the work of this thesis, a priority charging strategy is designed to mitigate
the impact of the BEB charging on the distribution power system. This priority charging
mitigates the BEB charging impact by aggerating the BEB charging demand for the
scheduled operational period.
2.9 Electric Vehicles Charging Demonstration with Microgrids

and Off-Grid Systems
With off-grid systems (e.g. mini-grids) playing an important role in energy provision in
developing countries, it is natural to consider them as part of the literature review of this
thesis. In many cases, there will be insufficient capacity available to accommodate EVs,
as this was never a consideration at the design stage. However, in some cases, there are
complementary benefits from considering renewable dense mini-grids with the charging
issues of EV that can read across to the BEB focus of his work.
In [238], microgrids (a collection of DERs, including RES and ESS, and loads that operate
locally as a single controlled entity) and their application with transportation
electrifications are discussed. It is proposed in this work that increasing EV adoption
would increase the microgrid's ability to promote renewable energy sources and the
vehicle's efficiency in providing grid-to-vehicle services. The authors argued that
electrifying the transportation system would minimise reliance on fossil fuels by

increasing access to a broader range of renewable energy sources. The coordination of

EV involvement in load frequency regulation in isolated microgrids MGs is discussed in
[275] to minimise frequency deviances and support the microgrids' device stability. The
authors added EVs to the MG to minimise frequency deviations and maintain device
stability by efficient charging/discharging scheduling of EVs. When the microgrid's
frequency is threatened, EVs will absorb (inject) excess (deficit) energy and act as energy
storage systems. The success of the proposed plan is evaluated under a variety of
operational scenarios. Additionally, this work suggested that implementing this proposed
strategy would increase the overall MG operator's benefit while substantially reducing
emissions, but this is not quantified.
In [276], an ‘ideal’ configuration of EV charging stations is considered by examining

different energy sources to minimise device lifecycle costs while minimising
environmental emissions. The EV charging stations that are being considered are those
that are powered by renewable energy and diesel. The isolated microgrid is planned to
meet the charging needs of electric vehicles and increase travel distance. Additionally,
this work looks at grid-connected electric vehicle charging stations, which are considered
the most viable alternative. The possibilities for rural areas to use electric Tuk-tuk battery
charging stations are discussed in [95]. A tuk-tuk is a three-wheeled vehicle primarily
used in Asia and Africa for public transportation. Wind and photovoltaic renewable
energy sources and standalone battery storage are used to power the planned EV charging
station. This study compared some EV charging scenarios using the HOMER framework.
These comparisons are based on equipment design, energy production, and financial
feasibility over a 20-year project. The system components are rated using the worst-case
scenario. The simulation results indicate that the optimal operating scenario is to charge
multiple vehicles consecutively when operating the station at maximum capacity for an
entire day or 24 hours continuously.
The authors of [277] investigate a direct current DC microgrid for future electric
vehicle charging stations using an energetic macroscopic representation and a maximum
control structure. The system is a photovoltaic-based urban DC microgrid that charges
plug-in electric vehicles and supplies a DC load. Additionally, the system considered the
connection to the public grid. The control system is designed to obtain the maximum
amount of energy possible from the photovoltaic system and regulate the power flow for

the state of charge of electric vehicles and the DC load power demand. The energetic
macroscopic representation (EMR) and maximum control structure are used to model the
urban DC microgrid. According to the simulation results, a DC microgrid device can
power five PEV chargers and a few household appliances. Additionally, the results
indicated that the amount of power drawn from the public grid could be monitored and
decreased by using the photovoltaic energy within the microgrid. [96] explains the design
of a low-cost electric vehicle charging station using four 255-watt solar panels, batteries,
a charge controller, and an inverter. The experiment lasted three months and involved a
SMART Electric Drive vehicle successfully charged solely with solar energy. The
proposed "Sun-Car System" offers a low-cost option for lower-income communities, such
as those found on Indian reservations in the southwestern United States and Africa. It
illustrates the fundamentals of using photovoltaic energy to charge electric vehicles.
However, the demonstrated capacity is small, as the car needed at least eight hours of
charging to gain 12 miles
[278] investigates the economics of microgrid planning algorithms in conjunction with

electric vehicle charging demands to determine the most economically feasible design
that maximises renewable energy use. This study uses HOMER to perform a formal
optimisation based on the optimal microgrid configurations for charging systems and
various EV penetration ratios. The authors of [279] discuss photovoltaic (PV) integration
into low voltage (LV) grid feeders in order to meet environmental targets. The energy
storage systems (ESS) within a public EV charging station regulate voltage, ensure power
quality during EV charging switching events, and store and dispatch generated energy
during the PV peak time (night or evening). [280] examines the design of a device that
automatically switches between solar, battery, and grid electricity, with the solar
generator taking precedence. The system considered is one in which an EV is connected
to the charger, and the PV system cannot produce enough power to charge the EV; the
battery or the utility grid will support it. While no electric vehicle is plugged in, PV energy
is stored in the battery. Excess PV energy is transmitted to the electricity grid when the
battery is completely charged. Grid power is used during off-peak hours to restore the
battery to its optimum level if it is depleted. [281] describes an 'optimal' energy
management scheme for EV integration into a microgrid that utilises an autonomous
controller to schedule the charging and discharging of the EV battery for optimal energy

management. The optimal energy management scheme considered is when the electric
vehicles (EVs) are connected to a charging station with a microgrid system. The peak
charging and discharging schedule is determined in this work by predicting the EV's trip
pattern.
As indicated in the introduction chapter, the focus of this thesis is on developing

countries where the security of grid supply is already a concern. As a result, proper
consideration for an energy system that can supply the energy demand of the
electrified bus transit system is critical. This proposed transit-based energy system is
expected to incorporate a variety of energy sources, most notably renewable energy
sources such as solar and wind energy, and an energy storage system (ESS). Additionally,
a grid source and a diesel generator can be added to supplement renewable energy sources
(The diesel generator is considered because of the regular 'black-outs' experienced by
domestic and industrial customers in developing countries [282]). As a result, the transit
energy system only buys from the grid when renewable energy sources are in limited
supply (i.e. when primary sources such as solar and ESS cannot meet the system's energy
requirement) and sells to the grid when there is a surplus of generated energy. The diesel
generator is only considered for usage in an emergency, such as when both the renewable
energy source and the grid cannot supply transit demand. The use of this diesel generator
is limited to reducing the system's CO2 emissions.
2.10 Chapter summary

This chapter provides the background for the BRT focus of this thesis. An overview of
the most recent studies on the modelling and simulation of battery-electric bus energy
consumption is included. The published literature examines various infrastructure options
and the design, implementation, and optimisation of their operational schedules and this
is linked to the work of this thesis. Most notably, a summary of the literature on various
strategies and technologies that can be used to promote the electrification of public transit
systems in developed countries to reduce GHG emissions and congestion was also
presented.
There are numerous drawbacks to adopting the bus manufacturer's nominal bus energy
demand for transit system modelling. The drawbacks include one or more of the

following:: 1) necessitate high-resolution speed profiles, which are difficult to obtain due
to a lack of operational data and related costs; 2) ignore the difficulty and variation of
detailed real-world operating conditions, such as route traffic conditions; and 3) depend
on standard driving cycles, which ignore randomness in speed and energy demand
volatility. The work of this thesis will incorporate five traffic flow conditions into the
model to address the above concerns, thus increasing the model's accuracy. The data is
focused on the volume of traffic and the road capability, with traffic conditions graded as
free, move, appropriate, inappropriate, and critical. The trip time, traffic conditions, driver
behaviour, and topography of the bus route will all be used to capture the effects of points
1 and 2 discretely.
Existing attempts to design charging infrastructures for BEB fleets in a transit system
have not fully accounted for: (i) energy consumption at each BEB fleet charging terminal
by analysing BEB energy demand and classifying the demand according to charging
priority; (ii) a multi-terminal model based on inductive charging infrastructure that
reduces the size of the electric vehicle's on-board battery. (iii) a hybrid solution that
incorporates both conductive and inductive charging systems within a transit network in
order to minimise operational downtime caused by BEB charging duration. In
developing countries, transit operators and power utilities face both technical and
operational challenges when integrating ready-made BEB systems. As a result, research
and development efforts must be made to study these problems and develop mitigation
strategies.
Thirdly, the efficient adoption of BEBs requires a multi-terminal deployment of transit

bus infrastructure. As a result, these BEBs charging stations must be optimally sized to
minimise capital expenditure and maximise operating profit. This profitability and
expenditure study would encourage private and public investments in the electrification
of public transit systems. As a result, both the government and private investors, including
the BEB fleet operators, must have analytical resources available to analyse BEB
charging stations and forecast financial returns based on the energy market and public
transport demand. Nonetheless, the literature has included several general models for
sizing charging infrastructures. However, prior to these studies, the presentation of a
mathematical model converted into a computational model was unclear. Most notably,
this study is focused on the electrification of bus rapid transit systems compared to the

previous work presented in [105], which explored an empirical model for estimating the
size of electric and hydrogen-based fuelling stations. Additionally, this study estimates
the transit energy demand profile that can be extended to evaluate renewable energy
systems' suitability by determining the renewable energy system's potential to electrify
public transit in developing countries with an already challenged grid. This resource
modelling approach is expected to differ from previous research that focused exclusively
on grid integration.

Chapter 3: BEB transit fleet design and mathematical modelling
3 BEB TRANSIT FLEET DESIGN

AND MATHEMATICAL
MODELLING
The preceding chapter presents a background and coverage of the relevant literature on
electrification of BRT systems and the gaps in the literature, which involve considering
various infrastructure options and the design, use, and optimisation of BEBs' operational
schedules. The chapter summarises recent research on the modelling and simulation of
the energy usage of battery-electric buses. The literature studies focus on various
infrastructure options and their design, implementation, and optimisation of operational
schedules. Notably, a literature review on different methods and innovations for
promoting the electrification of public transit systems in developing countries was also
addressed to minimise GHG emissions and congestion.
This chapter presents the details of the BRT system electrification modelling and the
mathematical model that manages the routing and charging for BEBs that will be used in
this thesis. In addition, a mathematical optimisation process for determining the critical
design variables of allocation of the conductive and inductive chargers is also presented.
BRTs are designed to serve the public in densely populated metropolitan areas where
traffic patterns and volumes permit service along defined routes at scheduled intervals
[283]. Therefore, it is critical to guarantee that the proposed model replicates and
integrates the practical elements of the standard BRT operation system while designing
charging infrastructure for the electrification of BRT systems. The operational timetable
for the transit network is part of the BEB's transit fleet concept. This would result in a
transit network electrification model that preserves the operational requirements of the

transit network. Furthermore, transit fleet data and network architecture (i.e., BEB
weight, efficiency, speed profiles, route distance, and topography) influence BEB energy
consumption. As a result, the output of the BEB transit fleet model and the energy
consumption model is fed into the model that allocates charging infrastructure for the
BEB transit network. The complete BRT transit model (Figure 3-1) aims to find the
optimal configuration for an electric BRT fleet, including BEB battery capacity, location,
and charger number. Furthermore, the model calculates the BEB transit energy demand
profile.
Figure 3-1 provides a comprehensive visual overview of the various modelling stages
for the BRT modelling that will be covered in this chapter. It starts on the left-hand side
with the design of the transit fleet routing model, then the BEB energy consumption
model, the modelling of charging infrastructure for the BRT transit fleets and the
estimation of transit energy demand.
Figure 3-1: The stepwise model of the BRT electrification process
The following sections will describe each of the different feature elements shown in
Figure 3.1. Firstly, the transit fleet routing model design that links the operation timetable
will be detailed. Secondly, the mathematical model for the BEB energy consumption will
be developed based on a longitudinal method that accounts for the dynamics. The
methodology for modelling charging infrastructure for the BRT transit fleets is then
developed along with the mathematical and optimisation formulations for the novel BEB
charging system model, multi-terminal charging model, the integrated charging model
and the BEB transit energy demand estimation. Finally, a novel mathematical model is
developed to allocate multi-terminal inductive charging infrastructure to optimise power

transmitter placement, inductive cable length, battery capacity, and the cost of an
electrified transit system.
3.1 Transit fleet mathematical model

This section describes the mathematical modelling of the BEBs transit feet. Prior work
that has influenced the approach of this thesis concerning the development of a
mathematical model for the bus transit system includes the work reported in [284]. These
authors present a broad and complete review of state-of-the-art models and approaches
for solving public transit problems at strategic, tactical and operational levels. Also, the
authors in [100] present an integrated model for optimal battery-electric bus systems
configuration. The transit network operational schedule is added to the BEBs transit fleet
model to preserve the transit network operational requirement. In [285], the authors
propose a classification of approaches dealing with the design, frequency setting, and
timetabling of transit lines and their combinations. As a result, the transit network
schedule is influenced by various elements, the most important of which are transit route
networks, passenger demand, transfer coordination, and fleet size. The authors of [286]
put together many contributions to vehicle routing and crew scheduling, schedule
optimisation, and service management for various modes of public transportation. This
study demonstrates that a successful transition to electric transit systems necessitates
adhering to the operation schedule.
These authors’ work emphasised the importance of public timetables in meeting public
transportation demands. This demand varies according to the hours of the day, the days
of the week, the season, and even the year. In addittion, it represents the community’s
transportation needs for enterprise, industry, culture, education, social, and recreational
activities. This section employs a modelling method similar to that described in [100] for
determining BRT charging requirements. In contrast, this section presents a mathematical
representation of the proposed bus transit timetable’s proposed model in an opportunity
charging scenario.
In this work, the number of electric buses is represented in the form of sets as:
ℬ = {1, 2, 3,. . . , 𝑏,. . . , 𝑁𝑏 } (3.1)

Where ℬ is the set of buses in the transit network, 𝑏 is the b-th bus in the network BEBs
and 𝑁𝑏 , is the maximum number of BEBs.
Similarly, the number of bus routes is represented in the form of sets as:
ℛ = {1, 2, 3,. . . , 𝑟,. . . , 𝑁𝑟 } (3.2)
Where ℛ is the set of routes in the transit network, 𝑟 is the indices for the r-th route and
𝑁𝑟 , is the maximum number of routes in the transit network.
The buses are allocated to bus routes based on the bus transit operational requirements,
which is mathematically represented in sets of assignments D. Each set of assignments
throughout the bus operating hours is given as:
𝐷 = {1, 2, 3,. . . , 𝑑,. . . , 𝑁𝑑 } ∀ 𝑑 = [𝑏, 𝑟, 𝑆𝑑 ] (3.3)
Where in each assignment d, bus b is scheduled for route ‘r’ for a prearranged scheduling
period, defined as follows:
𝑆𝑑 = [𝑡𝑑𝑠 , 𝑡𝑑𝑒𝑛𝑑 , 𝑡𝑑𝑡𝑟𝑖𝑝 , 𝑡𝑑𝑤𝑎𝑖𝑡 , 𝑁𝑑𝑡𝑟𝑖𝑝 ] ∀ 𝑑 ∈ 𝐷 (3.4)
Where 𝑆𝑑 is the set of the operational schedule for the assignment d. The Sd is like a
typical transit timetable.cla
Figure 3-2 shows a diagram for transit bus b allocation to the route presented
in (3.3) and (3.4). As depicted in the figure, the scheduled time interval for each trip
assignment ‘d’ is defined by a start time 𝑡𝑑𝑠 , and an end time 𝑡𝑑𝑒𝑛𝑑 for a specific allocated
route ‘r’ in minuties.
In sequence, each transit bus operates the given number of trips 𝑁𝑑𝑡𝑟𝑖𝑝 in a pre-
quantified time, which is followed by a given waiting/dwelling time 𝑡𝑑𝑤𝑎𝑖𝑡 in minuties
i.e., when the bus is dwelling at the bus terminal station after each trip. This waiting is
considered as being available for charging time for BEBs. Notably, each bus b may have
more than one assignment with dissimilar routes and/or schedule during the daily
operating hours.

Figure 3-2 Transit schedule for assignments d
Additionally, it is also important to note that the allocated buses and their operation
schedules for a specific route ‘r’ usually vary throughout the daily operation (i.e., peak
versus off-peak operation); the set of assigned trips for each bus in each assignment d can
be defined as:
𝐽𝑑 = {1, 2, 3 . . . , 𝑗, . . . , 𝑁𝑗,𝑑 } ∀ 𝑑 ∈ 𝐷 (3.5)
Where 𝐽𝑑 is the set of trips within assignment d and 𝑗 indices for the BEB trips.
The number of trips in each allocation is computed as:
(𝑡𝑑𝑒𝑛𝑑 − 𝑡𝑑𝑠 ) + 𝑡𝑑𝑤𝑎𝑖𝑡 (3.6)

𝑁𝑑𝑡𝑟𝑖𝑝 = 𝑐𝑦𝑐𝑙𝑒 ∀𝑑 ∈𝐷
𝑡𝑑
The cycle time for each assignment d is given as follows:
𝑐𝑦𝑐𝑙𝑒
𝑡𝑑 = 𝑡𝑑𝑡𝑟𝑖𝑝 + 𝑡𝑑𝑤𝑎𝑖𝑡 ∀ 𝑑 ∈ 𝐷 (3.7)

This mathematical approach will be used in this thesis to capture the transit data and
estimate the transit demand for BEBs that are operating in the transit network. It has been
structured in such a way to reflect operational aspects that planners use to define when
they are developing the timetable and to enable the analysis of transit data [100]. The next
section presents the BEB energy consumption model.
3.2 BEB Energy Consumption Model
The BEB transit fleet's Energy Consumption Model is the second stage of electrifying the
fleet, as identified in Figure 3-1. Accurate energy consumption calculation is critical for
BEB scheduling and service [287]. BEBs energy consumption is necessary to determine
the optimal battery and charger configuration to meet transit schedule requirements.
Although manufacturers specify a nominal energy consumption rate for their BEBs, these
values do not typically consider the transit network's unique characteristics, such as route
topography and speed. Calculating the actual energy consumption of each direction based
on the transit system's characteristics is worthwhile in this context [100].
Energy use is critical in assessing the operational strategy and cost analysis of BEBs
and directly impacts the cost of fuel and greenhouse gas emissions from electric bus fleets
[288]. BEB's energy use estimates are essential for planning, scheduling, and allocating
transit charging stations [44]. Typically, the route topography, battery weight,
temperature, and load variance all impact the energy consumption of bus fleets [239],
[244]. Additionally, external factors such as route traffic congestion, which is a
characteristic of a transit bus's speed profile, can affect the energy consumption of BEBs
[44]. As a result, to achieve a more accurate figure for BEB consumption, it is necessary
to estimate BEB energy consumption using transit route characteristics that account for
the uncertainty associated with speed profiles. BEB energy consumption is an input
parameter for charging facility allocation and transit energy demand modelling.
Additionally, by providing a model based on route traffic classification that accurately
reflects real-world scenarios. The transit operator can plan transit buses accurately,

adequately scale the BEB onboard battery, and allocate charging infrastructures while
preserving the BEB schedule timetable.
It was found in Chapter 2, as part of the literature review, that many of the works
identified as being relevant to the work of this thesis (for example, [100], [243]) had not
taken sufficient account of either the complexity and variations created by real-world
operating conditions. Additionally, most of the previous authors' work failed to account
for the effect of normal driving cycles, speed fluctuations, and variability in energy
demand. This thesis's BEB energy consumption model considers the impact of route
traffic conditions, HVAC system energy consumption, and bus route topographies. These
features can assist bus transit operators in deriving greater value from this modelling work
because of various traffic scenarios. This assessment is also expected to bolster the
accuracy of the onboard battery sizing and charging infrastructure design.
Hence, this section presents a longitudinal dynamic BEB energy consumption model.
The longitudinal dynamic method is generally based on the forces acting on the bus and
the traction forces that overcome those forces to maintain the BEB velocity during the
trip [250]. The traction force on the BEB in motion can be calculated using Newton's
second law, which describes translational and rotational systems. This mathematical
method is used to calculate the total amount of energy taken from the onboard battery of
an electric bus to power the bus. The amount of energy used by the bus varies according
to driving conditions. The overview of the powertrain system configuration of BEB is
shown in Figure 3-3.
Figure 3-3 BEB Powertrain System Configuration (Adapted from [289])

As shown in Figure 3-3, the BEB's primary source of energy is the battery. Traction
motors can be used in either drive or regenerative-braking mode. In order to maximise
performance, the bus's braking power can be captured in the regenerative-brake mode.
The traction motor provides propulsion for the vehicle and recovers braking energy to
charge the battery. While the traction motor is in drive mode, the electrical power
produced by the motor is positive; when the motor is in regenerative-brake mode, the
electrical power generated by the motor is negative. The efficiency map of a traction
motor [290] is converted to a look-up table in the simulation tool, and the efficiency is
calculated based on the speed and torque.
The converter acts as an ac/dc power interface between the traction motor and the
battery. Converting ac to dc and dc to dc is modelled as an ideal voltage transformer with
constant efficiency values. The accessory is represented mathematically as a constant
power load. The transmission mechanism connects the electric motor to the vehicle
dynamics. The bus transmission can be modelled in single or multiple gear ratio
configurations. When a transmission with several gear ratios is modelled, an external
controller controls gear shifting based on the vehicle's speed. The transmission efficiency
is determined using an efficiency map and the current speed and torque values. Constant
efficiency is employed in this work via the dc-dc converter—the battery powers the
auxiliary devices [289], [291].
3.2.1 The longitudinal dynamic model for the BEB energy

consumption
As mentioned previously, the energy consumption of a bus is usually determined by the

forces acting on it and the traction force that overcomes those forces to maintain the BEB
velocity during the trip [250]. The traction force on the BEB in motion can be calculated
using Newton's second law, which is used to describe translational and rotational systems.

Figure 3-4 Free body diagram of a BEB in motion (Adapted from [291])
Figure 3-4 shows a free body diagram of a BEB in motion. The top and bottom rigid
bodies represented are equivalent, according to Newton's second law. The main forces
𝑔 𝑟
acting on the body are the grade force 𝐹𝑡,𝑗,𝑑 , the rolling resistance force, 𝐹𝑡,𝑗,𝑑 and the
𝑎 𝑝
aerodynamic force 𝐹𝑡,𝑗,𝑑 , and 𝐹𝑡,𝑗,𝑑 represents the force supplied by the motor to propel
the vehicle forward by overcoming the external resistive forces (traction force).
According to Newton's second law [250], [292], the traction force is expressed as.
𝑝 𝑎 𝑔 𝑟
𝑀𝑡,𝑗,𝑑 𝑎𝑡,𝑗,𝑑 = 𝐹𝑡,𝑗,𝑑 − (𝐹𝑡,𝑗,𝑑 + 𝐹𝑡,𝑗,𝑑 + 𝐹𝑡,𝑗,𝑑 ) ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏 (3.8)
Where 𝑀𝑡,𝑗,𝑑 , is the mass of BEB at time t, trip j, and assignment d, 𝑎𝑡,𝑗,𝑑 in m/𝑠 2 is the
𝑎 𝑔 𝑟 𝑝
BEB acceleration at time t, trip j, and assignment d, and 𝐹𝑡,𝑗,𝑑 , 𝐹𝑡,𝑗,𝑑 , 𝐹𝑡,𝑗,𝑑 , 𝐹𝑡,𝑗,𝑑 , is the
aerodynamic, grade, rolling resistance and traction forces at time t, trip j, and assignment
d.
Therefore, the traction force at time t, trip j, and assignment d is given as:
𝑝 1 2 (3.9)
𝐹𝑡,𝑗,𝑑 = 𝜌𝐴𝐶𝑡𝐷 𝑣𝑡,𝑗,𝑑 + 𝑀𝑡,𝑗,𝑑 𝑔𝑐𝑜𝑠 ∅𝑡,𝑗,𝑑 + 𝐶𝑟 𝑣𝑡,𝑗,𝑑 𝑀𝑡,𝑗,𝑑 𝑔𝑠𝑖𝑛∅𝑡,𝑗,𝑑
2
+ 𝑀𝑡,𝑗,𝑑 𝑎𝑡,𝑗,𝑑 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝑝
Where 𝐹𝑡,𝑗,𝑑 in (N) is the BEB traction force at time t, trip j, and assignment d, 𝜌 is the
Air density (kg/𝑚3 ), 𝐴 is the BEB cross-section area (𝑚2 ), and 𝐶𝑡𝐷 , 𝐶𝑟 , is the drag, and
rolling resistance coefficient, respectively. The 𝑣𝑡,𝑗,𝑑 in (m/s) is the BEB speed at time t,
trip j, and assignment d, 𝑀𝑡,𝑗,𝑑 in kg, is the mass of BEB at any instance at time t, trip j,
and assignment d g is the gravitational force (m/𝑠 2 ), and ∅𝒕,𝒋,𝒅 in (deg.), is the road slip
angle at time t, trip j, and assignment d.

During acceleration, the traction force is positive. Therefore, power is transferred from
the BEB battery to the wheel. The wheel and motor torque are expressed as:
𝑤 𝑝
𝑇𝑡,𝑗,𝑑 = 𝐹𝑡,𝑗,𝑑 . 𝑅𝑤ℎ𝑒𝑒𝑙 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏 (3.10)
𝑤
where 𝑇𝑡,𝑗,𝑑 is the wheel torque and 𝑅𝑤ℎ𝑒𝑒𝑙 , is the wheel radius.
and
𝑤
𝑚
𝑇𝑡,𝑗,𝑑 (3.11)
𝑇𝑡,𝑗,𝑑 = ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝜂𝑇
𝑚
𝑇𝑡,𝑗,𝑑 represent the motor torque and 𝜂𝑇 , is the transmission efficiency.
Respectively, their respective rotational speed is described as
𝑤
𝑣𝑡,𝑗,𝑑 (3.12)
𝑤𝑡,𝑗,𝑑 = ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝑅𝑤ℎ𝑒𝑒𝑙
𝑤
where 𝑤𝑡,𝑗,𝑑 , is the wheel rotational speed.
𝑚 𝑤
𝑤𝑡,𝑗,𝑑 = 𝑤𝑡,𝑗,𝑑 × GR ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏 (3.13)
𝑚
where 𝑤𝑡,𝑗,𝑑 , is the motor rotational speed, and GR is the gear ratio.
After obtaining the motor efficiency using an efficiency map, the BEB’s instantaneous
power consumption can be computed using the following formula:
𝑚 𝑚
𝑐𝑜𝑛𝑠
𝑇𝑡,𝑗,𝑑 . 𝑤𝑡,𝑗,𝑑 𝑃𝑎𝑢𝑥 (3.14)
𝑃𝑡,𝑗,𝑑 = + ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝜂𝑚 . 𝜂𝑐𝑜𝑛 𝜂𝑐𝑜𝑛
𝑃𝑎𝑢𝑥 is the BEB auxiliary power demand and 𝜂𝑐𝑜𝑛 , 𝜂𝑚 , is the conversion and motor
efficiency, respectively. The energy supply for auxiliary loads such as heating, cooling,
lighting, and sound systems is referred to as auxiliary power.
This mathematical model (3.8) to (3.14) is used to estimate the total energy drawn from
the electric bus onboard battery to power the bus. The energy consumed by the bus can
be varied depending on the driving conditions.
The energy consumption rate is given by

𝑡=𝜏 (3.15)
𝑐𝑜𝑛𝑠 𝑐𝑜𝑛𝑠
𝐸𝑡,𝑗,𝑑 = ∫ 𝑃𝑡,𝑗,𝑑 (𝑡)𝑑𝑡 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝑡=0
The energy use per route in kWh/km is given as:

𝑡=𝜏 (3.16)
𝑡𝑟𝑖𝑝 1 𝑐𝑜𝑛𝑠
𝐸𝑙𝑟 = ∫ 𝑃𝑡,𝑗,𝑑 (𝑡)𝑑𝑡 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝑙𝑟
𝑡=0
The SOC of the battery is expressed as a percentage of the total remaining battery energy
capacity (100% state-of-charge means that the battery charge level is full, while 0% SOC
implies that the battery charge level is empty), and this is represented as follows
𝑐𝑜𝑛
𝐸𝑡,𝑗,𝑑 (3.17)
𝑆𝑂𝐶𝑡,𝑗,𝑑 = 𝑆𝑂𝐶𝑡−1,𝑗,𝑑 − (100 . . 𝑙𝑟 ) ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝐸𝑏𝑐𝑎𝑝
𝐸𝑏𝑐𝑎𝑝 is the BEB battery energy capacity and 𝑙𝑟 is the length of the route
The traction force is negative during deceleration, and the kinetic energy is transferred
to recharge the battery. Thus (3.18) and (3.19) introduce the evaluation of this
regenerative braking.
𝑏 𝑝
𝑀𝑡,𝑗,𝑑 𝑎𝑡,𝑗,𝑑 = 𝐹𝑡,𝑗,𝑑 − 𝐹𝑡,𝑗,𝑑 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏 (3.18)
𝑏
Where 𝐹𝑡,𝑗,𝑑 , is the BEB brake force at time t, trip j, during assignment d. The unit of this
force is newton (N).
Hence, the regenerative force at any time instant in trip j is given as:
𝑟𝑒𝑔 𝑏
𝐹𝑡,𝑗,𝑑 = 𝑆. 𝐹𝑡,𝑗,𝑑 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏 (3.19)
S is the power slip ratio between friction brake and regenerative brakes. The manufacturer
usually estimates the split friction and regenerative to be 60 to 40 [292].
This mathematical model shows that energy consumption is mostly determined by the
forces acting on the bus and the required traction force to overcome those forces and

maintain the BEB velocity during the journey. BEB's energy consumption figure is
required to find the optimal battery and charger configuration to meet the transit timetable
requirement. As a result, the generic roadmap for calculating BEB energy consumption
is shown in this mathematical model. The advanced vehicle simulator (ADVISOR) is a
commonly used tool that allows the integration of critical factors (route topography,
weather conditions, passenger occupancy rate, and traffic conditions) that can improve
the accuracy of a BEB battery-to-wheel energy consumption figure. As a result, the
simulation example in the next chapter applied the ADVISOR modelling tool to a specific
case study. The BEB energy consumption figure is estimated to be kWh/km and used as
an input parameter in the novel charger allocation optimisation model proposed in this
thesis, as well as to estimate the potential reduction in GHG emissions along the route
under consideration.
3.3 MODELS OF CHARGING INFRASTRUCTURES

The previous section presents the procedure for evaluating the BEB transit fleet Energy
Consumption figure. The BEBs energy consumption figure is one of the input parameters
required for the optimum configuration of batteries and chargers to meet transit schedule
requirements.
The mathematical and optimisation formulations for the novel BEB charging system
model, multi-terminal charging model, integrated charging model, and BEB transit
energy demand estimation are presented in this section. Moreover, a mathematical model
is developed to allocate multi-terminal inductive charging infrastructure in order to
optimise power transmitter location, inductive cable length, battery capacity, and an
electrified transit system's cost.
3.3.1 NOVEL DESIGN OF CHARGING INFRASTRUCTURE FOR

BUS RAPID TRANSIT SYSTEM
This section presents the methodology for designing charging infrastructure for the
BRT transit network [33]. Also, this section gives the overview and the procedures for
multi-terminal charging for the BRT transit network. The system presented in this work

integrates both inductive and conductive priority charging to ensure effective energy
management without disrupting the operational bus schedule.
The charging at the BEB terminals is assigned based on the priority that depends on
the state of charge of the BEB battery on arrival. Priority charging is a charging strategy
that is designed to address the charging challenges in the electric bus transit system. The
priority charging proposed in this thesis first classified the charging need of the BEB on
arrival at the bus terminal. This charging classification is based on the BEB’s computed
energy demand for the next scheduled trip and the state of charge on arrival at the bus
terminals where the proposed conductive charging infrastructures are located. Then a
decision is made on whether the charging requirements of the BEB’s are classified as low,
medium or high level charging, where this charging requirement classification is based
on the calculated energy consumption expected for the next scheduled trip, the available
capacity of the charger at the bus terminals and availability of on-the-road integrated
inductive chargers on the route.
The term low-priority charger relates to a set of BEB’s onboard battery chargers with
the lowest power capacity. The medium-priority charger relates to a set of onboard battery
chargers with mid-range power capacity, and high priority chargers relate to onboard
chargers with the highest power capacity. It is worth noting that the priority-charging
strategy is designed to ensure that BEB’s on-board battery quickly recovers to maintain
the operational timetable while aggregating the charging demand over the daily transit
operation scheduled and reducing the burden of the transit BEBs charging demand on the
power distribution system.
Moreover, the priority-charging model presented in this thesis also has an on-the-road
integrated inductive charger. These inductive chargers are intended to be integrated into
the transit system to improve the system's reliability by eliminating the impact of charging
downtime. In this thesis, the allocation of the inductive charger [101] to a route is based
on the length of the route, the onboard battery size of the BEB and the trade-off between
the available conductive chargers and the route charging demand. This inductive charging
stage occurs when the bus moves slowly or is stopped for a few minutes to pick up and
drop off passengers. The BEB that can be charged inductively will contain additional
equipment installed that acts as a 'pickup' device to collect electrical energy from the

installed on-road power transmitters. This 'scavenged' charging will enable the topping
up of the BEBs batteries to extend their range and protect the BEBs batteries from deep
discharge. This proposed model is illustrated in Figure 3-6.
Figure 3-5 Proposed allocation of charging infrastructure for the bus transit system
The schematic diagram that illustrates multiterminal-based charging that integrated on-
road inductive charger is shown in Figure 3-5. In this figure, items number 1-6 are
considered the bus terminals along the BRT route. This proposed bus terminal is the
location of the conductive chargers, and the green-coloured lines between some of the
terminals are the proposed locations of the integrated on-road induction charger along the
transit route.
The process flow diagram in Figure 3-6 illustrates the terminal-based model described
above. The process starts with the daily schedule, where it is assumed that the BEBs start
with a fully charged onboard battery. On arrival from the trip, the allocation of charging
infrastructure is based on the process flow diagram in Figure 3-6 described as follows:
1. Given the set of BEBs (ℬ) assigned to a set of routes (ℛ) and a set of assignments
(𝐷) with a set of trips ( 𝐽𝑑 ). Hence, the number of buses 𝑏 ∈ ℬ; the assignment 𝑑 ∈
𝐷; the trip 𝑗 ∈ 𝐽𝑑 ; and the time 𝑡 ∈ 𝜏.
2. Each BEB operates on a fixed route within the BRT network, a subset of ℛ; this
fixed route has a terminal at both ends where the BEB can use conductive chargers.
Additionally, inductive chargers are allocated to the routes based on the length of
the route, the onboard battery size of the BEB and the trade-off between the
available conductive chargers and the route charging demand. In addition, there
are multiple bus stops to load and unload passengers along each route.
3. The b-th BEB is equipped with one battery unit denoted with (𝐸𝑏𝑐𝑎𝑝 ) its capacity
in kWh. Each of the BEB batteries is taken to be of the same size. In this work, the

BEBs batteries have a minimum (𝐸𝑏𝑚𝑖𝑛 ) and maximum (𝐸𝑏𝑚𝑎𝑥 ), range of the total
battery energy capacity.
𝑎𝑟𝑟,𝑖 𝑎𝑟𝑟,𝑖
4. The arrival time (𝑡𝑏,𝑗,𝑑 ), the battery energy capacity on arrival (𝐸𝑏,𝑗,𝑑 ) and the
battery capacity (𝐸𝑏𝑐𝑎𝑝 ) are logged on the arrival of the b-th BEB on the j-th trip of
assignment d at the i-th terminal (𝑖). The duration between the arrival and the next
schedule is called the stay or dwell time which is the maximum time allowed for
BEB charging
𝑤𝑎𝑖𝑡,𝑖 𝑠,𝑖 𝑎𝑟𝑟,𝑖
𝑡𝑏,𝑗,𝑑 = 𝑡𝑏,𝑗,𝑑 − 𝑡𝑏,𝑗,𝑑 ∀𝑏 ∈ 𝐵 ∧ ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 (3.20)
𝑐ℎ,𝑖 𝑤𝑎𝑖𝑡,𝑖
𝑡𝑏,𝑗,𝑑 ≤ 𝑡𝑏,𝑗,𝑑 ∀𝑏 ∈ 𝐵 ∧ ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
(3.21)
5. Then, the energy consumption of the BEB for the next scheduled trip 𝑗 + 1 in an
assignment 𝑑 of the b-th BEB moving from terminal i to i+n is given as:
𝑡=𝑖+𝑛
𝑖,𝑖+𝑛 𝑐𝑜𝑛𝑠
𝐸𝑏,𝑗,𝑑 =∫ 𝑃𝑡,𝑗,𝑑 (𝑡)𝑑𝑡 ∀𝑏 ∈ 𝐵 ∧ ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡
𝑡=𝑖 (3.22)
∈𝜏
Here 'n' is an integer indicating the subsequent number of terminals until the end of the
scheduled trip.
𝑎𝑟𝑟,𝑖
6. If the 𝐸𝑏,𝑗,𝑑 satisfies the next scheduled trip of the BEB energy demand, the
proposed algorithm will also check capacity constraints 𝐸𝑏𝑚𝑖𝑛 , associated with the
b-th battery are met. If the constraint 𝐸𝑏𝑚𝑖𝑛 is also satisfied, the charging of the
BEB depends on the availability of the charging unit before the next scheduled trip
𝑗 + 1, and the BEB charging in this circumstance is given low priority.
7. However, if the constraint 𝐸𝑏𝑚𝑖𝑛 is not satisfied, the model verifies if the next
scheduled route has a midway inductive charger that can top up the BEB battery
to satisfy the 𝐸𝑏𝑚𝑖𝑛 , constraint. With an inductive charger's availability along the
route, the BEB is scheduled to charge at a low priority (depending on the charging
unit's availability before the next scheduled trip). Otherwise, the BEB is allocated
to be charged at a medium-priority.

𝑎𝑟𝑟
8. If the 𝐸𝑏,𝑗,𝑑 , cannot meet the energy consumption demands of the next scheduled
trip 𝑗 + 1 from i to i+n, the BEB is allocated for charging with high priority
charging.
Thus, on the arrival of BEB at the terminal, the charging priority is selected to be high,
medium, and low is expressed in the form of energy dynamics as follows:
𝑎𝑟𝑟.𝑖 𝑖,𝑖+𝑛
𝐻𝑖𝑔ℎ 𝐸𝑏,𝑗,𝑑 ≤ 𝐸𝑏,𝑗,𝑑
𝑠𝑒𝑙𝑒𝑐𝑡
𝐶ℎ𝑏,𝑑,𝑗 = {𝑀𝑒𝑑𝑖𝑢𝑚 𝑎𝑟𝑟,𝑖
𝐸𝑏,𝑗,𝑑 𝑖,𝑖+𝑛
− 𝐸𝑏,𝑗,𝑑 ≤ 𝐸𝑏𝑚𝑖𝑛 } ∀𝑏
(3.23)
𝐿𝑜𝑤 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
∈ 𝐵 ∧ ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
𝑠𝑒𝑙𝑒𝑐𝑡 𝑖,𝑖+𝑛
Where 𝐶ℎ𝑏,𝑑,𝑗 , is the charging priority selector. 𝐸𝑏,𝑗,𝑑 , is the energy required for the
BEB's next scheduled trip from a terminal, i to i+n. The nomenclature of high is high-
priority charging, the medium is medium-priority charging, and the low is low-priority
charging.
𝑎𝑟𝑟,𝑖
The battery energy capacity on arrival at the bus terminal is (𝐸𝑏,𝑗,𝑑 ) , and the minimum
battery capacity (𝐸𝑏𝑚𝑖𝑛 ) is the tolerable minimum energy level that should be in the BEB
onboard. In this work, the minimum energy level of the battery is expressed as 20% of the
battery's state of charge. This is necessary to avoid over-draining of BEB battery to prolong
the battery lifetime.
The equation (3.23) gives a simplified overview of how the priority chargers is selected
at each bus terminal and ensures that each BEB uses appropriate charging priority during
their recovery time at the bus terminal.
In summary, this section describes the priority-charging methodology that classifies

the BEB's charging requirements upon arrival at the bus terminal. This charging
classification is determined by the BEB's calculated energy requirement for the next
scheduled trip, as well as the level of charge at the bus terminals. Furthermore, the
priority-charging model described in this thesis includes an on-the-road integrated
inductive charger. These inductive chargers are designed to be incorporated into the
transit system to enhance system dependability by removing the impact of charging
downtime.

Any Start
BEB exist daily NO scheduled YES
End trip for the
schedule
BEB
BEB start scheduled
trip at the terminal i
Inductive charger
BEB depart at 𝑡𝑏𝑠 Route BEB arrive at i+n
Compute the
𝐸𝑏𝑎𝑟𝑟 ,𝑡𝑏𝑤𝑎𝑖𝑡 ,
𝑖,𝑖+𝑛 𝑐𝑎𝑝
𝑡𝑏𝑎𝑟𝑟 , 𝐸𝑏 & 𝐸𝑏
Assign BEB to the Charge BEB & update
𝐵𝐸𝐵 𝑐ℎ 𝑑𝑒𝑝
scheduled route 𝑁𝑐ℎ , 𝑡𝑏 , 𝐸𝑏
NO Is next
High priority
trip
NO
charging satisfy
YES
Charger will the

YES Low priority
available 𝐸𝑏𝑡 after NO
before charging next trip
next trip <𝐸𝑏𝑚𝑖𝑛
YES
Scheduled for
mid-way Inductive
NO Route has
charging Medium priority
inductive
charging charger
YES
Figure 3-6 Flow chart for the terminal-based charging model
The next subsections (i.e. 3.3.1.1-3.3.1.3) give the details of the objectives equations
and associated constraints based on this priority-charging model. First, an objective
equation for allocating priority-charging infrastructure in the BEB transit model is given
in sub-section (3.3.1.1). This objective equation is extended to represent the transit
network as a multi-terminal allocation of priority-charging infrastructures in (3.3.1.2),
and subsection (3.3.1.3) presents an integrated model that takes into account the

allocation of both the multi-terminal priority conductive charging system and the on-road
inductive system.
3.3.2 Optimisation problem for the BEB transit model's priority-

charging infrastructure allocation.
This thesis is aimed to address the infrastructure need of the electrified BRT transit
systems. Hence, considering a set of BEB ℬ with each BEB b operates with an onboard
battery capacity 𝐸𝑏𝑐𝑎𝑝 (kWh) that is valued at 𝐶𝑏𝑏𝑎𝑡 per kWh required processing a set of
trip 𝐽𝑑 . Each BEB b can process, at most, one trip j at a time. In the scheduled trip the
𝐸𝑏𝑐𝑎𝑝 is expected to be decreasing at the average rate of BEB energy consumption 𝐸𝑡,𝑗,𝑑
𝑐𝑜𝑛𝑠
,
estimated for the scheduled trip j. In order to prolong the battery lifespan, the BEB on-
board battery is constrained by the maximum capacity (Ebmax ) and minimum capacity
(Ebmin ) limits [293].
In addition, there are three categories of chargers located at the bus terminal for the
purposed of charging the BEBs during the scheduled operational dwell time 𝑡𝑑𝑤𝑎𝑖𝑡 . These
𝜄
chargers are represented as low-priority chargers with power capacity 𝑃(𝑡) and associated
𝑙 𝑚 𝑚
cost 𝐶𝑐ℎ , medium-priority chargers with power capacity 𝑃(𝑡) and associated cost 𝐶𝑐ℎ , and
ℎ ℎ
high-priority chargers with power capacity 𝑃(𝑡) and associated cost 𝐶𝑐ℎ respectively. The
𝑎𝑟𝑟
charger utilisation is selected based on the state of charge on arrival 𝐸𝑏,𝑗,𝑑 at the BEB
terminal (as described in chapter 3, section (3.3.1)) and the demand for the next scheduled
𝑐𝑜𝑛𝑠
trip j, in which the product 𝐸𝑡,𝑗,𝑑 with the length of the scheduled trip 𝑙𝑟 . Moreover,
installing these charging infrastructures is associated with a varied costs with the
classified types of charging infrastructure.
Therefore, this problem formulation allows the infrastructure requirements for

allocating the minimum priority-charging infrastructure in the BEB transit model to be
determined-this is captured formally as:

𝑀𝑖𝑛 ∑ ∑ 𝑁𝑏 . 𝐸𝑏𝑐𝑎𝑝 . 𝐶𝑏𝑏𝑎𝑡 + 𝑁𝑐ℎ

𝚤 𝜄
. 𝑃(𝑡) 𝑙
. ∝𝚤 . 𝐶𝑐ℎ 𝓂
+ 𝑁𝑐ℎ 𝑚
. 𝑃(𝑡) 𝑚
. ∝𝑚 . 𝐶𝑐ℎ (3.24)
𝑏∈𝐵 𝑡∈𝑇
ℎ ℎ ℎ 𝑙 𝑙 𝑚 𝑚 ℎ ℎ
+ 𝑁𝑐ℎ . 𝑃(𝑡) . ∝ℎ . 𝐶𝑐ℎ + 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 + 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 + 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡
+ 𝑁𝑏 . 𝐶𝑏
 and the various terms in this equation are as follows: The first term is the cost of
the BEBs battery packs (𝑁𝑏 . 𝐸𝑏𝑐𝑎𝑝 . 𝐶𝑏𝑏𝑎𝑡 ), expressed as the product of the number
of BEBs, their battery capacity, and the battery cost.
 𝚤
The second term is the cost of the low-priority charging units, (𝑁𝑐ℎ 𝜄
. 𝑃(𝑡) 𝑙
. ∝𝚤 . 𝐶𝑐ℎ ),
as the product of the number of low-priority chargers, low-priority charger rated
power, and the low-priority charger expenditure cost.
 The third term is the cost of the medium-priority charging units,
𝑚 𝑚 𝑚
(𝑁𝑐ℎ . 𝑃(𝑡) . ∝𝑚 . 𝐶𝑐ℎ ), as the product of the number of medium-priority chargers,
medium-priority charger rated power, and the medium-priority charger
expenditure cost.
 ℎ
The fourth term is the cost of the high-priority charging units, (𝑁𝑐ℎ ℎ
. 𝑃(𝑡) ℎ
. ∝ℎ . 𝐶𝑐ℎ )
as the product of the number of high-priority chargers, high-priority charger rated
power, and the high-priority charger expenditure cost.
 𝑙
The remaining terms describe the charger installation cost (𝑁𝑐ℎ 𝑙
. 𝐶𝑖𝑛𝑠𝑡 𝑚
, 𝑁𝑐ℎ 𝑚
. 𝐶𝑖𝑛𝑠𝑡 ,
ℎ ℎ
𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 ) for the low, medium, and high-priority charging, respectively. The
charger installation cost is the product of the number of chargers and the
installation cost per charger. The installation cost is considered to vary for each
charging priority.
It is worth noting that the transit fleet optimisation problem aims to find the minimum
cost for the transit infrastructure while optimising the charging infrastructures and the
BEB onboard battery capacity. This optimisation problem is subjected to associated
constraints that are given in equations 3.25 to 3.41.
For simplicity, it is assumed in this thesis that BEBs in the fleet are identical, i.e., BEBs
have the same battery size to offer transit network operators flexibility to allocate the
𝑑𝑒𝑝
BEBs to different routes. Hence, for each BEB b at trip j, the departure (𝐸𝑏,𝑗,𝑑 ) and arrival
𝑎𝑟𝑟
(𝐸𝑏,𝑗,𝑑 ) battery energy capacity (from/to the bus terminal) of the BEBs on-board battery

is constrained by the maximum capacity (𝐸𝑏𝑚𝑎𝑥 ) and minimum capacity (𝐸𝑏𝑚𝑖𝑛 ). The
BEB manufacturer usually enforces limits to prolong the battery lifespan. These
constraints are given as follows:
𝑎𝑟𝑟 𝑑𝑒𝑝 𝑐𝑜𝑛𝑠 (3.25)

𝐸𝑏,𝑗,𝑑 = 𝐸𝑏,𝑗,𝑑 − 𝐸𝑡,𝑗,𝑑 . 𝑙𝑟 ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝑎𝑟𝑟
𝐸𝑏,𝑗,𝑑 ≥ 𝐸𝑏𝑚𝑖𝑛 ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 (3.26)
𝑑𝑒𝑝
𝐸𝑏,𝑗,𝑑 ≤ 𝐸𝑏𝑚𝑎𝑥 ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 (3.27)
Here (3.25) gives the trip arrival energy state of each BEB on arrival at the bus terminal
as the difference between the state of energy of the battery on departure and the trip
consumption rate with distance. Therefore, the BEB battery size is constrained by the
capacity limits of the onboard bus battery as follows:
𝐸𝑏𝑚𝑖𝑛 ≤ 𝐸𝑏𝑐𝑎𝑝 ≤ 𝐸𝑏𝑚𝑎𝑥 ∀𝑏 ∈ 𝐵 (3.28)
At the BEB arrival at the bus terminal, the bus is expected to charge before the next trip
j+1. The charging equation is expressed as follows:
𝑑𝑒𝑝 𝑎𝑟𝑟 𝑐ℎ (3.29)
𝐸𝑏,𝑗+1,𝑑 = 𝐸𝑏,𝑗,𝑑 + 𝐸𝑏,𝑗,𝑑 ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
The energy gain during the charging with low-priority chargers is given as:
𝑐ℎ,𝑙
𝑡=𝑡𝑏,𝑗,𝑑 (3.30)
𝑐ℎ,𝑙 𝑙
𝐸𝑏,𝑗,𝑑 = 𝜂𝑐ℎ × ∫ 𝑃(𝑡) 𝑑𝑡 ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝑡=0
The energy gain during the charging with medium-priority chargers is given as:
𝑐ℎ,𝑚
𝑡=𝑡𝑏,𝑗,𝑑 (3.31)
𝑐ℎ,𝑚 𝑚
𝑡=0
The energy gain during the charging with high-priority chargers is given as:
𝑐ℎ,ℎ
𝑡=𝑡𝑏,𝑗,𝑑 (3.32)
𝑐ℎ,ℎ ℎ
𝑡=0
The charging power is constrained by the minimum and maximum capacity of the
specified charger. The constrained equation for the minimum and maximum capacity of
the low-priority, medium-priority and high-priority charger is given by:

0 ≤ 𝑃𝜄 ≤ 𝑃𝑚𝑎𝑥
𝜄 (3.33)
𝜄
𝑃𝑚𝑎𝑥 ≤ 𝑃𝑚 ≤ 𝑃𝑚𝑎𝑥
𝑚 (3.34)
𝜄
𝑃𝑚𝑎𝑥 ≤ 𝑃ℎ ≤ 𝑃𝑚𝑎𝑥
ℎ (3.35)
Respectively, the summation of the energy used by all BEBs that charge at a particular
terminal can be described as follows:
𝑛
(3.36)
𝑐ℎ,𝑙 𝑐ℎ,𝑚 𝑐ℎ,ℎ 𝑡𝑜𝑡𝑎𝑙 𝑐ℎ,𝑡𝑜𝑡𝑎𝑙
∑(𝐸𝑏,𝑗,𝑑 + 𝐸𝑏,𝑗,𝑑 + 𝐸𝑏,𝑗,𝑑 ≤ 𝑃(𝑡) . 𝑡𝑏,𝑗,𝑑 ) ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑
𝑗
∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
𝑙
∑(𝑃(𝑡) 𝑚
+ 𝑃(𝑡) ℎ
+ 𝑃(𝑡) 𝑡𝑜𝑡𝑎𝑙
= 𝑃(𝑡) ∀𝑡 ∈ 𝜏 (3.37)
𝑡∈𝑇
𝑛
(3.38)
𝑐ℎ,𝑙 𝑐ℎ,𝑚 𝑐ℎ,ℎ 𝑐ℎ,𝑡𝑜𝑡𝑎𝑙
∑(𝑡𝑏,𝑗,𝑑 + 𝑡𝑏,𝑗,𝑑 + 𝑡𝑏,𝑗,𝑑 = 𝑡𝑏,𝑗,𝑑 ) ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
𝑗
When the BEB arrives at any designated BEB terminals, and it receives a charge, the
constraints are given as:
𝑎𝑟𝑟 𝑐ℎ,𝑙
𝐸𝑏,𝑗,𝑑 + 𝐸𝑏,𝑗,𝑑 ≤ 𝐸𝑏𝑚𝑎𝑥 ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 (3.39)
𝑎𝑟𝑟 𝑐ℎ,𝑚
𝑎𝑟𝑟 𝑐ℎ,ℎ
The ∝𝚤 , ∝𝑚 and ∝ℎ , are the probability variables for low-priority, medium-priority,

and high-priority charging, respectively. In equations 3.42 to 3.49, the details of these
probability variables are given.
The value of ∝𝚤 , ∝𝑚 and ∝ℎ are is based on the possible numbers of the trips in the
transit network schedule. With the short distance attributed to low-priority or low power
charger within the dwelling time at the bus terminal, the medium-priority for median
distances and the high-priority for the far distances.

Figure 3-7 Multi-terminal route based charging model
Using Figure 3-7 to show how the values of ∝𝚤 , ∝𝑚 and ∝ℎ are calculated: Assuming that
some BEBs will travel from T1 to T2, T2 to T3, T3 to T4, T4 to T5, or vice versa. Because
the trips occur between two terminals and span a short distance, the energy consumed is
proportionate to the distance travelled, resulting in low kWh consumption that may be
recovered with a low-priority charger during the dwell period.
Similarly, if BEBs are scheduled to go from T1 to T3, T2 to T4, T3 to T5, or vice versa,
the BEB will require more energy to recover appropriately; thus, the BEBs are charged
using the medium-priority charger.
Furthermore, if BEBs are scheduled for a long trip, such as from T1 to T4, without
being charged the entire journey. The BEBs will be charged using a high-priority charger
in order for them to recover quickly within the allotted dwell period.
Hence, a mathematical model is developed to calculate the proposed value of ∝𝚤 , ∝𝑚

and ∝ℎ using Figure 3-7 as an example. The (3.42) combines all the transit routes in the
network. The (3.43) is the proportion of BEBs that used low priority chargers, (3.44) is
the proportion of BEBs that used medium-priority chargers and (3.45) is the proportion
of BEBs that used high priority chargers.
𝑇11 𝑇12 ⋯ 𝑇1𝑛 (3.42)

𝑇21 𝑇22 𝑇2𝑛
𝑇=| |
⋮ ⋮ ⋯ ⋮
𝑇𝑚1 𝑇𝑚1 𝑇𝑚𝑛
𝑛(𝑇 𝑙 ) (3.43)
∝𝚤 =
𝑛(𝑀) − 𝑛(𝐷)
Where 𝑛(𝑇 𝑙 ) is the number of possible routes with the distance that require a low-priority
charger, 𝑛(𝑀) is the total number of routes combined in a transit schedule system that is

represented with matrix (𝑇) shown in (3.42), and 𝑛(𝐷) are diagonal elements of the
matrix in (3.42).
Similarly,
𝑛(𝑇 𝑚 ) (3.44)
∝𝑚 =
𝑛(𝑀) − 𝑛(𝐷)
ℎ
𝑛(𝑇 ℎ ) (3.45)
∝ =
𝑛(𝑀) − 𝑛(𝐷)
Where,
𝑛(𝑇 𝑙 ) = 𝑛(𝑇𝑚𝑛+1 , 𝑇𝑚+1𝑛 , 𝑇𝑚+1𝑛+2 … … . 𝑇𝑚+5𝑛+4 … ) (3.46)
𝑛(𝑇 𝑚 ) = 𝑛(𝑇𝑚𝑛+2 , 𝑇𝑚+1𝑛+3 , 𝑇𝑚+2𝑛 … … . 𝑇𝑚+5𝑛+3 … ) (3.47)
𝑛(𝑇 ℎ ) = 𝑛(𝑇𝑚𝑛+3 , 𝑇𝑚𝑛+4 , 𝑇𝑚𝑛+5 … … . 𝑇𝑚+5𝑛+2 … ) (3.48)
It is also a constraint that a BEB can only use one type of charger during the bus terminal's
waiting time.
Therefore,
∝𝚤 + ∝𝑚 + ∝ℎ = 1 ∝∈ [0, 1] (3.49)
This section described the optimisation model of the multi-terminals priority charging
strategy and the associated constraints. The implementation of this optimisation algorithm
and the result of a simulation example is presented in the next chapter. The following
section considered an integrated model that added an on-road inductive charging system
to the multi-terminal priority-charging present in this section.
The introduced problem is based on optimising the infrastructure needs for the
electrification of the BRT system. The developed mathematical formulation helps
quantify the charging infrastructure needed to meet the electrified BEB demand as it
services the various routes in the timetable. The solution to this problem will provide the
necessary infrastructure needed to optimise the investment cost (least assets) for the BEB
transit system.
This mathematical formulation only considers a priority-charging conductive single

terminal case; the next subsection extends these optimisation parameters for the multi-
terminal transit system.

3.3.3 Optimisation problem for the allocation of the priority-charging

infrastructure in a multi-terminal transit model
The proposed BEB transit system multiterminal charging configuration is the design
of chargers at the major bus stop (called the bus terminal). The terminal is where the buses
stop for 10-30 minutes at the end or start of a scheduled trip to off-board and onboard
passengers. This waiting time is considered as the charging duration to avoid disruption
of the BEB operational schedule. A transit network usually has many terminals
strategically positioned to distribute buses to meet customer demand effectively. Hence,
the multiterminal priority charging strategy proposed in this thesis is aimed to allocate
priority charging infrastructures to meet the transit BEBs charging demand by positioning
adequate resources at each BEB terminal. An example that illustrates this is shown in
Figure 3-5, which shows six terminals transit network where charging infrastructures
would be allocated.
The objective equation for the multi-terminal charging model is expressed,

𝑛
(3.50)
𝑀𝑖𝑛 ∑ ∑ ∑ 𝑁𝑏 . 𝐸𝑏𝑐𝑎𝑝 . 𝐶𝑏𝑏𝑎𝑡 + 𝑁𝑐ℎ
𝚤 𝜄
. 𝑃(𝑡) 𝑙
. ∝𝚤 . 𝐶𝑐ℎ 𝓂
+ 𝑁𝑐ℎ 𝑚
. 𝑃(𝑡) 𝑚
. ∝𝑚 . 𝐶𝑐ℎ
𝑖=1 𝑏∈𝐵 𝑡∈𝑇
ℎ ℎ ℎ 𝑙 𝑙 𝑚 𝑚
+ 𝑁𝑐ℎ . 𝑃(𝑡) . ∝ℎ . 𝐶𝑐ℎ + 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 + 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡
ℎ ℎ
+ 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 + 𝑁𝑏 . 𝐶𝑏
It is noted that (3.50) is similar in structure to (3.24) that provides the infrastructure
requirements for allocating the minimum priority-charging infrastructure in the BEB
transit model; the only difference is the fact that (3.50) is applied across the whole transit
network system (i.e., all the terminals in the transit network).
Here
𝑙 𝜄
𝐶𝑐ℎ = 𝑐𝑓𝑙 + 𝑐. 𝑝(𝑡) (3.51)
𝑚 𝑚
𝐶𝑐ℎ = 𝑐𝑓𝑚 + 𝑐. 𝑝(𝑡) (3.52)
ℎ ℎ
𝐶𝑐ℎ = 𝑐𝑓ℎ + 𝑐. 𝑝(𝑡) (3.53)

𝑙 𝑚 ℎ
The cost components 𝐶𝑐ℎ , 𝐶𝑐ℎ and 𝐶𝑐ℎ are considered to have two components; the fixed
cost (𝑐𝑓𝑙 , 𝑐𝑓𝑚 , and 𝑐𝑓ℎ )charged for using either low priority, medium priority, and high
priority, as described in (3.51) to (3.53), respectively. These fixed costs are considered
varied with the charger's priority. The lowest cost is allocated as a fixed cost for the
lowpriority charger and the highest for the high-priority charger. This fixed cost is only
charged once per charging, and it is used to enforce the use of an appropriate charger (i.e.,
BEB that needs to charge with low priority should not consider high priority). The other
𝜄 𝑚 ℎ
component of the charging cost (i.e.,𝑐. 𝑝(𝑡) , 𝑐. 𝑝(𝑡) , and 𝑐. 𝑝(𝑡) ) is the cost of energy per
kWh. This cost is varied with the capacity of the charger, and hence, the objective
function (3.50) can now be expressed as follows:
𝑛
(3.54)
𝚤 𝜄
. ∝𝚤 . (𝑐𝑓𝑙 + 𝑐. 𝑝(𝑡) )
𝓂 𝑚 ℎ
+ 𝑁𝑐ℎ . ∝𝑚 . (𝑐𝑓𝑚 + 𝑐. 𝑝(𝑡) ℎ
) + 𝑁𝑐ℎ . ∝ℎ . (𝑐𝑓ℎ + 𝑐. 𝑝(𝑡) )
𝑙 𝑙 𝑚 𝑚 ℎ ℎ
+ 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 + 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 + 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 + 𝑁𝑏 . 𝐶𝑏
This section described the optimisation model of the multi-terminals priority charging
strategy and the associated constraints. The implementation of this optimisation algorithm
and the result of a simulation example is presented in the next chapter. The following
section considered an integrated model that added an on-road inductive charging system
to the multi-terminal priority-charging present in this section.
3.3.4 Optimisation problem for the allocation of the priority integrated

charging model
In this section, the transit electrification system considered is the multi-terminal

allocation of charging infrastructure for a bus transit system where both conductive and
inductive charging systems are employed. This approach is considered with the aim to
reduce the cost of investment and makes the operational service reliable by optimising
the charging station infrastructure: charger-operating capacities, the onboard battery
capacity, location of inductive transmitters, length of inductive cables and the numbers

of chargers. This section presents the mathematical model that integrates conductive and
inductive charging infrastructure for the multi-terminal BRT system (Figure 3-5).
The objective equation for the integrated charging model is expressed as:
𝑛
(3.55)
𝚤 𝜄
. ∝𝚤 . 𝑃(𝑡) 𝜄
. (𝑐𝑓𝑙 + 𝑐. 𝑝(𝑡) )
𝓂 𝑚 𝑚
+ 𝑁𝑐ℎ . ∝𝑚 . 𝑝(𝑡) . (𝑐𝑓𝑚 + 𝑐. 𝑝(𝑡) )
ℎ ℎ ℎ 𝑙 𝑚
+ 𝑁𝑐ℎ . ∝ℎ . 𝑝(𝑡) . (𝑐𝑓ℎ + 𝑐. 𝑝(𝑡) 𝑙
) + 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 𝑚
+ 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡
ℎ ℎ
+ 𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 + 𝑁𝑏 . 𝐶𝑏 + 𝑁𝑏 . 𝐶𝑊𝑃𝑅 + 𝑁𝑊𝑃𝑇 . 𝑇𝑐𝑎𝑝 . 𝐶𝑊𝑃𝑇
𝑁
+ ∑ 𝐶𝑐𝑎𝑏𝑙𝑒 (𝑑𝑖𝑒 − 𝑑𝑖𝑠 ) + 𝑁𝑐ℎ

𝑖𝑛𝑑 𝑖𝑛𝑑
. 𝐶𝑖𝑛𝑠𝑡
𝑖=1
In addition, the various terms in this equation are as follows:
 The first term is the cost of the BEBs battery packs (𝑁𝑏 . 𝐸𝑏𝑐𝑎𝑝 . 𝐶𝑏𝑏𝑎𝑡 ), expressed
as the product of the number of BEBs, their battery capacity, and the battery cost.
 The second term is the cost of the low-priority charging units,
𝚤 𝜄 𝜄
(𝑁𝑐ℎ . 𝑃(𝑡) . ∝𝚤 . (𝑐𝑓𝑙 + 𝑐. 𝑝(𝑡) )), as the product of the number of low-priority
chargers, low-priority charger rated power, and the low-priority charger
expenditure cost.
 The third term is the cost of the medium-priority charging units,
𝑚 𝑚 𝑚
(𝑁𝑐ℎ . 𝑃(𝑡) . ∝𝑚 . (𝑐𝑓𝑚 + 𝑐. 𝑝(𝑡) )), as the product of the number of medium-priority
chargers, medium-priority charger rated power, and the medium-priority charger
expenditure cost.
 The fourth term is the cost of the high-priority charging units,
ℎ ℎ ℎ
(𝑁𝑐ℎ . 𝑃(𝑡) . ∝ℎ . (𝑐𝑓ℎ + 𝑐. 𝑝(𝑡) )) as the product of the number of high-priority
chargers, high-priority charger rated power, and the high-priority charger
expenditure cost.
 𝑙
The fifth terms describe the charger installation cost ( 𝑁𝑐ℎ 𝑙
. 𝐶𝑖𝑛𝑠𝑡 𝑚
, 𝑁𝑐ℎ 𝑚
. 𝐶𝑖𝑛𝑠𝑡 ,
ℎ ℎ
𝑁𝑐ℎ . 𝐶𝑖𝑛𝑠𝑡 ) for the low, medium, and high-priority charging, respectively. The
charger installation cost is the product of the number of chargers and the

installation cost per charger. The installation cost is considered to vary for each
charging priority.
 The 𝑁𝑏 . 𝐶𝑏 , 𝐶𝑊𝑃𝑅 is the numbers of BEB, cost of each BEB, and cost of wireless
power receiver respectively.
 The term 𝑁𝑊𝑃𝑇 . 𝑇𝑐𝑎𝑝 . 𝐶𝑊𝑃𝑇 describe the cost of the wireless power transmitters-
where 𝑁𝑊𝑃𝑇, 𝑇𝑐𝑎𝑝 , 𝐶𝑊𝑃𝑇 is the number, capacity and cost of wireless power
transmitters, is the respectively.
 The term 𝐶𝑐𝑎𝑏𝑙𝑒 (𝑑𝑖𝑒 − 𝑑𝑖𝑠 ) is used to describe the cost of the inductive cables.
 𝑖𝑛𝑑
And the 𝑁𝑐ℎ 𝑖𝑛𝑑
. 𝐶𝑖𝑛𝑠𝑡 represent the cost of installing the inductive chargers.
It is worth noting that all the inductive transmitters considered in this work are of identical
power capacity. The duration for the inductive charging for the b-th BEB on the j-th trip
𝑖𝑛𝑑
in d-th assignment is represented with 𝑡𝑐ℎ(𝑏.𝑗.𝑑) which is equivalent to the time for the
𝑡𝑟𝑎𝑣
BEB to travel distance (𝑑𝑖𝑛𝑑(𝑏,𝑗,𝑑) ) as shown in Figure 3-8. The figure illustrates the
location of the inductive chargers between two BEB terminals; the BEB is considered to
receive a charge at the start of the trip, and the on-road inductive charger top-up the BEB
onboard battery while moving on the part of the route that has an integrated on-road
inductive charger.
Figure 3-8 Model of two-transit terminal
Here, the charge added to the BEB battery as it passes across the inductive charger is
proportional to the time the BEB takes to travel above the transmitter.
Hence, the energy gain is a result of the inductive charging depends on the distance
𝑡𝑟𝑎𝑣
covered with the inductive cable (represented with 𝑑𝑖𝑛𝑑(𝑏,𝑗,𝑑) in Figure 3-8), the time to

𝑡𝑟𝑎𝑣 𝑡𝑟𝑎𝑣
travel the distance 𝑑𝑖𝑛𝑑(𝑏,𝑗,𝑑) represented with 𝑡𝑖𝑛𝑑(𝑏,𝑗,𝑑) and the power capacity of the
𝑖𝑛𝑑
inductive charger 𝑃𝑐ℎ . In a route that is equipped with an inductive charging system
between terminals (as shown in Figure 3-8), the BEB battery energy level after the
𝑡𝑟𝑎𝑣
travelling distance 𝑑𝑖𝑛𝑑(𝑏,𝑗,𝑑) is expressed as:
𝑡𝑟𝑎𝑣
𝑑𝐸 (𝑡𝑖𝑛𝑑(𝑏,𝑗,𝑑) ) 𝑖𝑛𝑑 𝑖𝑛𝑑 𝑡𝑟𝑎𝑣 (3.56)
= −𝑃𝑢𝑠𝑒(𝑏,𝑗,𝑑) + 𝑃𝑐ℎ (𝑡𝑖𝑛𝑑(𝑏,𝑗.𝑑) ) ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
𝑑𝑡
𝑖𝑛𝑑
Where 𝑃𝑢𝑠𝑒(𝑏,𝑗,𝑑) is the rate of consumption when the b-th BEB on the j-th trip in d-th
𝑡𝑟𝑎𝑣 𝑡𝑟𝑎𝑣 𝑖𝑛𝑑
assignment has travelled the distance 𝑑𝑖𝑛𝑑(𝑏,𝑗,𝑑) in the time (𝑡𝑖𝑛𝑑(𝑏,𝑗,𝑑) ) and 𝑃𝑐ℎ is the
𝑡𝑟𝑎𝑣
energy supply to the battery at the time 𝑡𝑖𝑛𝑑(𝑏,𝑗,𝑑) that is required to travel a distance
𝑡𝑟𝑎𝑣
𝑑𝑖𝑛𝑑(𝑏,𝑗,𝑑) . Then integrating (3.56) gives
𝑡𝑟𝑎𝑣 𝑡𝑟𝑎𝑣
𝑡=𝑡𝑖𝑛𝑑(𝑏,𝑗,𝑑) 𝑡=𝑡𝑖𝑛𝑑(𝑏,𝑗,𝑑) (3.57)
𝑎𝑟𝑟 𝑖𝑛𝑑 𝑖𝑛𝑑
𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) −∫ 𝑃𝑢𝑠𝑒(𝑏,𝑗,𝑑) 𝑑𝑡 + ∫ 𝑃𝑐ℎ 𝑑𝑡 ≤ 𝐸𝑏𝑚𝑎𝑥 ∀𝑏
𝑡=0 𝑡=0
∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
𝑎𝑟𝑟
Where 𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) is the energy capacity in the BEB battery at the start of
inductive charging.
Hence, equation 3.57 can be rewritten as:
𝑎𝑟𝑟 𝑢𝑠𝑒 𝑟𝑒𝑐

𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) − 𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) + 𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) ≤ 𝐸𝑏𝑚𝑎𝑥 ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 (3.58)
𝑔𝑎𝑖𝑛 𝑟𝑒𝑐 𝑢𝑠𝑒 (3.59)

𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) = 𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) − 𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
𝑢𝑠𝑒 𝑔𝑎𝑖𝑛 (3.60)

𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) ≤ 𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
𝑔𝑎𝑖𝑛 𝑟𝑒𝑐 (3.61)

𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) ≤ 𝐸𝑖𝑛𝑑(𝑏,𝑗,𝑑) ∀𝑏 ∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷
In order to calculate the length of the inductive cable, let the starting and ending points
of the inductive cable of the 𝑖𝑡ℎ power transmitter be 𝑑𝑖𝑠 and 𝑑𝑖𝑒 , respectively. The
inductive cable length of the 𝑖𝑡ℎ power transmitter is expressed as 𝑑𝑖𝑒 − 𝑑𝑖𝑠 , as shown in
Figure 3-9. This expression should satisfy the following:

Figure 3-9 Allocation of inductive cable length
𝑑𝑖𝑒 > 𝑑𝑖𝑠 𝑓𝑜𝑟 𝑖 = 1, … … , 𝑁 (3.62)
𝑑𝑖𝑒 < 𝑑𝑖+1

𝑠
𝑓𝑜𝑟 𝑖 = 1, … … , 𝑁 − 1 (3.63)
Another constraint is the number of wireless power transmitters (𝑁𝑊𝑃𝑇 ). The number of
power transmitters can be expressed as the transit energy ratio to the power transmitter's
installed capacity as given in (3.64). This energy ratio is evaluated by subtracting the
energy demand of the conductive chargers from the transit system energy demand and
dividing by the transmitter’s power capacity.
𝑡𝑗 (3.64)
𝑛 𝑛
1 𝑐ℎ,𝑙 𝑐ℎ,𝑚 𝑐ℎ,ℎ
𝑁𝑊𝑃𝑇 ≥ {∑ ∫ 𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡 − ∑ 𝐸𝑏,𝑗,𝑑 + 𝐸𝑏,𝑗,𝑑 + 𝐸𝑏,𝑗,𝑑 } ∀𝑏
𝑇𝑐𝑎𝑝
𝑗=1 𝑡1 𝑗
∈ 𝐵 ∀𝑗 ∈ 𝐽𝑑 ∧ ∀𝑑 ∈ 𝐷 ∧ ∀𝑡 ∈ 𝜏
This section has presented the multi-terminal priority charging optimisation model in
combination with an on-road inductive charging. This integrated model is considered a
trade-off option for the infrastructure need of a BEB transit system. The simulation work
present in the next chapter compared these models while considering various charging
scenarios.
3.4 Design of Inductive Charging System for Bus Rapid Transit

Network
Generally, BEBs are more financially expensive than diesel buses [2], and the
conventional charging system limits the driving range [81]. The charging systems that

support long driving distances are still in the early phases of development and very
expensive [14]. Moreover, the schedule for BEBs needs to consider a variety of
parameters about the energy system [81],[14]. Mitigating these issues while observing
the operational bus schedules and the adequate provision for trip demand energy need to
be considered. Different strategies have been proposed in the literature, and the common
theme includes charging and battery swapping [100] and repurposing infrastructure [295].
Under a battery swapping system, the BEB can replace its depleted battery pack with a
full one within a few minutes. The drained battery packs are then recharged at the station
and later swapped for other arriving BEBs [295],[296]. However, the battery swapping
system requires a significant extra investment and depot investment [297]. Hence, most
of the research work on BEBs has been concentrated on the optimal sizing of battery and
charging infrastructures to satisfy the BEBs' transit operation plan [81],[82],[298].
Specifically, the authors in [299] develop an optimisation model for charging station
design in fully electric transport networks. Moreover, in [298] a Spatio-temporal
optimisation model for determining the best deployment methods for the BEB system, is
presented.
The wireless charging technology implies transmitting energy from a power transmitter
to the BEB that charges on-road during its motion over the inductive cable. The
improvements in wireless power transfer (WPT) technologies have aided various
commercial applications. The application of WPT is common in small electronic devices
such as mobile phones and small personal devices [300], [301], [302]. WPT technology
can charge the batteries in such equipment without a wired connection [302]. Recently,
the WPT has applications in the charging of electric vehicles [302]. In 2006, a team of
MIT scholars introduced a resonant coupler that transmits a considerable quantity of
power at a low frequency for a mid-length distance [10]. The authors in [11] developed
the commercial applications of WPT for charging vehicles in parks and garages. Hino
Motors and Showa Aircraft in Japan developed inductive coupling wireless power
transfer to an electric bus in 2009. In [13], the authors examine the power transmitter's
allocation for the on-line electric vehicle (OLEV). This work (ref [13]) considered the
trade-off between battery size and the positions of power transmitters on a circular route.
However, the authors fail to account for the multi-terminal bus transit system with the

bus terminals as the charging transmitter's primary location. The authors also fail to
consider the impact of route traffic situations.
In this section, a new mathematical model is developed to allocate multi-terminal

inductive charging infrastructure to optimise power transmitters' location, the length of
the inductive cable, battery capacity, and the electrified transit system cost. The
formulated optimisation model minimises capital cost while considering the system
operational constraints imposed by the transit system.
3.4.1 The BRT inductive charging system overview

The inductive charging system is an integrated on-road charging system that charges
the BEB battery remotely while the bus moves over the installed inductive cable area.
The length of the inductive cable, the size of BEB's battery, and the number of transmitters
installed on the road directly affect the transit system's overall performance and cost.
Figure 3-10 BRT route with buses
The focus of this work is to introduce the BRT network inductive charging system design
problem. This design problem is illustrated with a mathematical model used to evaluate
the optimal combination of the inductive transit system's critical design parameters (i.e.
battery size, length of the inductive cable, and the power transmitters' allocation) as
shown in Figure 3-11.

Figure 3-11 Design of the BEB inductive transit system
3.4.2 Wireless Charging System

WPT system generally consists of a power supply, transmitter (primary coil), receiver
(secondary coil), oscillator circuit and matching circuit [210]. The A.C. supplied by a
power source is changed into low-frequency A.C. using an oscillator. The oscillator's
output is fed into the push-pull circuit that supplies the transmitter coil. This transmitter
coil is also referred to as the primary coil. This primary coil transmits this power to the
receiver coil that is separated by a certain distance; the power received by the secondary
coil is then rectified and regulated before output to the BEB battery charger [305], [306].
The inductive charging system is environmentally suitable and cheap to maintain because
water, ice, dirt and chemicals do not affect them. A more detailed description of the
system's power transfer mechanism and hardware configuration can be found in [211],
[212], [301], [305], [306].
The wireless power transfer technology enables the motor to obtain power remotely
while the vehicle is on track, eliminating the need for the vehicle to idle while the battery
is recharged. As a result, the BEB does not need a large battery. The downside of wireless
power transfer charging systems (i.e. inductive chargers) is the high initial investment.
Compared to the existing plug-in electric vehicle infrastructure, inductive chargers need
a significant initial investment to install the power transmitters. However, the smaller
battery size and improved running time achieved by eliminating the vehicles' recharging
downtime can compensate for this initial expense. Specifically, there is a trade-off
between battery capacity and power transmitter allocation. If the vehicle has a large
battery, fewer transmitters are needed. However, a vehicle with a small battery may need
more frequent recharges, necessitating the installation of additional transmitters. Given

that the battery and power transmitter costs account for a sizable portion of the inductive
charging transit system's overall investment cost, it is vital to investigate this trade-off.
Figure 3-12 Proposed BRT inductive charging system
In this work, the BEB is considered to operate on a road with an installed power
transmitter. As shown in Figure 3-12, the wireless power receiver device is deemed to be
installed at the lower part of the BEB bus. This wireless power receiver device remotely
collects electricity from the transmitter. The wireless receiver delivers power to the motor
or the battery, or both, depending on the motor's power requirement and the battery
charging level [12]. When the BEB is moving on a busway where no power transmitter
unit is installed, the motor in the BEB uses the power from the battery. To eliminate
energy waste, the BEBs that charge via inductive means is considered to have an infra-
red sensor that turns 'ON' the supply that activates the energy exchange between the
transmitter and the receiver circuit. When the BEB moves away from the inductive cable,
the infra-red sensor sends the signal to turn 'OFF' the power supply [211].
3.4.3 Mathematical optimisation model
This optimisation model aims to minimise the total cost of the electrified transit system;
this includes BEB cost 𝐶𝐵𝐸𝐵 , inductive charger cost (the receiver 𝐶𝑊𝑃𝑅 and the transmitter
𝐶𝑊𝑃𝑇 ), battery cost 𝐶𝑏𝑎𝑡 and the inductive cable cost 𝐶𝑐𝑎𝑏𝑙𝑒 , as follows
𝑀𝑖𝑛(𝑁𝐵𝐸𝐵 . 𝐶𝐵𝐸𝐵 + 𝑁𝐵𝐸𝐵 . 𝐶𝑊𝑃𝑅 + 𝑁𝐵𝐸𝐵 . 𝐵𝑐𝑎𝑝 . 𝐶𝑏𝑎𝑡 + 𝑛. 𝑇𝑐𝑎𝑝 . 𝐶𝑊𝑃𝑇 (3.66)
𝑛
+ ∑ 𝐶𝑐𝑎𝑏𝑙𝑒. (𝑑𝑖𝑒 − 𝑑𝑖𝑠 ))

𝑖=1

The first term in (3.66) represents the cost of BEBs for the number of BEBs 𝑁𝐵𝐸𝐵 . The
second expression refers to the price of Wireless power receivers. The wireless power
receivers (𝑊𝑃𝑅𝑠) are devices mounted beneath BEBs that accept electricity from the
transmitters and use it to charge the BEB's onboard battery. The received power is
delivered to the motor or the battery, or both. The power transferred to the motor or the
battery depends on the motor's power requirement and the battery state of charge; hence,
each BEB operating on the BRT route is equipped with a WPR. The third term is the
battery cost multiplied by the number of BEBs operating on the BRT route. There
are 𝑁𝐵𝐸𝐵 BEBs, each with a single identical battery. The battery's cost is proportional to
its maximum capacity, which is specified by the decision variable 𝐵𝑐𝑎𝑝 . The fourth term
denotes the transmitter's cost. Each transmitter unit is connected to an inductive cable;
hence, if a transmitter unit is installed, there is a cost for the transmitter and the inductive
cable. There are n transmitter units; thus, the terms describing the cost of the transmitter
are multiplied by n. It is worth noting that the cost of the cable increases linearly with the
cable's length. As a result, the cost of each cable unit is denoted by the 𝐶𝑐𝑎𝑏𝑙𝑒. (𝑑𝑖𝑒 − 𝑑𝑖𝑠 ).
As a result, the fifth term represents the total cost of the cables.
The BEBs operate on a fixed route with multiple terminals and stops to load and unload
passengers. The BEBs considered in this work are identical and operated according to the
operational schedule or timetable set by the transit operators. For a given BEB, the rate
of using energy is represented by the power used over the operational time ( 𝑃𝑢𝑠𝑒 (𝑡)), and
the energy via charging is given as the power gain for the charging duration 𝑃𝑔𝑎𝑖𝑛 (𝑡).
Figure 3-13 Mathematical model transit inductive system
Consider the situation in which a BEB is about to leave an area that has a power
transmitter. In this modelling concept, the BEB is located at 𝑑𝑖𝑒 as shown in Figure 3-14,
and the time at which the BEB is at this location is symbolised by 𝑡𝑖+1 as shown in Figure
3-13. The battery's state of charge at this point is expressed by the equation given

in (3.67). Once the BEB begins going along a route without a transmitter, the energy in
the battery is used at the rate of 𝑃𝑢𝑠𝑒 (𝑡), i.e., the BEB's power consumption rate at time t.
𝑡𝑖+1 𝑡𝑖+1
(3.67)
𝐸𝑡𝑖−𝑖+1 = ∫ 𝑃𝑔𝑎𝑖𝑛 (𝑡)𝑑𝑡 − ∫ 𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡
𝑡𝑖 𝑡𝑖
Here 𝐸𝑡𝑖−𝑖+1 represents the energy received while the BEB is moving from 𝑡𝑖 to 𝑡𝑖+1 as
shown in Figure 3-13. The received energy is considered higher than the energy used.
𝑃𝑔𝑎𝑖𝑛 (𝑡) > 𝑃𝑢𝑠𝑒 (𝑡) (3.68)
When the BEB reaches the transmitter's starting point at a time 𝑡𝑖+2 , the total energy
consumption from the point 𝑡𝑗 to point 𝑡𝑗+1 , is calculated by integrating 𝑃𝑢𝑠𝑒 (𝑡) over 𝑡𝑗
and 𝑡𝑗+1 . This number should be more than the minimum charge level 𝐸𝑚𝑖𝑛 ; thus, we can
state the following constraint:
𝑡𝑗+1
(3.69)
𝐸𝑡𝑖−𝑖+1 − ∫ 𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡 ≥ 𝐸𝑚𝑖𝑛
𝑡𝑗
If the BEB continues to travel from point 𝑡𝑖+2 , the charge level will increase since it
is now moving on the path equipped with a transmitter, and the battery is receiving energy
wirelessly from the transmitter. 𝑃𝑔𝑎𝑖𝑛 (𝑡), is the quantity of charge added to the battery is
proportional to the time spent by the BEB moving over the transmitter. As a result, the
motor consumes charge at the rate of 𝑃𝑢𝑠𝑒 (𝑡), while the BEB adds charges at the rate of
𝑃𝑔𝑎𝑖𝑛 (𝑡) . When the battery charge level reaches its maximum, i.e., 𝐸𝑚𝑎𝑥 , the pickup
device's power is disconnected to protect the battery from overcharging. As a result, at
any point along the BEB transit route, the charge level in the onboard battery is given as:
𝑡𝑖+1 𝑡𝑖+1 𝑡𝑗+1
𝐸𝑡𝑖−𝑖+1 − ∫ 𝑃𝑔𝑎𝑖𝑛 (𝑡)𝑑𝑡 − ∫ 𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡 + ∫ 𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡 ≤ 𝐸𝑚𝑎𝑥

𝑡𝑖 𝑡𝑖 𝑡𝑗
𝑓𝑜𝑟 𝑖 = 1, … … , 𝑛 − 1 𝑎𝑛𝑑 𝑓𝑜𝑟 𝑗 = 1, … … , 𝑚 − 1

In this work, the cost of battery per BEB accounts for a substantial portion of the transit
system's electrification total investment cost. The battery capacity (𝐵𝑐𝑎𝑝 ) is considered as
one of the critical decision variables of the objective equation. This battery capacity
(𝐵𝑐𝑎𝑝 ) is expressed as the function of the transit system energy dynamics as
𝑡𝑖+1 𝑡𝑖+1 𝑡𝑗+1

𝑛 𝑚 (3.70)
𝐵𝑐𝑎𝑝 ≥ ∑(∫ 𝑃𝑔𝑎𝑖𝑛 (𝑡)𝑑𝑡 − ∫ 𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡) + ∑ ∫ 𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡
𝑖=1 𝑡𝑖 𝑡𝑖 𝑗=1 𝑡𝑗
Another constraint is the number of wireless power transmitters, 𝑁𝑊𝑃𝑇 . The number of
power transmitters can be expressed as the transit energy ratio to the power transmitter's
installed capacity as given in (3.71).
𝑡 𝑡 (3.71)
{∑𝑛𝑖=1 ∫𝑡 𝑖+1 𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡) + ∑𝑚
𝑗=1 ∫𝑡
𝑗+1
𝑃𝑢𝑠𝑒 (𝑡)𝑑𝑡 }
𝑖 𝑗
𝑁𝑊𝑃𝑇 ≥
𝑇𝑐𝑎𝑝
Also, consider the starting point and endpoint of the inductive cable of the 𝑖𝑡ℎ power
transmitter be 𝑑𝑖𝑠 and 𝑑𝑖𝑒 , respectively. The inductive cable length of the 𝑖𝑡ℎ power
transmitter is expressed as 𝑑𝑖𝑒 − 𝑑𝑖𝑠 , as shown in Figure 3-14. This expression should
satisfy (3.72) and (3.73).
Figure 3-14 Allocation of inductive cable
𝑑𝑖𝑒 > 𝑑𝑖𝑠 𝑓𝑜𝑟 𝑖 = 1, … … , 𝑛 (3.72)
𝑑𝑖𝑒 < 𝑑𝑖+1

𝑠
𝑓𝑜𝑟 𝑖 = 1, … … , 𝑛 − 1 (3.73)
𝑛
(3.74)
∑(𝑑𝑖𝑒 − 𝑑𝑖𝑠 ) < 𝐿𝑟
𝑖=1

The section looks at a multi-terminal inductive base model of BEB's transit charging
infrastructure. The mathematical optimisation model examines the provision of inductive
charging infrastructure, including battery sizes, transmitter position, and inductive cable
length. With the trade-off analysis of the component of an inductive transit system, the
next chapter gives a simulation example that explores several scenarios that maximise the
utilisation of the inductive charging infrastructure while minimising the infrastructure
cost.
3.5 Chapter summary

First, section 3.1 presents the transit fleet routing model related to the operating
timetable. The operational schedule for the transit network is fed into the BEBs transit
fleet model. This fleet routing model would result in transit network electrification
parameters that meet the operational needs of the BEB transit system.
Second, in section 3.2, the mathematical model for BEB energy consumption is
described based on a longitudinal technique that accounts for dynamics. The energy
consumption model is used to determine the BEBs' energy consumption in kWh/km using
transit fleet data and network topology data (i.e., BEBs weight, efficiencies, speed
profiles, and routes distance and elevation profile). Therefore, the output of the BEBs
transit fleet model and the energy consumption model and operational needs are
incorporated into the BEB transit system’s proposed novel charging infrastructure model.
Section 3.3 presents the technique for modelling charging infrastructure for BRT
transport fleets and the mathematical and optimisation formulations for the novel BEB
charging system model 3.3.2, multi-terminal charging model 3.3.3, and integrated
charging model 3.3.4. The integrated charging model is a multi-terminal charging
infrastructure allocation strategy using conductive and inductive charging methods for a
bus transit system. This approach has been considered to cut the cost of investment and
increase the reliability of the operational service by optimising the charging station
infrastructure. At the same time, examine the trade-off of the charger-operating
capacities, onboard battery capacity, number of conductive priority chargers and number
of inductive chargers. In summary, the general aim of the three objectives of the

optimisation problem introduced in this section is to minimise the electrification

expenditure cost of bus transit networks while considering different charging options.
Figure 3-15 Summary of BEB transit fleet design and mathematical modelling
Finally, in section 3.4, a mathematical model for assigning multi-terminal inductive

charging infrastructure is presented to optimise power transmitter location, inductive
cable length, battery capacity, and the cost of an electrified transportation system. The
inductive charging system is an integrated on-road charging technology that charges the
BEB battery while the bus moves through the inductive cable route. The overall
efficiency and cost of the transit system are directly influenced by the length of the
inductive cable, the size of BEB’s battery, and the number of transmitters deployed on
the road. The summary of the sections covered in this chapter is given in Figure 3-15.
The simulation study in the next chapter looked at these models while considering
various charge scenarios.

Chapter 4: BEB Transit fleet electrification numerical Case Studies
4 BEB TRANSIT FLEET

ELECTRIFICATION
NUMERICAL CASE STUDIES
4.1 Introduction
The numerical case studies shown in this chapter used the optimisation model
described in Chapter 3 and analysed the several modelling options described in Chapter
3. First, a simulation is performed to determine the average power consumption of BEBs
given a classified traffic situation and speed profile for a particular case study. Then, a
novel charging optimisation tool is developed to demonstrate how multiple charging
strategies can be optimally employed to study essential factors of the transit system: such
as the trade-offs between alternative charger designs, charger locations, battery sizes, and
cost.
In order to estimate the average power requirement of the BEB, simulation tools and
algorithms are developed to evaluate the power consumption of BEBs travelling on
defined driving cycles. The BEB power requirement simulation is accomplished using
the 'Advanced Vehicle Simulator' (Advisor), an open-source software application tool
developed for the US Department of Energy by the National Renewable Energy
Laboratory [30]. Numerous authors and international laboratories have proven its
accuracy [31], [32].
The derived energy requirements figures are combined with transit data to define the
system's performance requirements. The systems under consideration involve designing
novel charging infrastructures for the BEB transit system to minimise charging costs and

downtime. Additionally, the inductive charging simulation system that optimises the
location of power transmitters, the length of the inductive wire, battery capacity, and the
cost of the electrified transit system is also studied. The developed optimisation model
minimises capital costs while considering the transportation system's operational
constraints. Moreover, the GHG emission reduction is estimated. This study aims to
determine whether such systems are financially reasonable the extent to which this can,
contribute to direct transport GHG emission reduction programmes in developing nations,
and improve transit infrastructure and alleviate congestion in emerging cities.
The BEB transit fleet optimisation model developed is based on the particle swarm
algorithm PSO. Figure 4-1 shows the overview of the simulation work present in this
chapter. The figure shows that the Advisor modelling tool simulates the BEB energy
consumption and estimates the GHG emission figure. The PSO optimisation is used to
allocate charging infrastructure, trade-off the transit system components and optimise for
the minimum cost of infrastructure configuration. It utilises the three proposed priority
charging models described in Chapter 3 and the proposed inductive charging model as
the bases for studying the BEB fleet electrification.
Figure 4-1 Modelling elements and simulation overview

The Lagos BRT case study is considered because both the Lagos state government and
the federal government of Nigeria have been eager to adopt measures to alleviate traffic
congestion, promote transit use, and significantly reduce CO2 emissions along this route.
Nonetheless, the technique used in this study could be used to examine the electrification
of road transit systems in other comparable countries.
This is examined in detail in this thesis, and up-to-date data was gathered from a reputable
sources, including the Lagos Metropolitan Area Transport Authority (LAMATA) and
PRIMERO Transport Services Limited [307]. This simulation tool and method also
employed national traffic statistics, road length data, and driving cycle profiles. Valuable
data was gathered from the websites of BEB manufacturers (examples include BYD
ADL), and some cost estimates were made using the relevant project as a baseline.
The remainder of this chapter begins with a summary of the case study. The Advanced
Vehicle Simulator (ADVISOR) modelling tool and BEB energy consumption simulation,
are discussed in detail. The novel method for BEB charging simulation (Scenario I, II, III
and IV) is then provided.
4.2 Overview of Lagos Bus Rapid Transit-Case study

Lagos is the most populated city in Nigeria and Africa, with 21 million inhabitants
[308]. The rapid growth in Lagos's population (the current estimated growth rate is 6%)
generates increasing travel demand, longer trip distances, more trips by car and more car
ownership, all of which add to more traffic congestion and pollution in Lagos city [309].
BRT is a sustainable transit system that can compete with and supplement rail systems
in providing high-quality transport services. It is a high-quality bus-based transit system
that offers fast, comfortable, and cost-effective urban mobility through segregated right-
of-way infrastructure and rapid, reliable and frequent operations [31]. The BRT system's
unique advantages over the metro systems include lower operating and capital costs,
higher flexibility, and short implementation time. These benefits have made BRT
predominant in developing countries with insufficient funds for public infrastructure.

Figure 4-2 Lagos BRT route with the location of bus stops (Source: PRIMERO)
The Lagos BRT scheme is on the Ikorodu-TBS passageway in the Lagos metropolis.
This route is one of the busiest and most essential routes with high passenger traffic. The
Ikorodu-TBS passageway cuts across seven local government areas and leads to the major
Lagos Central Business District (CBD). The route is divided by a median and has several
intersections, which are traffic signalised. The route from Ikorodu provides one of the
main principal links within Lagos State, linking the exterior settlement of Ikorodu to
Lagos Island and other intermediate settlements. The entire BRT Scheme is 38km long,
running from Ikorodu to TBS, as shown in Figure 4-2. The transit system is equipped
with Electronic Ticketing and Intelligent Transport System. The Lagos BRT currently has
44 bus shelters in both directions, six terminals, 19 intersections, two bus depots with a
maintenance bay, an automatic washing bay, administrative offices, a fuel dump, and
other appurtenances smooth operation of the system [310] [311].
Primero Transport Services Limited (PTSL), a private company that took over the
transit operations in November 2015, manages the Lagos BRT scheme. PTSL has added
434 buses to the transit operations from Ikorodu to TBS and has shuttled over 101 million

commuters from November 2015 to date [310]. Since its inception, the BRT scheme in
2008 has conveyed over 400 million passengers [310]. The BRT Operates 16 hours every
day from 6:00 am to 10:00 pm. The transit operation schedule is divided into two shifts,
with 8 hours of driving per shift (AM & PM). In 2020, pre-covid, the government planned
to extend the busway to Oworonshoki-Apapa, Ota-CMS, TBS-Berger, Berger-Iyana
Ipaja-Ikotun and Jbowu-Iddo [310] with some delays still in place.
The greening of the Lagos BRT system is an attractive option to reduce pollution and
promote policies that can eliminate driveway congestion. Therefore, given its relevance,
the data used for this work is for the Lagos BRT network.
4.3 BEB Energy Consumption Simulation

A mathematical analysis based on the longitudinal dynamic is provided in Chapter 3.
This section presents the simulation of this mathematical model, which is based on the
ADVISOR simulation tool. As a result, this section briefly explains the ADVISOR
modelling tool in relation to the energy consumption simulation. In addition, the details
of vehicle specification modelling, energy storage system and driving cycles, as well as
HVAC and route elevation, were provided. It is worth noting that the BEB energy
consumption figure is a vital input parameter in the charging infrastructure
allocation. Hence, the BEB energy consumption simulation presented in this section is an
essential step in designing and simulating charging infrastructure for the BEB transit
system.
4.3.1 The Advanced Vehicle Simulator (ADVISOR)
A study that investigates transit route-based BEB energy consumption and GHG
emission is carried out as part of the work of this thesis; because the simulated BEB
energy consumption figure is one of the model's input parameters that allocates charging
infrastructure for the BEB transit network. The transit fleet routing model design, which
links the route and operation circumstances, is provided in Chapter 3, along with the
mathematical model for the BEB energy consumption, which is based on a longitudinal

method that accounts for dynamics. Hence to simulate this developed model based on the
route characteristic of the case study, an ADVISOR simulation tool is adopted.
The National Renewable Energy Laboratory's Advanced Vehicle Simulator

(ADVISOR) was first created in November 1994 [312]. It was created as an analysis tool
to assist the US Department of Energy (DOE) in developing hybrid electric vehicle (HEV)
technologies through the Hybrid Electric Vehicle Propulsion System contracts with Ford,
General Motors Daimler Chrysler. Its primary function is to demonstrate the system-level
interactions between hybrid and electric vehicle components and their impact on vehicle
efficiency and fuel economy[257], [312].
The National Renewable Energy Laboratory developed ADVISOR in the

MATLAB/Simulink environment. ADVISOR offers an easy-to-use, scalable, robust, and
supported research kit for advanced vehicle modelling to the vehicle engineering
community. It is primarily used to measure the fuel efficiency, output, and emissions of
vehicles equipped with alternative technologies such as fuel cells, batteries, electric
motors, and internal combustion engines operating in hybrid (multiple power source)
configurations. It is particularly adept at quantifying the relative change that can be
anticipated due to technology adoption compared to a baseline scenario [257], [312],
[313].
The previous study has demonstrated the value of the ADVISOR modelling tool. For
example, the authors of [314] present the ADVISOR software's design, explain its
combined backward–forward computation approach, and demonstrate its accuracy,
speed, and adaptability. [315], gives a practical overview of ADVISOR, including the
layout of the graphical user interface (GUI), the program's capabilities and limits, and
power source modelling options. Additionally, [316] detailed ADVISOR's battery
modelling capabilities and indicated that a resistance-capacitance model is preferred for
vehicle simulations. The authors of [317] used Simulink tools to model a hybrid drive
train for a postal service delivery vehicle. The authors of [318] used the ADVISOR tool
to analyse a basic vehicle used for studying hybrid fuel cell/battery passenger cars.
Furthermore, the authors of [319] altered the parallel block diagram in ADVISOR to
create a simple electric vehicle. The vehicle required controls and components that were
not included in the software. The simulation and data analysis tools were found to be

convenient and helpful in establishing the capabilities of their design. Using the
ADVISOR model, the authors in [100] computed the BEB's kWh/km energy usage along
specific routes. The authors improved the accuracy of the BEB energy consumption value
by using the BEB speed profile, route terrain, and auxiliary load rating in this work.
Nonetheless, the author fails to consider the impact of weather-related uncertainty on the
BEB energy consumption value.
As a result, the usage of ADVISOR in this work is in a competitive context and distinct
from past research. This work incorporates five traffic flow conditions that mimic the
case study situation into the model in order to compare the impact of traffic conditions on
the BEB energy consumption figure and to improve the model's accuracy. The data
focuses on traffic volume and road capacity, with traffic conditions classified as free,
move, appropriate, inappropriate, and critical. The matching average speed profile from
the ADVISOR database was used to select this graded traffic condition.
4.3.2 The Advanced Vehicle Simulator (ADVISOR) structure and

capabilities
ADVISOR was written in MATLAB/Simulink. MATLAB is a simple-to-use matrix-

based programming environment for performing calculations, while Simulink is used to
graphically model complex structures using block diagrams [257]. The software uses an
iterative calculation approach to produce outputs of a vehicle's velocity and energy
consumption at all moments during a simulation. ADVISOR is the only simulation engine
that integrates backwards and forwards simulation features.
In the Forward-facing approach, the vehicle simulators provide a driver model that
considers the necessary speed and the current speed while developing acceptable throttle
and brake commands (often through a PI controller). The throttle command is then
converted into engine (and/or motor) torque and energy consumption rate. The engine's
torque is fed into the transmission model, converting it to the transmission's efficiency
and gear ratio. The computed torque is then transmitted forward through the drivetrain in
the vehicle's physical power flow direction until it reaches the tire/road interface as a
tractive force. The forward-facing method is very useful when creating hardware and
running extensive control simulations. Because forward-facing models deal with

quantities that can be measured in a real-world drivetrain, such as control signals and true
torques. The fundamental drawback of the forward-looking strategy is its slow simulation
speed. Drivetrain power calculations are based on vehicle states, which include computed
drivetrain component speeds via integration. Higher-order integration strategies with
small time steps are thus required to achieve robust and accurate simulation results. As a
result, forward-facing simulation for analysing BEB energy consumption figures can be
excessively time-consuming [257].
In contrast to the forward-facing technique that starts with a driver model, the required
vehicle velocity drives the backwards-facing approach. A driver model is not included in
the backwards-facing method. The force needed to accelerate the vehicle through the time
step is derived directly from the speed trace of the simulated driving cycle. The simulation
calculates the torque and speed of drive train components required to overcome the
vehicle's inertial forces and achieve the target velocity. The calculation is performed
backwards, beginning at the tire/road interface and finishing at the energy source. The
backwards-facing technique is unsuitable for researching control systems due to the lack
of throttle and brake information. A pure backward method is unsuitable for analysing
best-effort performance since the backwards-facing model implies that the vehicle
achieves the requisite speed trace (i.e., acceleration tests). ADVISOR can effectively take
advantage of improved battery and component models while retaining a comparably rapid
simulation speed by merging the forward- and backwards-facing approaches [320].
The user interacts with a succession of graphical user interface (GUI) windows to enter
various vehicle characteristics and drive cycle requirements and evaluate their impact on
vehicle performance, energy consumption, fuel economy, and emissions. The vehicle
input screen, the simulation setup screen, and the results screen are the three primary GUI
screens in ADVISOR. Figure 4-3 and Figure 4-4, illustrate examples of these screens.
The user builds a vehicle of interest on the vehicle input screen (Figure 4-3) by selecting
options from drop-down menus. Each list contains many pre-programmed items for use
in the vehicle. By altering the attributes of each part, the user can easily construct new
components. ADVISOR's revolutionary vehicle design and simulation are more
accessible by this functionality [1] [320].

Figure 4-3 ADVISOR vehicle input window
Figure 4-4 ADVISOR simulation setup window

The user defines the drive cycle settings for the event over which the vehicle is to be
modelled in the simulation setup screen (Figure 4-4). The results screen (Figure 4-5)
displays fuel economy and emissions and detailed charts of time-dependent outputs for
reviewing vehicle performance. The user can choose from a variety of output options
related to speed and torque, energy usage, emissions, battery charge level, Etc., and up to
four plots can be displayed simultaneously [257].
Figure 4-5 ADVISOR results window
4.3.3 ADVISOR BEB energy consumption simulation

In this work, The ADVISOR is used to compute the BEB energy consumption for the
Lagos BRT case study using the BEB parameters given in Table 4-1, the BEB speed
profile, route topography, trip duration, and the corresponding traffic flow conditions.
Hence the next section presents the BEB energy consumption simulation relevant to the
case studies for this thesis Advisor's database has a sizable number of regular vehicle
types, including light and heavy-duty vehicles equipped with conventional or all-electric
drivetrains. The user chooses components such as motors, batteries, vehicle mass, and

additional electric loads to mimic a particular vehicle's performance, fuel economy, and
emissions.
4.3.3.1 Vehicle design

The advisor vehicle input interface for vehicle design allows the user to pick and
customise the drive train configuration to emulate the real-world counterpart. This work
uses the ADVISOR vehicle input interface to pick a default EV, and the BEB parameters
are entered, as shown in Table 4-1.
All BEBs are assumed to be the same size and capacity in this work. The current bus
specification for the Lagos bus rapid transit network is 12m, with a total capacity of 80
passengers with up to 40 seats. It is worth noting that the passage capacity is used to
compute the cargo mass in kg. This mass is computed using the global average body mass
of 62kg [320].
Table 4-1 BEB Parameter (Source: BYD ADL Enviro200EV)
Vehicle Parameter Value
Vehicle type Pure electric
Dimensions ~12m
Gross Veh. Weight 18700kg
Passenger capacity 80 (40 seats)
Electric motors 2×90kW
Battery system lithium iron phosphate 348kWh
Charging system dual plug 2×40kW AC
Operational range Up to 160 miles
Aux. Power About 15kW
𝑆𝑂𝐶𝑏𝑚𝑎𝑥 , 𝑆𝑂𝐶𝑏𝑚𝑖𝑛 80%, 20%
4.3.3.2 Battery selection

The energy storage capacity is based on the battery capacity values on the
specifications of the BYD ADL Enviro200 EV. The lithium-iron-phosphate battery

capacity of the BYD ADL Enviro200EV pure electric bus is 348kWh [321]. The lithium
ion battery is selected as the battery for this simulation because the ADVISOR does not
have a lithium-iron-phosphate battery option. On the other hand, lithium-ion batteries
have a higher energy-to-weight ratio than other chemistries used in electric vehicles (such
as nickel-metal hydride or lead-acid) [322]. In ADVISOR, the default lithium-ion battery
pack is made up of 25 modules, each with a capacity of 6 Ah. The battery pack's nominal
voltage is 267 V, and capacity in kilowatt-hours may be determined using the formula
kWh=AhV. According to this kilowatt-hours computation, the default lithium-ion battery
model in ADVISOR has a capacity of around 1.6 kWh. As a result, the energy storage
capacity input variable ("ess_cap_scale") must be changed to the desired capacity. In this
study, a 348kWh capacity is used to keep the BYD ADL Enviro200 EV in its real-world
form.
4.3.3.3 Driving Cycle

The prime objective of the BEB energy consumption simulation is to replicate the
proposed case study's real-world conditions (i.e. Lagos BRT). As a result, the ADVISOR
was precisely set by employing specific driving cycles that mimicked the Lagos route
traffic condition. This traffic situation is classified into five categories that are based on
the ratio (v/c) of traffic volume (v) to road capacity (c). These traffic conditions are free,
move, appropriate, inappropriate and critical [286], [323], [324]. Although, the Lagos
BRT BEBs prototype covered in this thesis operates on a designated fixed-route known
as Transit-Way. The Transit-Way separates bus lanes from other traffic with concrete
walls, as illustrated in Figure 4-2; nonetheless, these lanes are not absolute. For instance,
there are instances when buses connect with other buses on the opposite side of the road.
As a result, the speed profile is not always consistent throughout the journeys. In order to
evaluate the impact of the traffic conditions on the BEB's energy consumption figure, the
speed profiles for different Levels of service LoS (i.e., (v/c) values- The vehicle road
capacity is the maximum number of vehicles that can go through a particular point on
the road for a specific period under the prevailing roadway, traffic, and control
conditions [325]). Quality of service involves quantitative procedures to characterise
operational requirements within a traffic stream. Level of service(LoS) is a quality
measure describing operating conditions within a traffic stream, usually in terms of such
service measures as speed and travel time, freedom to manoeuvre, traffic interruptions,

and comfort and convenience [325]) are considered. The five-speed profiles classification
for Lagos BRT's route are free, move, appropriate, inappropriate and critical
conditions that correspond to <0.6, <0.9, <1.1, <3.1 and >3.1, LoS respectively. The
Advisor database lists driving cycles that can be adopted for various simulation purposes.
The driving cycles selected for this work are classified based on the average speed. The
corresponding average speed used for free, move, appropriate, inappropriate, and critical
traffic conditions are 33km/h (Figure 4-6), 21 km/h (Figure 4-7), 18km/h (Figure 4-8),
12km/h (Figure 4-9), and 7km/h (Figure 4-10). As shown in the Figures, the average
speed of the BEB's driving cycles decreases with the increasing traffic congestion. It is
worth noting that the corresponding average speed used for this simulation is based on
the estimation that is focused only on Lagos, Nigeria, traffic scenarios [326]–[331].
Figure 4-6: Free traffic (average speed: 33km/h)

Figure 4-7 Move traffic (average speed: 21km/h)
Figure 4-8: Appropriate traffic (average speed: 18km/h)

Figure 4-9: Inappropriate traffic (average speed: 12km/h)
Figure 4-10 Critical traffic (average speed: 7km/h)
4.3.3.4 Other input parameters

The heating, ventilation, air conditioning (HVAC) and route topography are two more
input characteristics taken into account to improve the simulation's accuracy. The rated
power of the HVAC system and auxiliary systems (i.e., light, sound, and mechanical
systems) is assumed to be 15kW [10], which is the worst-case scenario imaginable given

the case study's average ambient temperature (> 20 degrees Celsius). The value is inputted
into ADVISOR by modifying the default value of Elect_Aux_Loads.
Figure 4-11 depicts the route topography for this case study (i.e., the Lagos BRT route
topography). The accuracy of the BEB energy consumption figure is improved by
mimicking route topographies. As a result, this is entered into the ADVISOR by changing
cyc_elevation_init table.
Figure 4-11 The route elevation profile [332]
4.3.4 Simulation results

Individual driving cycles used in this simulation are shorter than the case study transit
route to simulate the case study route length. Several driving cycles are simulated in
succession. The simulated number of cycles is determined by the multiple selected
driving cycles that are equivalent to the proposed length of the case study considered
(40km).
The ADVISOR simulations' primary outputs are displays of the velocity profile, State
of Charge, emission, and energy consumption as a function of time. The top plots (i.e.
Label-I) show the input drive cycle versus the BEB's real simulated velocity. The second
(i.e. Label-II) graphs show the energy storage system's State of Charge. The third graph
(i.e. Label-III) depicts the BEB's GHG emissions, while the fourth one (i.e. Label-
IV) depicts its energy use. These profiles accurately predict how the BEB will perform
on the specified course. The BEB's energy consumption is calculated using this
simulation. The output represents the energy consumed between the battery pack and the
vehicle driving train. As a result, this approach is referred to as BEB Electricity-Tank-to-

Wheel, and the details of the five scenarios simulated result is given Table 4-2. As earlier
presented, this table compare the BEB energy consumption simulation result for the five
categories of studied traffic situations (i.e., free (Figure 4-6), move (Figure 4-7),
appropriate (Figure 4-8), inappropriate (Figure 4-9) and critical (Figure 4-10).
Figure 4-12 Result (Free traffic)
The Figure 4-12 is the simulation result for the free traffic scenario. The figure is labels
(I)-(IV); (I) is the speed profile, (II) shows the BEB state of charge (SoC), (III) shows the
BEBs emission profile, and (IV) is the BEBs power consumption profile. The power
consumption profile is the BEB battery's power to drive the BEB and supply the auxiliary
loads. In contrast, the negative power represents the recovered power during deceleration.
The kWh/km results presented in Table 4-2 represent the average BEB power
consumption simulated for the 40 km route considered as the case study.

Table 4-2 Average BEBs Energy Consumption Based on Traffic Flow classification
Parameters Traffic condition

Free move Appropriate Inappropriate Critical
v/c <0.6 <0.9 <1.1 <3.1 >3.1
kWh/km 1.05 1.54 1.88 2.22 2.95
Period(hr) 23-5 10-15 5-7 9-10 8-9
22-23 21-22 16-17 17-21
Table 4-2 shows Lagos traffic characteristics (free, move, appropriate, inappropriate
and critical) and the average energy consumption rate for the studied transit system under
different traffic conditions. Auxiliary power is regarded to be a constant value. The model
does not attempt to account for the uncertainty associated with weather conditions. The
energy usage figures in Table 4-2 are an average for the driving cycle. The average energy
consumption of the transit system is required from the transit operator's assessment in
order to standardise their BEBs route energy demand. As indicated in Table 4-2, the
average energy usage under free traffic conditions (the BEB is likely to make a few stops)
is 1.05kWh/km. The critical traffic scenario consumes the most energy at 2.95kWh/km.
The BEBs' energy usage increases when the route traffic situation worsens, as the BEBs
take longer to traverse each kilometre.
4.3.5 Discussion and Summary

The simulated energy consumption analysis reflects the impact of roadway level of
service on the average traffic speed to simulate different real-world traffic conditions.
Five different traffic scenarios have been presented to quantitatively assess the parameters
that affect the energy consumption of BEBs in different real-world traffic. The case
studies indicate that different traffic scenarios (i.e., free, move, appropriate,
inappropriate and critical traffic) might have different energy consumption figures due
to the BEB speed and trip time variation, and these have been generated to be the
reference of conditions on the Lagos BRT. The numerical studies also reveal that traffic
conditions could affect the energy consumption figures by increasing the BEB energy
consumption by about 180% of its value in critical traffic situations due to lower speed
values and longer trip duration at high traffic conditions. The evaluated BEB energy
consumption figure is part of the input parameter for the developed optimisation model

for the allocation of charging infrastructures for the BEB transit system. Hence, the next
section presents the detail of the simulation BEB Transit Charging Infrastructures.
4.4 Simulation BEB Transit Charging Infrastructures

The previous section presents the procedure for evaluating the BEB transit fleet Energy
Consumption figure. The BEBs energy consumption figure is one of the input parameters
required for the optimum configuration of batteries and chargers to meet transit schedule
requirements.
The mathematical and optimisation formulations for the novel BEB charging system
model, multi-terminal charging model, integrated charging model, and BEB transit
energy demand estimation are described in the third chapter. A mathematical model is
also developed in Chapter 3 to allocate multi-terminal inductive charging infrastructure
in order to optimise power transmitter placement, inductive cable length, battery capacity,
and the cost of an electrified transit system.
Based on these mathematical models in chapter 3, in this section, simulation tools are
developed using a PSO algorithm to optimally determine the numbers, locations, power
capacity, and costs while considering alternative scenarios and charging strategies, as
well as the trade-offs among these output parameters in the designing of charging
infrastructures for the BEB transit system.
The PSO approach is simple to implement and can rapidly converge to a good solution.
It uses only primitive mathematical operators and does not require any gradient
information about the function to be optimised. It is faster, cheaper, and more efficient
than other optimisation methods. Furthermore, it requires few initial parameters and is
highly suited to solving non-linear, non-convex, continuous, discrete, integer variable
problems [333]–[335]. The optimisation equations in this work are not linear and non-
differential, and the case studies studied have limited data. As a result, the PSO algorithm
is the ultimate pick for solving this optimisation problem. The appendix contains
information on the PSO approach, including a brief introduction and PSO fundamental
modelling ideas.

4.4.1 PSO Modeling

The optimisation problems formulated in this thesis are nonlinear programming
problems with the objective of:
Solve for 𝑥⃗ which optimises 𝑓( 𝑥⃗ )
Subject to:
𝒈𝒊 ( 𝒙
⃗⃗ ) ≤ 𝟎, 𝒊 = 𝟏, 𝟐, … . . , 𝒎 (4.8)
𝒉𝒊 ( 𝒙
⃗⃗ ) = 𝟎, 𝒊 = 𝟏, 𝟐, … . . , 𝒑 (4.9)
𝑙𝑏𝑖 ≤ 𝑥𝑖 ≤ 𝑢𝑏𝑖 , 𝑖 = 1,2, … . . 𝑛 (4.10)
Where 𝑥⃗ is the denotes the solutions vector 𝑥⃗ = [𝑥1 , 𝑥2 , … … . . , 𝑥𝑛 ] 𝑇 , m is the total

number of inequality constraints, and p is the total number of equality constraints (in both
cases, constraints could be linear or nonlinear). In the relevant literature on evolutionary
optimisation, it is usual to convert equality constraints to inequalities of the form:
|𝒉𝒊 ( 𝒙
⃗⃗ )|−∈ ≤ 𝟎 (4.11)
Where ϵ is the permissible tolerance (a very small value), this enables us to focus
exclusively on inequality constraints. A similar transformation is achievable when only
equality constraints are involved. However, this type of modification is uncommon since,
when utilising evolutionary algorithms, it is typically easier to deal with inequality
constraints (for example, using exterior penalty functions [15]) than it is with equality
constraints.

Figure 4-13 Pseudocode of the PSO algorithm
Assuming F denotes the feasible area and S denotes the whole search space, it should
be obvious that F⊆S. For an inequality constraint that satisfies 𝑔𝑖 ( 𝑥⃗ ) = 0, hence 𝐹 is
active at 𝑥⃗ . At all points of F, all equality constraints ℎ𝑖 (independent of the value of 𝑥⃗
used) are considered active.
4.4.1.1 The modelling approach

The PSO algorithm used in this research is depicted in Figure 4-13. The technique is
essentially a straightforward PSO implementation, with three notable exceptions: the way
velocity is determined, the turbulence operator and the mechanism used to handle
restrictions. These points are covered in considerable detail in the following subsections.

4.4.1.2 Calculating Velocity

The expression proposed by the authors in [336] is used to determine the velocity of a
particle:
𝑽𝒊𝒅 = 𝝎 × 𝑽𝒊𝒅 + 𝒄𝟏 × 𝒓𝒂𝒏𝒅𝟏 () × (𝒑𝒃𝒆𝒔𝒕,𝒊𝒅 − 𝒙𝒊𝒅 ) + 𝒄𝟐

(4.12)
× 𝒓𝒂𝒏𝒅𝟐 () × (𝒈𝒃𝒆𝒔𝒕,𝒊𝒅 − 𝒙𝒊𝒅 )
where 𝑉𝑖𝑑 is the velocity of the id dimension, 𝑐1 and 𝑐2 are two values randomly
generated in the range [1.5,2.5] (this range are empirically
determined), 𝑟𝑎𝑛𝑑1 () and 𝑟𝑎𝑛𝑑2 () refer to functions that return a random value in the
range [0.0, 1.0], is the inertia weight, which in this case takes a randomly generated value
in the range [0.1, 0.5] (this range was determined empirically), pbest is the best position
of the current particle found so far, and gbest is the best position of the best particle found
so far.
4.4.1.3 The Turbulence Operator

Turbulence is defined as a change in the travel velocity of a particle. This alteration
occurs across all dimensions (i.e., across all choice variables), allowing the particle to
migrate to an entirely isolated location (something much more difficult to achieve by the
mere use of the velocity adjustment formula described above). This method perturbs the
swarm in order to prevent particles from becoming trapped in local optima. The
turbulence operator behaves probabilistically, considering the current generation and the
overall number of iterations required. The objective is to have a substantially higher
possibility of perturbing the particle's trajectory at the start of the search. This probability
will decline over time as we advance in the search.
Turbulence can be viewed as a mutation operator, and it is defined as follows:
𝒕𝒆𝒎𝒑 = 𝒄𝒖𝒓𝒓𝒆𝒏𝒕𝒈𝒆𝒏𝒆𝒓𝒂𝒕𝒊𝒐𝒏𝒔 /𝒕𝒐𝒕𝒂𝒍𝒈𝒆𝒏𝒆𝒓𝒂𝒕𝒊𝒐𝒏𝒔 (4.13)
𝑝𝑟𝑜𝑏𝑡𝑢𝑟𝑏𝑢𝑙𝑒𝑛𝑐𝑒 = 𝑡𝑒𝑚𝑝1.7 − 2.0𝑥(𝑡𝑒𝑚𝑝) + 1.0 (4.14)
where 𝑡𝑒𝑚𝑝 is used as a temporary variable, 𝑐𝑢𝑟𝑟𝑒𝑛𝑡𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑠 is the current generation

number, 𝑡𝑜𝑡𝑎𝑙𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑠 denotes the total number of generations
and 𝑝𝑟𝑜𝑏𝑡𝑢𝑟𝑏𝑢𝑙𝑒𝑛𝑐𝑒 denotes the likelihood of employing the turbulence operator to

impact the trajectory of a particle. The values for this expression were determined
empirically through a series of studies.
4.4.1.4 Handling Constraints

The technique proposed in this thesis to handle constraints is applied when selecting a
leader [337], [338]. The fitness function is modified slightly so that the particle with the
highest fitness value wins when two viable particles are compared. If one of the particles
is deemed infeasible and the other is deemed feasible, the feasible particle becomes the
winner. If both particles under consideration are infeasible, the particle with the lowest
overall violation of constraints (normalised by the maximum violation of each constraint
attained by any particle in the present population) prevails. The objective is to choose the
particle closest to the feasible region as a leader, even if it is infeasible.
Figure 4-14 Graphical representation explains the constraint-handling mechanism

incorporated into our PSO algorithm (Adapted from [339]).
To better understand this concept, consider three particles and two constraints scenario
presented in [339]: particle one breaches the first constraint in 30 units and the second in
40 units. Particle 2 does not violate the first constraint but does violate the second
constraint by 100 units. Finally, particle 3 breaches the first constraint by 130 units, but
not the second. Additionally, the largest violation of the first restriction is 200, whereas
the largest violation of the second constraint is 120. Thus, particle 1's fitness is
30/200+40/120=0.48333. Particle 2 has a fitness of 0+100/120=0.83333. Particle 3 has a

fitness of 130/200+0=0.65000. Thus, particle 1 has higher fitness than particle 2 and
particle 3 (remember that a smaller value indicates that the particle is closer to the feasible
region in this case), even though this particle violated both of the problem's constraints.
In contrast, the other two particles violated only one of them. Figure 4-14 illustrates this
behaviour graphically.
4.4.2 Transit fleet electrification simulation studies

Chapter three presents the design and the mathematical modelling that includes the
optimisation model for the BEB transit fleet. The modelling considers diverse BEB’s
charging configurations, including studying the use of both the conductive and inductive
charging infrastructure and the design of the novel alternative multi-terminal priority
charging model. Each of these modelling attributes is used as a planning objective for the
BEB transit system optimisation problem.
In this section, each of these modelling objections is demonstrated as a BEB charging

configuration scenario. These studies shows the trade-off between the cost, location of
the charging infrastructures, the charging duration, on-board battery size, and the transit
operation schedule. The first scenario illustrates the trade-off between the chargers' three
possible capacities (i.e. 50kW, 200kW, and 400kW) and the optimised battery sizes and
the cost.
The second scenario is the novel priority charging strategy that considers allocating
the high-priority, medium priority and low priority chargers at each BEB terminal based
on the state of charge on arrival. The detail of this strategy is illustrated in the flow
diagram in Figure 3-6. In this work, the high priority chargers are configured to have a
maximum capacity of 400kW. The medium priority chargers are considered to have a
maximum capacity of 200kW, and the low priority chargers are configured to have a
maximum capacity of 50kW. These different capacities chargers are considered because
of the nature of the schedule in a bus transit system, i.e. buses are usually scheduled to
different routes, resulting in different energy demand and charging needs. As a result, to
manage the impact of of the BEB charging demand on the grid while maintaining the
operational bus schedule, the use of different capacity charges based on the charging
priority need of the bus is to aggregate the charging demand over daily operational hours

and to avoid unnecessary peak that can impact the grid while ensuring that the transit
operational schedule is met.
The third scenario is the novel priority charging strategy that integrates both the
conductive and inductive chargers. The allocation of the conductive chargers at the bus
terminals is based on the charging strategy described in the second scenario. The addition
of strategic placement of inductive chargers to the system illustrates the trade-off in cost,
battery size, and infrastructure need for the transit network. These inductive charges are
considered located near major bus stops. Because the vehicles slow down as they
approach the stations and then eventually stop to offload and load passengers - these allow
a short but sufficient time to 'top-up' the BEB battery, and several buses can be charged
simultaneously.
The case study considered is the Lagos BRT system, as described in section 4.2. The
first part of this study evaluates the BEB average energy consumption value, already
presented in section 4.4. This BEB energy consumption figure is part of the input
parameter for evaluating the BEB charger configuration. Each of these BEB charging
configuration scenarios is a model with the PSO algorithm (the detail of the PSO
algorithm is presented in section 4.5).
4.4.2.1 BEB Transit data and model input parameters

As presented in section 4.2, the BRT route is about 40km long, running from Ikorodu
to TBS. The Lagos BRT currently has 44 bus shelters in both directions, six terminals, 19
intersections and two bus depots [4] [5]. Primero Transport Services Limited (PTSL), a
private company that took over the transit operations in November 2015, manages the
Lagos BRT scheme. PTSL has added 434 buses to the transit operations from Ikorodu to
TBS and has shuttled over 101 million commuters from November 2015 to date. It
operates 16 hours every day from 6:00 am to 10:00 pm [4].
Figure 4-15 Lagos transit network showing bus terminals (Source: LAMATA)

Table 4-3 presents the details of the service lines with the average cycle time. The route
has 29 bus stops and six terminals. The terminals' locations are Ikorodu, Agric, Mile 12,
Moshaladhi, CMS and TBS, respectively. This data, combined with the cycle time, the
BEB energy consumption figure and the bus transit schedule data, are used as input
parameters for the optimisation model. The multi-terminal characteristic of the Lagos
BRT is shown in Figure 4-15. From the figure, the total distance of the transit network is
about 40km.
Table 4-3 Route distance and runtime (Source: LAMATA)
Route Distance and Runtime
Starting point End point Length Km Runtime
IKORODU ARUNA 1 0:04:00
ARUNA AGRIC 1 0:01:30
AGRIC OWUTU IDIROKO 1 0:03:00
OWUTU IDIROKO OGOLONTO 2 0:03:00
OGOLONTO MAJIDUN 1 0:02:30
MAJIDUN AJEGUNLE 3 0:08:00
AJEGUNLE IRAWO 1 0:03:00
IRAWO1 IDERA 1 0:01:40
IDERA OWODEO 1 0:02:00
OWODEO MILE12 1 0:02:30
MILE12 KETU 3 0:05:00
KETU OJOTA 1 0:03:11
OJOTA NEW GARAGE 0.5 0:02:16
NEW GARAGE MARYLAND 1 0:02:20
MARYLAND IDIROKO 1 0:01:50

IDIROKO ANTHONY 1 0:02:25
ANTHONY OBANIKORO 1 0:02:30
OBANIKORO PALMGROOVE 1 0:01:18
PALMGROOVE ONIPANU 1 0:02:15
ONIPANU FADEYI 1 0:03:42
FADEYI MOSHALASI 0.9 0:01:00
FADEYI BARRACKS 2 0:05:00
MOSHALASI BARRACKS 1.1 0:03:00
BARRACKS STADIUM 1 0:01:30
STADIUM IPONRI 2 0:03:00
IPONRI COSTAIN 0.4 0:01:00
COSTAIN LEVENTIS 4 0:05:30
LEVENTIS CMS 1 0:04:00
CMS TBS 2 0:04:00
Total Travel Time & Length 38 1:25:57
Number of Terminals: 6
Number of Depot: 2
Average Speed: 45Km/h
Average Passenger volume: 116,000 per day
Other input parameters for the case study considered in this work are shown in
Table 4-3. Detailed data of the Lagos BRT transit data is provided in Appendix A.

Table 4-4 Modelling and simulation parameter (Adapted from [245], [340], [100])
Input Parameters Value
Number of BEBs 400

Cost of battery per kWh (£) 250
Cost of inductive Cable per meter (£) 50
Cost of BEB (£) 100,000
Cost of wireless power receiver (£) 2,000
Cost of wireless power transmitter per kWh (£) 200
Transmitter capacity (kW) 500
Power of conductive chargers (low, medium & high priority)
50, 200, 400
(kW)
Cost of charger installation (low, medium & high priority) (£) 5,000, 15,000 & 30,000
fixed cost (cfl , cfm , and cfh ) charged for using either low, medium
£0.0, £1.0 & £2.0
and high priority, respectively
Charging cost (c) is the cost of energy per kWh (£) 0.2
Cost of charger installation of inductive transmitter(£) 30,000
It is also worth noting that all the inductive chargers are configured to have the same
capacity of 500kW. The inductive chargers' location and the length of the inductive cables
are to be obtained from the optimisation result. However, the approach considered in this
work gives priority for the inductive chargers to be located near major bus stops where
possible. Because the vehicles slow down as they approach these stops and then
eventually stop to offload and load passengers - these allow sufficient time to 'top-up' the
BEB battery, and several buses can be charged simultaneously. The inductive chargers
are considered necessary because of the following reasons:
As shown in Figure 4-15, the Lagos BRT network has six terminals, with each terminal
in the modelled case will have charging infrastructures that BEB can use at the start and
the end of a scheduled trip. The chargers at the terminals are designed with various
capacities, and the capacities are allocated based on the BEB energy demand and the
power capacity of the charging infrastructures. The BEBs will be subject to the high,
medium and low priority charging described previously. The high priority chargers in this

study are set to a maximum capacity of 400kW. Chargers with a maximum capacity of
200kW are called medium capacity, while those with a maximum capacity of 50kW are
considered low capacity.
4.4.2.2 BEB energy consumption

The Lagos BRT network's current bus specification is 12m with up to 80 total
passengers with up to 40 seats [341]. While there is some generational variation in the
existing bus fleet for the Lagos BRT, with different buses being purchased at different
times, in this work, all the BEB are assumed to be of the same size and capacity. This is
largely for ease of modelling characterisation. In this work, the equivalent electric drive
considered is the BYD ADL Enviro200EV pure electric, zero emission single deck bus
from BYD and Alexander Dennis [167]. This BEB energy consumption modelling
associated with the modelled vehicles have already been presented in section 4.3.4 using
the ADVISOR modelling tool. The evaluated value of 1.88kWh/km for the 'appropriate
traffic' is considered as the input parameter for the optimisation model (see Table 4-2).
4.4.3 Charging infrastructure simulation studies

The simulation research presented in this section is based on the mathematical analysis
of charging infrastructure allocation to the electric bus transit network system presented
in chapter 3, section (3.3.1). The proposed analysis is based on the design and modelling
of multi-terminal charging infrastructure to meet the charging demand of the bus transit
network system. The simulation model considered different scenarios that include fixed
charger capacities (scenario I). The second scenario is the simulation priority charging
(scenario II). Also considered is a model that integrates both inductive and conductive
priority charging (scenario III) to ensure effective energy management without disrupting
the operational bus schedule. In the fourth scenario (scenario IV)), only inductive
charging infrastructure is investigated for comparison. The inductive model determines
the best position and length for the inductive cable, as well as the lowest infrastructure
cost. The optimisation problem is made up of the objective functions and constraints
given in section (3.3.1) of chapter 3.

4.4.3.1 Scenario I – Simulation of fixed terminal charger capacity

In this case, the transit optimisation problem assumes that all six terminals have the
same charger capacity. The simulation model allocates a sufficient number of charging
infrastructure to each bus terminal in the most efficient way possible. This charging
infrastructure allocation is based on the mathematical model presented in chapter 3,
section (3.3.1) and implemented using the PSO optimisation method. The derived BEB
energy consumption figure (Table 4.2 in section 4.1), as well as transit data such as route
length, transit schedules, and cycle duration, were used as input parameters. In addition,
the estimated cost value in Table 4-4, the number of buses, and the charger's capacity are
also input parameters for the simulation model. Additionally, the model generates results
for the optimal battery size and the cost associated with different types of chargers
considered in this scenario.
In this scenario, a comparative evaluation of simulation studies of three different types

of chargers is classified based on their power capacities. The charger capacities are
400kW, 200kW and 50kW, respectively, as shown in Table 4-5. The results show that the
whole transit network system in the case studies requires about 47 chargers of 400kW
capacity, and each BEB will require an onboard battery of 96kWh. At 200kW, the transit
network needs to have 76 chargers, and the optimal onboard battery is evaluated to
126kWh. Charging with 50kW chargers requires 116 charges, and an onboard battery
capacity is estimated to be 164kWh.
Table 4-5 Simulation result for the scenario I

Variables Terminals Result Summary
1 2 3 4 5 6 Total Battery Size Optimal
(kWh) Cost (£)
No of 400kW conductive 7 6 9 10 9 6 47 96 51,070,720
chargers
chargers
chargers
In this scenario, the optimal cost for each charger capacities classification is shown in
Table 4-5, with the lowest value of £51,070,720.00 associated with the 400kW chargers.

It shows that a fast charger can benefit the cost of installation. Without the loss of
generality, it is important to indicate that fast charging can negatively impact the grid if
not properly managed and promote the BEB on-board battery's ageing [343].
4.4.3.2 Trade-off Analysis (Scenario I)

This study has shown the impact of onboard battery reduction with varied transit
charger configurations on the related costs. This first scenario presents the trade-off
between the chargers' three possible capacities (i.e. 50kW, 200kW, and 400kW) and the
optimised battery sizes and the cost, as shown in Table 4-5.
According to the findings, the entire transport network system will require
approximately 47 charges of 400kW capacity each, with each BEB requiring a 96kWh
onboard battery. In addition, when considering a 200kW charger for the transit network,
it is determined that it will require 76 chargers. The optimum onboard battery capacity of
the BEB is estimated to be 126kWh. The onboard battery capacity is estimated to be
164kWh while charging with 50kW chargers, and the transit network requires about 116
units’ charges. In this scenario, the optimal cost for each charger capacities classification
is shown in table 5.3, with the lowest value of £51,070,720.00 associated with the 400kW
chargers.
600
500
400
300
200
100
0
Battery Size (kWh) Optimal Cost (£100,000)
400kW 200kW 50kW
Figure 4-16 Scenario I, trade-off analysis
As illustrated in Figure 4-16, having many low-capacity chargers may be financially

costly and inefficient than having fewer, higher-capacity ones (as seen in the situation of
400kW and 50kW configurations). This inefficiency is caused by long charging durations

that are inconvenient for bus transit schedules. In order to fulfil transit schedules, the
number of charging facilities increases, as does the cost.
Furthermore, the greater capacity charger demonstrates that the BEB onboard battery can
be swiftly recharged. As a result, a lower on-board battery capacity is possible with high-
capacity chargers. Furthermore, because there are fewer charger units and a smaller BEB
onboard battery, the investment cost is cheaper than low-capacity chargers requiring more
charging points and BEBs with a large onboard battery. For example, this simulation
result shows that implementing 400kW is estimated to be £5,909,888.00 less expensive
than implementing 50kW in this case study.
4.4.3.3 Scenario II- Simulation of priority charging strategy

The modelling of the multiterminal BEB transit system in this scenario is based on
priority charging set up at the bus terminal. The terminal is where buses halt for 10-30
minutes at the end or beginning of a scheduled trip to allow passengers to off-board and
on-board. This waiting period is considered the charging duration to prevent disrupting
the BEB operational timetable. A transportation network typically has multiple terminals
strategically placed to adequately distribute buses to meet passenger demand. As a result,
the multiterminal priority charging approach proposed in this thesis aims to allocate
priority charging infrastructures to meet the charging demand for transit BEBs by
positioning adequate resources at each BEB terminal.
The charging infrastructure allocation of high (H), medium (M), and low (L) priority
chargers are based on the mathematical model (objective equation and constraints)
described in chapter 3, section (3.3.1.2) and is implemented using the PSO optimisation
method. The input parameters include the derived BEB energy consumption figure (Table
4.2 in section 4.1) and bus transit data such as route length, transit timetables, and cycle
duration. Also, the estimated cost value in Table 4-4, the number of buses, and the
charger's capacity are part of the input parameters for the simulation model. Moreover,
the model gives results for the optimal battery size and the optimal cost.
This allocation of high-priority, medium-priority, and low-priority chargers at each

BEB terminal is based on charge condition on arrival. The flow diagram Figure 3-6
depicts the strategy in full depth. The high-priority chargers in this study are set to a

maximum capacity of 400kW. Chargers with a maximum capacity of 200kW are termed
medium-capacity, while those with a capacity of 50kW are considered low-capacity.
In this scenario, the model considered the use of three different chargers capacities
strategically, as described in chapter 3 and illustrated in Figure 3-6. This strategy is
considered to efficiently manage the transit energy demand and mitigate the power grid's
transit demand impact because the power demand at every point in time depends on the
total capacities of the operating chargers. For example, if 18 BEBs arrive at the terminal
for charging at the same time, and all of the chargers are 400kW, the charger demand will
be (400kW ×18 = 7,200kW). However, utilising the proposed priority charging technique,
the total demand will be (400kW ×1 + 200kW ×6 +50kW ×11 = 2,150kW), relieving the
grid of approximately 5,050kW of immediate demand. Although a lower number of
400kW may be sufficient to meet the charging demand, BEBs will be forced to queue in
this situation. This condition was depicted in scenario-I above, so the following
subsection compares the behaviour of terminal one in scenarios-I and II.
Table 4-6 Simulation result for priority charging (Scenario II)

Terminals
Variable
1 2 3 4 5 6
H 1 1 3 3 3 1
No of conductive chargers M 6 6 6 8 7 8
L 11 9 9 9 9 9
Total no. of terminal chargers 18 16 18 20 19 18
Optimal Cost (£) 47,338,900.00
Optimal battery capacity (kWh) 64
The optimisation model generates the optimal cost for the bus transit infrastructure needs
to be £47,338,900.00. Table 4-6 shows the result per terminal—the optimal onboard
battery capacity in this scenario is 64kW. It is worth noting that the result charger
allocation simulation across the terminals is different because the terminal's demand is
based on the distance travelled by bus and the varying operational schedule.

4.4.3.4 Comparison of charging behaviour of Scenario I and II

This study demonstrates the charging of ten BEBs using the charger allocation result
from the optimisation algorithm. In this case study, a comparative evaluation of priority
and the fixed capacity charging of BEB is carried out at terminal one. The priority
charging of ten BEBs is compared with 50kW, 200kw, and 400kW capacity chargers.
The result is shown in Figure 4-17.
Figure 4-17 Charging duration (min) and SoC (%) of BEBs priority and non-priority
(50kW) scenarios
Figure 4-17 shows the arrival time and departure time and the BEB on-board battery state-
of-charge on arrival and the state-of-charge (SoC) on departure for both priority and fixed
capacity (50kW) scenarios. The fixed capacity of 50kW charger’s capacities specify that
none of the BEBs can receive charge to total capacity before departure within the
maximum schedule duration of 15minute (dwell time). However, some of the BEB may
be able to meet the next scheduled trip demand. For example, the 10th BEB in the Figure

4-17 could meet the demand of the next scheduled trip depending on the trip distance.
However, it could be difficult to achieve an optimal BEB schedule (like the existing diesel
buses) using the 50kW capacities chargers in most cases. In the priority scenario, seven
out of the ten BEBs charged to their full capacities while the remaining three received
energy sufficient for the next scheduled trip. These three BEBs cannot be fully charged
because they receive a charge at a medium or low priority that depends on their battery
capacity on arrival; however, the received energy is sufficient to keep the transit schedule
going without delay.
Figure 4-18 shows the arrival time and departure time, the BEB on-board battery state-
of-charge on arrival, and the state-of-charge (SoC) on departure for both priority and non-
priority (200kW) set-ups. The non-priority scenario of 200kW charger’s capacities shows
that 5 of the 10 BEBs can receive charge to total capacity before departure within the
maximum schedule duration of 15minute. However, the remaining 5 BEBs that arrived
at the charging station with relatively low SoC did not charge to the total capacity.
Although the received energy may be adequate to meet the next trip demand in some
cases, this is not always the case since some BEB with a deficient state of charge may be
unable to gain enough energy for the next scheduled trip.

(200kW) scenarios
In the priority scenario, seven out of the ten BEBs charged to their full capacities while
the remaining three received energy sufficient for the next scheduled trip. These three
BEBs cannot be fully charged because they are receiving a charge at a medium or low
priority that depends on their battery capacity on arrival. However, the received energy is
sufficient for the next transit scheduled trip. Compared to the priority charger with the
non-priority 200kW chargers, the priority configuration systems are cheaper and ensure
that all the buses are accurately planned to receive charges to meet scheduled trips without
any operational delay. In addition, a minimum number of buses can be operated, similar
to the diesel bus transit system.

(400kW) scenarios
Also, Figure 4-19 shows the arrival time and departure time and the BEB on-board
battery state-of-charge on arrival and the state-of-charge (SoC) on departure for both
priority and non-priority (400kW) scenarios. The non-priority 400kW charger’s
capacities specify that all the BEBs received a charge to total capacity before departure
within the maximum schedule duration of 15minute (dwell time). However, compared to
the novel priority-charging charging strategy proposed in this work, the use of several
high-capacity chargers simultaneously could negatively impact the grid and distribution
network infrastructure [344]. The priority charging strategy proposed in this thesis
reduces the demand from the grid while charging several BEBs simultaneously. Hence,
it potentially mitigates the valley and peak effect on the grid [344] and reduces the impact
of BEBs' charging demand on the distribution network and the grid because the peak load

could be shifted as priority-charging spreads the transit demand for the duration of the
operational bus schedule.
Moreover, the cost of providing these proposed charging infrastructures (priority

chargers) is cheaper than the needed investment cost for providing the 400kW capacities
chargers.
4.4.3.5 Scenario III-Simulation of the multi-terminal integrated

charging model
The simulation of the transit electrification system considered in this scenario is the
multiterminal allocation of charging infrastructure for a bus transit system that uses
conductive and inductive charging techniques. This simulation approach demonstrates
the cost-benefit trade-off and improves operational service reliability by optimising
charging station infrastructure: charger-operating capacities, on-board battery capacity,
the position of inductive transmitters, length of inductive cables, and the number of
chargers. This simulation is based on the mathematical model and objective equation
described in chapter 3, section (3.3.1.3)-equations (3.55) - (3.65).
The PSO optimisation method is used to implement this mathematical model. Input
parameters included the computed BEB energy consumption figure (Table 4-2 in section
4.1), as well as transit data such as route length, transit timetables, and cycle duration. In
addition, the simulation model takes into account the estimated cost value in Table 4-4,
the number of buses, and the charger's capacity. The model gives findings for the
optimum battery size, total infrastructure cost, and the number of different types of
chargers considered in this scenario.
Table 4-7 Simulation result for the integrated model (scenario III)
Terminals
Variable
1 2 3 4 5 6
H 1 2 2 3 3 1
No of conductive chargers M 6 6 6 6 6 5
L 8 10 8 9 8 10
Locations of the inductive
0 0 1 1 1 0
transmitter (500kW)
Locations of inductive cable 7900- 20425- 31630-
(m) 9039 21526 32852
Optimal Cost (£) 44,377,000.00
Optimal battery capacity
33
(kWh)

It is worth noting that inductive charges should be located near major bus stops. Because
the vehicles slow down as they approach the stop and then eventually stop to offload and
load passengers - these allow a short but sufficient time to 'top-up' the BEB battery, and
several buses can be charged simultaneously.
The optimisation model generates the optimal cost for the bus transit infrastructure
required in this scenario to be £44,377,000.00. The result per terminal is shown in Table
4-7; the ideal onboard battery capacity, in this case, is 33kWh. The optimisation result
shows that inductive chargers were only allocated to terminals 3, 4, and 5 of the transit
network because these positions are associated with 13 km, 12.5 km, and 10 km (Figure
4-15), respectively. Compared to terminals 1 and 2, where the BEB travel distance is
about 2km and can be completed without recharging the BEB on-board battery. In
addition to cost-saving, this integrated model also reduces the size of the onboard battery
of the BEB, as shown in (Table 4-7). Hence, reducing the cost and the weight of the BEBs.
As a result, it minimises BEB energy usage per km.
Figure 4-20 Optimal design Lagos BRT network electrification
The summary of the optimal design (i.e. the result) of the novel priority charging
strategy for Scenario III that integrates both the conductive and inductive chargers for the
Lagos BRT network is shown in Figure 4-20. This figure gives an overview of the
simulation result that can be implemented for this case study. Moreover, it can be inferred

that this framework demonstrates that large, expensive onboard batteries are unnecessary
and that recharge periods can be reduced. This allows BEBs to be utilised in the same
cycle as conventional diesel buses without compromising the system's effectiveness.
4.4.3.6 Scenario IV- Simulation of inductive charging
This scenario's transit electrification simulation is based on utilising inductive charging

infrastructure for a bus transit system. This simulation approach demonstrates the cost-
benefit trade-off and increases operational service reliability by optimising charging
station infrastructure: onboard battery capacity, inductive transmitter position, and
inductive cable length. The mathematical model and objective equation provided in
chapter 3, equations (3.67) - (3.74) are the frameworks for this simulation studies.
Table 4-8 Simulation result for the inductive charging model (scenario IV)
Terminals
Variable
1 2 3 4 5 6
Locations of inductive cable 0- 2120- 14120- 26120- 38020- 39120-
(m) 660 2980 15400 28381 38983 39480
Optimal Cost (£) 52,155,184.98
Optimal battery capacity
24
(kWh)
This mathematical model is implemented using the PSO optimisation method. The input
parameters include the computed BEB energy consumption figure (Table 4.2 in section
4.1) and bus transit data such as route length, transit schedules, and cycle duration.
Furthermore, the simulation model considers the estimated cost value in Table 4-4, the
number of buses, and the capacity of the charger. The model provides results for the
optimal battery size, overall infrastructure cost, and length with inductive cable locations.
The optimisation model generates the optimal cost for the bus transit infrastructure
needs to be £52,155,184.98. Table 4-8 shows the optimal locations and the length of the
inductive cable. The optimal onboard battery capacity in this scenario is 24kWh. It should
be noted that the inductive simulation in this section is intended to be used as a
comparison tool with alternative charging models that have been suggested, primarily to

examine the financial impact of each scenario. The following chapter gives a
comprehensive framework for the simulation of an inductive transit system.
4.4.3.7 Trade-off Analysis (Scenario I, II, III, &IV)

Figure 4-21 depicts the optimal battery sizes and costs in various charging scenarios
explored in this work. This chart illustrates the trade-off between the BEB's battery
configuration and the optimal cost for the proposed electrified bus transportation system.
As shown in Figure 4-21, the lowest cost is related to the smallest on-board battery BEB
except for scenario IV, which considers an inductive only system, and this cost is
associated with a longer length of the inductive wire, which could be a trade-off if the
number of buses increases.
600
500
400
300
200
100
0
Battery Size (kWh) Optimal Cost (£100,000)
Scenario I Scenario II Scenario III Scenario IV
Figure 4-21 Trade-off analysis of Scenario I, II, III & IV
In summary, having a robust charging infrastructure can lower the cost of electrifying
bus transit systems. Scenario III, in which the system integrates conductive and inductive
charging techniques, demonstrates that investing in charging infrastructure is more
advantageous than increasing on-board BEB battery capacity.
As a result, increasing the number of BEB fleets could offset the cost of charging
infrastructures because numerous BEB can use the same charging infrastructures,
particularly on-the-fly-inductive chargers. This, of course, excludes some installation fees
and any complications that may develop as a result.

4.4.4 Transit charging demand profile

The power consumption of onboard battery chargers and the available charges are
utilised to calculate the demand profile of the charging process. BEBs begin to use
chargers one by one as demand progressively rises with varying SOCs, and the aggregated
power demand curve begins with low demand. Figure 4-22 illustrates the aggregated
power demand curve from around 6 a.m. to 10 p.m. These demand profiles are calculated
based on the daily bus schedule and charging demand for the priority and non-priority
charging scenarios. Most BEBs are expected to begin their daily schedules with a charged
battery; hence, charging buses after their daily operations may be appropriate, resulting
in a demand profile that extends to roughly 11 p.m. through overnight charging.
Figure 4-22 BEB charging demand profile
Furthermore, because not all buses begin service at the same time, the transit
operational schedule data has an impact on the demand profile (see Appendix A). This
operational schedule data is utilised to calculate and determine when the BEB will need
to charge, how much charge is required, and how much charging time is available. The
total energy consumed for charging is estimated to be 142880 kWh. This estimate is based
on the amount of energy consumed per kilometre and the type of charger utilised during
each recovery time.

Priority charging may not have a substantial impact on overall charging demand.
Because it evenly distributes peak demand and minimises the impact of charging transit
buses on the grid and power distribution infrastructures.
4.4.5 Summary and discussion

This work proposed designing the multi-terminal base model of BEB's charging
infrastructure for the transit network. In this work, both inductive and conductive chargers
are considered to manage the transit network's charging demand. The charging
infrastructure allocation is based on the PSO algorithm that is used to find the minimum
allocation value per terminal.
The priority (scenario II) method for preferential charging is introduced in this work
and is compared with the non-priority (scenario I) model. The result shows that scenario
II saves cost and is a key for energy management for the BEB transit network. Combining
the conductive and inductive charger (scenario III) in this model helps keep the BEB
batteries from being deep discharge. This priority strategy also lowers the cost of having
a large, expensive conventional onboard battery, reducing the overall weight of the BEBs
and contributing to the BEBs using less energy per km. This work's priority charging
system has the potential to be developed into an efficient, affordable energy management
system that aggregates energy usage over the operational schedule time as shown in
Figure 4-22. This priority charging strategy also reduces the power infrastructure
investment because the model ensures that minimum energy is used from each terminal
at every time.
Through the investigation of Scenarios I, II, III and IV, it is found that charging BEBs
in the Nigerian city of Lagos's BRT network can be accomplished using an opportunity
charging method that does not affect the operational bus timetable. During their stay at
the BEB's terminal, all BEBs are scheduled to charge. This research focuses on
underdeveloped countries with unreliable networks that make overnight charging of large
energy storage systems difficult and sometimes impossible.
It is worth noting that preliminary studies have shown that using different charging
capacities at different terminals along the route could not meet the transit network's
demand. The transit timetable made it clear that each terminal had a different schedule

for each bus, which resulted in varying energy demand and, consequently, different
recovery times. As a result, having different chargers at different transit network terminals
will cause disruptions in the BEB transit schedule.
4.5 Chapter Summary

This chapter's numerical case studies are based on the optimisation model provided in
Chapter 3. First, a simulation is performed to evaluate the average power consumption
of BEBs given a defined traffic situation and speed profile. Then, a novel charging
optimisation tool is created to demonstrate how diverse charging techniques may be
advantageous to investigating key transit system elements such as the trade-offs between
alternative charger designs, charger locations, battery sizes, and cost.
The chapter describes how ADVISOR modelling works before simulating various traffic
scenarios (i.e., free, move, appropriate, inappropriate, and critical traffic) based on BEB
speed and trip time variation using the Lagos BRT as a case study. The results show that
the average BEB energy consumption per km ranges from 1.05 to 2.95kWh and that
traffic conditions can impact energy consumption figures up to 180 %.
The evaluated BEB energy consumption figure is part of the input parameter for the
developed optimisation model that uses the PSO algorithm. As a result, a brief overview
of the PSO algorithm is provided.
The numerical study compares several charging choices such as varied charging
capabilities, priority charging, and integrated charging. The results suggest scenario I, fast
chargers are less expensive for powering transport fleets. However, the impact of this
charger on the grid is not investigated. The numerical study of priority charging reduces
costs and optimises energy use for BEBs in the transit network. The combination of the
conductive and inductive chargers in this model helps to keep the BEB batteries from
being overdrawn and reduces costs because the BEB optimal battery size is small
compared to other proposed models.

Chapter 5: BEB Transit fleet electrification numerical Case Studies (2)-inductive charging and carbon
emissions savings
5 BEB TRANSIT FLEET

ELECTRIFICATION
NUMERICAL CASE STUDIES
(2)-INDUCTIVE CHARGING
AND CARBON EMISSIONS
SAVINGS
5.1 Introduction
This chapter continues the numerical studies presented in the previous chapter (Chapter
4). The work presented in this chapter is explicitly based on the simulation of a case study
of an inductive charging system simulation and a case study of estimating carbon
emission savings for the transit bus system. Hence, this chapter's numerical case studies
is based on the optimisation model provided in Chapter 3.

emissions savings
5.2 Case Study of inductive charging simulation

The simulation of the inductive charger allocation for the bus transit system is detailed
in this section. This inductive charging infrastructure allocation sought to optimise power
transmitter placement, inductive cable length, battery capacity, and the cost of an
electrified transit system.
Three different scenarios based on the case studies are discussed in this section. The
first scenario envisions adding an inductive charging system to the current Lagos BRT
system, with inductive chargers primarily located at the bus terminal. In the second
scenario, the optimal design of inductive chargers at six equal-distance locations along
the Lagos BRT transit network route is investigated. The third scenario looks into the
possibility of having four inductive charger locations. These three scenarios are compared
using the optimisation parameters (i.e. cost, the battery size, length and the location of the
inductive cable and power transmitters).
A state of charge analysis that depicts battery depletion and recovery rate, as well as a
sensitivity analysis that investigates the impact of the trade-off between; the BEB's on-
board battery capacity and the number of bus terminals, transmitter numbers, inductive
cable length, and inductive cable length, are also presented.
5.2.1 Introduction
As described in chapter 3 (section 3.4), the inductive charging system is an operated
on-road charging technology that charges the BEB battery while the bus moves through
the inductive cable region. The length of the inductive cable and the size of BEB’s battery
directly influence the overall performance and cost of the transit system and the number
of transmitters deployed on the road. The novel mathematical model for the allocation of
inductive charging infrastructure is established in (3.4.1) to optimise the position of power
transmitters, the length of the inductive cable, battery capacity, and the cost of an
electrified transit system. The stated optimisation model (3.66) minimises capital cost
while taking into account the transportation system's operating constraints. The simulated
system improves accuracy by adding BEB energy usage based on route factors. This
designed problem is demonstrated with a mathematical model that is used to determine

emissions savings
the best combination of crucial inductive transit system design characteristics (i.e. battery
size, length of the inductive cable and the allocation of the power transmitters)
The described particle swarm optimization (PSO) algorithm (in section 4.5) is used to
determine the optimal transit system cost, cable length, battery size, number of
transmitters, and total cost. This cost includes the BEB cost, inductive charger cost (the
receiver and the transmitter), battery cost, and the inductive cable cost, as illustrated with
the objective equation (3.66) and the associated constraints in equations (3.67-3.74).
The inductive-based simulation present in this section is based on the Lagos BRT case
study described in (4.1). As shown in Figure 4-15, the transit system has six terminals
located at various points (2km, 15km, 27.5km, 37.9km, and 40km) within the 40km route
transit system. Because the vehicles slow down as they approach the bus terminals and
then stop to offload and load commuters, these are considered the locations for inductive
charges. These allow a short but sufficient time to ‘top-up’ the BEB battery, and
several buses can be charged simultaneously.
This thesis investigates the optimal values of the optimisation parameters (i.e. battery
size, length of the inductive cable and the allocation of the power transmitters and the
cost) based on the bus terminals. In addition, the trade-off of these optimisation factors
was investigated further. Three scenarios are addressed in this study. The first scenario
considers introducing an inductive charging system to the existing Lagos BRT system,
with the bus terminal illustrated in Figure 4-15 serving as the primary location of the
inductive chargers.
Figure 5-1 Redesigned transit system (Scenario II)
The second inductive charging scenario investigates the optimal configuration of the
inductive chargers at six equal-distance locations from each other along the Lagos BRT
transit network route Figure 5-1. The third scenario investigates the possibility of having
four locations for the inductive chargers Figure 5-2. These three scenarios are compared
using the optimisation parameters (i.e. cost, the battery size, length and the location of the
inductive cable and power transmitters). Moreover, this thesis further investigates the

emissions savings
trade-off between the possible number of bus terminals (i.e. location of power
transmitters), numbers of transmitters, inductive cable length and the on-board battery
capacity of the BEB for the Lagos transit system.
Figure 5-2 Redesigned transit system (Scenario III)
5.2.2 Data and input parameters
In addition to the data in Figure 4-15, Figure 5-1, Figure 5-2 and
Table 4-3 presented. The additional input parameters considered for this case study is
shown in Table 5-1. The BRT line, which runs from Ikorodu to TBS, is approximately 40
kilometres long, as shown in section 4.2. There are 44 bus shelters in both directions, six
terminals, 19 intersections, and two bus depots on the Lagos BRT [4] [5]. Table 5.1 also
shows the details of the service lines as well as the average cycle time. There are 29 bus
stops and six terminals along the route. Ikorodu, Agric, Mile 12, Moshaladhi, CMS, and
TBS are the terminal locations. This data, along with the cycle time, BEB energy usage,
and bus transit schedules, is used as input to the optimisation model.
Table 5-1 BEB Parameter (adopted from [101])

Input Parameters Value
Number of BEBs 100
Cost of battery per kWh (£) 125
Cost of inductive Cable per meter (£) 20
Length of the transit route (m) 40,000
Cost of BEB (£) 50,000
Cost of wireless power receiver (£) 1,000
Cost of wireless power transmitter per kWh (£) 100

emissions savings
Transmitter capacity (kW) 500
5.2.3 Inductive charging simulation studies

The optimal cost of the transit system, cable length, battery size, the number of
transmitters, and total cost are all determined using a particle swarm optimization (PSO)
algorithm. The total cost includes the BEB cost, inductive charger cost (the receiver and
the transmitter), battery cost, and inductive cable cost. The objective equation described
in (3.66) reflects the details. The PSO algorithm operated in this case study using 200
iterations and 1000 particles; the key optimisation results for the three scenarios
investigated are listed in Table 5-2. The simulations converge in about 15seconds, and
the simulation results obtained from the PSO implementation for allocating the inductive
chargers for the Lagos BRT electrification model are shown in Table 5-2.
The simulation result for the scenario 1 – inductive charging at terminals (Figure 4-15
Lagos transit network showing bus terminals (Source: LAMATA)) gives an optimal
solution that indicates that the total cost is £5,619,550 and the optimal battery size is
26kWh. Scenario 2 – inductive charging at equidistant locations on the route considered
is when each inductive charger proposed for the Lagos BRT transit system is at an equal
distance Figure 5-1. The results suggest that a smaller onboard battery of 16kWh capacity
should be sufficient for each BEB running on the proposed inductive transit route due to
the equidistant location of the inductive transmitters, resulting in a shorter inductive cable
and hence a lower total cost in comparison to the first scenario.
In the third scenario (Figure 5-2), the optimal cost is the lowest as a result of using four
transmitters at equal distances from each other. Hence, some savings on the transmitters
cost in comparison to scenarios II & I. This optimisation result is shown in Table 5-2 and
is aimed to support the transit operator in considering the various trade-offs between the
parameters when investigating the electrification of transit routes—in addition, using the
model to determine the appropriate position and length of inductive cable required for the
specific transit route.
Table 5-2 Result of allocation of inductive simulation

Variable Scenarios
I II III

emissions savings
Total cost (£) 5,619,550 5,501,450 5,492,200

Battery Size (kWh) 26 16 22
Number of Power Transmitter 6 6 4
Total length of cable (m) 4034 3691 2140
Position 1 of Transmitter with cable 0-340 0-220 0-221
(m)
(m)
(m)
(m)
Position 5 of Transmitter with cable 37882-38720 31982-32700
(m)
Position 6 of Transmitter with cable 39282-40120 39481-40080
(m)
5.2.3.1 State of charge analysis of electrified BRT inductive model

Figure 5-3, Figure 5-4 and Figure 5-5 depict the SoC of the on-board battery of the
BEB along the Lagos BRT transit route using the inductive charging model. It should be
noted that the onboard battery SOC is calculated using the linear model proposed in [345],
[346] and [217]. This state of charge analysis is based on the battery depletion and
recovery rate as it travels along the transit route. The battery's state of health is not
considered in the SOC calculation, as proposed in [317], because for this initial design
assessment, the BEBs would have new batteries. Figure 5-3 (scenario I) indicates that the
BEB on-board battery is fully charged at the start of the scheduled trip and the SoC drops
to 65% before recovery to 80%. Also, a full recovery of the BEB onboard battery SoC
occurs at 4717 seconds toward the end of the trip. This reduces to 66% before another
full recovery occurs as a result of another inductive transmitter located about 2km from
the 5th transmitter. Therefore, it is observed that the 1st and 6th transmitter’s location are
not optimal for this bus. Thus, it is important to investigate how to maximize the use of
the inductive charger along the transit route across the fleet.

emissions savings
Figure 5-3 Scenario I- State of charge analysis
The inductive power transmitters are set at an equal distance from one another in the
second scenario, which takes into account six different locations. Because the battery SoC
changes between the set maximum SoC of 80 % and the set minimum SoC of 20 %, as
illustrated in Figure 5-3, the locations maximise the BEB on-board battery capacity
compared to the scenario I. Hence led to the consideration of a smaller onboard battery
and reduced cost.
Figure 5-4 Scenario II- State of charge analysis

emissions savings
In the third scenario, the location of the inductive chargers is at four points along the
transit route, and the system optimally maximises the use of the onboard battery; as shown
in Figure 5-4, the battery SoC changes from 80% to 20%. Hence a small onboard battery
reduces cost, as shown in Table 5-2.
Figure 5-5 Scenario III- State of charge analysis
5.2.3.2 Sensitivity analysis of inductive charging case study

The results of the different inductive charging scenarios show the trade-off between
the available number of bus terminals (i.e. location of power transmitters), numbers of
transmitters, inductive cable length and the on-board battery capacity of the BEB for the
Lagos transit system. To further investigate the impact of the trade-off between the BEB's
onboard battery capacity and the number of bus terminals, transmitter numbers, inductive
cable length, and inductive cable length, a sensitivity analysis is presented in this section
using the developed optimisation model.
As demonstrated in the three scenarios considered in this simulation work, the varied
equidistance locations of the inductive charger can significantly impact the cost of
infrastructure. Hence, sensitivity analysis is considered to investigate different locations

emissions savings
for allocating power transmitters and inductive cables within the transit network. This
study varied the numbers of the power transmitter within the Lagos bus transit network
between 3 and 9, as shown in Figure 5-6 and Table 5-3. As shown in Table 5-3, the
number of bus terminals is usually the same as the number of the power transmitter except
for the circumstance where three transmitters are considered because the transmitter's
capacity is proportional to the length of the inductive cable. The total length of the
inductive cable for the four terminals transit is 2140m. The three-terminal configuration
needs an inductive cable length of 2141m; hence a terminal location requires two power
transmitters of 500kW (making four power transmitters due to the constraints imposed
on the inductive cable length).
Figure 5-6 Sensitivity analysis at different number of terminals
In considering the varied location of power transmitters (Bus Terminals), as shown in

Figure 5-7. The three-terminal configuration requires a sizeable onboard battery capacity
of 30kWh compared to other configurations in the transit system simulation. Because
three-terminal configurations require that each transit BEB have a huge battery capacity,
increasing cost, the cost keeps increasing proportionally with an increase in the number
of BEBs. The four-terminal configurations show that an optimum transit electrification

emissions savings
cost can be achieved with a 22kWh onboard battery, an inductive cable length of 2140m,
and four power transmitters. As shown in Figure 5-6, the cost keeps increasing
progressively on increasing the terminal configuration to 5, 6, 7 and 8. Using nine
terminals configuration led to decreased cost because the design requires a small on-board
battery capacity (i.e. 12kWh) BEBs for the transit operation.
Figure 5-7 Alternative configurations of inductive chargers for Lagos BRT network
This sensitivity analysis result can help the transit company consider the trade-offs
between the parameters and cost and decide based on their need.
Table 5-3 Sensitivity analysis of the multi-terminal configuration of inductive charging

transit design
Terminals
Variable
9 8 7 6 5 4 3
Total Cost (£) 5,603,600 5,649,600 5,552,500 5,501,450 5,498,900 5,492,200 5,628,196
Battery Size
12 18 16 16 17 22 30
(kWh)
Number of
Power 9 8 7 6 5 4 4
Transmitter
Total length of
3167 2912 3288 3691 2855 2140 2141
Cable (m)
Position 1 of
Power
0-190 0-220 0-221 0-220 0-340 0-221 0-460
Transmitter
with cable
Position 2 of 9980- 13280- 19980-
4980-5580 5690-6280 6680-7280 7980-8700
Power 10580 13760 20940

emissions savings
Transmitter
with cable
Position 3 of
Power 9985- 11402- 13280- 15986- 19980- 26580- 39360-
Transmitter 10580 11980 14000 16700 20580 27180 40080
with cable
Position 4 of
Power 14980- 17085- 20007- 23980- 29984- 39360-
Transmitter 15580 17680 20580 24700 30580 40080
with cable
Position 5 of
Power 19980- 22786- 26702- 31982- 39360-
Transmitter 20579 23260 27280 32700 40080
with cable
Position 6 of
Power 24988- 28604- 33380- 39481-
Transmitter 25570 29060 33978 40080
with cable
Position 7 of
Power 30046- 34282- 39495-
Transmitter 30580 34760 40080
with cable
Position 8 of
Power 35063- 39502-
Transmitter 35580 40080
with cable
Position 9 of
Power 39480-
Transmitter 40080
with cable
Each column in Table 5-3 presents the result for each configuration investigated. The
optimal battery capacity follows the optimal cost, the number of power transmitters, and
then inductive cable length and position.

The work presented in this section considers the design of a multi-terminal inductive
base model of BEB's charging infrastructure for the transit network in Lagos, Nigeria.
The solution is obtained using a PSO algorithm that seeks to find the minimum allocation
of charging resources to meet the requisite route demands. The optimisation model
developed for this research makes a significant contribution by optimally allocating
inductive chargers and forms the basis of a practical approach in the design of inductive

emissions savings
charging system for the bus transit network; this is in the provision of inductive charging
infrastructure that allocates battery sizes, location of the transmitter, and the inductive
cable's length.
When comparing the original location, Figure 4-15, with the bus terminal's equal
distance locations (Figure 5-1), the result shows that more savings can be achieved if the
inductive chargers are located equidistant from each other. The option in Figure 5-2
(scenario III) shows that having four terminals in the Lagos transit system can give an
optimal trade-off between the bus charging terminal and the onboard battery's size.
The inductive charger's location at the bus at the significant stops maximises the use of
the inductive charging infrastructure. Positioning the charging infrastructure at an equal
distance from each other, as shown in the example of Lagos infrastructure, reduces cost.
In this case, the battery size required is smaller. This study's allocation of inductive
charging infrastructure is a cost-cutting method that saves money in the short and long
run.
5.3 A case study of carbon emissions savings of electrified BRT

system
In section (4.2), the BEB energy consumption simulation result for the five categories
of studied traffic situations (i.e., free, move, appropriate, inappropriate and critical). The
power consumption profile represents the power supplied by the BEB battery to drive the
BEB and supply the auxiliary loads. This simulation produced values for buses' energy
consumption when powered by electricity. In this section, an additional simulation is
performed to estimate the energy consumption of the diesel equivalent of the BEB bus in
five categories of studied traffic situations. The findings are utilised to compute the BEB
buses' emission savings compared to the diesel equivalent. It is worth noting that these
energy consumption estimates represent the energy consumed between the diesel-fuel
tank/battery pack and the vehicle drive train (i.e. tank-to-wheel). Table 5-4 shows the
tank-to-wheel electricity and diesel consumption in various traffic scenarios.

emissions savings
Table 5-4 Comparison of Diesel bus and BEB GHG emission (gCO2/km)
Parameters Traffic condition
Free move Appro Inappro Critical
kWh/km of electricity 1.05 1.54 1.88 2.22 2.95
Litre/km of diesel 0.17 0.25 0.30 0.35 0.47
BEB Emission (gco2/km) 269 394 481 568 755
Diesel bus emission (gco2/km) 459 675 810 945 1269
BEB Vs Diesel bus GHG Saving (gco2/km) 190 281 329 377 514
The diesel engines produce approximately 2.7kg of CO2 emission per litre of diesel fuel
consumed [348] [349]. Using the UK government 2019 figure for carbon emission per
kWh of electricity on the average energy fuel mix is 0.256kg CO2 [350],[351].
These GHG figures for diesel and electricity per km must be updated based on the
energy consumption estimates for the BEB and diesel bus values from the simulated case
study.
Hence
𝐾𝑔𝐶𝑂2 2.7𝐾𝑔𝐶𝑂2 𝐴𝐷𝑉𝐼𝑆𝑂𝑅 𝑠𝑖𝑚𝑢𝑙𝑎𝑡𝑒𝑑 𝐿

( ) = × (5.1)
𝐾𝑚 𝐷𝑖𝑒𝑠𝑒𝑙 𝐵𝑢𝑠 𝐿 𝐾𝑚
Similarly;
𝐾𝑔𝐶𝑂2 0.256𝐾𝑔𝐶𝑂2 𝐴𝐷𝑉𝐼𝑆𝑂𝑅 𝑠𝑖𝑚𝑢𝑙𝑎𝑡𝑒𝑑 𝑘𝑊ℎ

( ) = ×
𝐾𝑚 𝐵𝐸𝐵 𝑘𝑊ℎ 𝐾𝑚 (5.2)
The figure for CO2 emission comparing diesel and BEB bus at different traffic scenarios
is also given in Figure 5-8.

emissions savings
Figure 5-8 Comparison of Diesel bus and BEBs GHG emission at different Traffic
Condition
In Table 5-4 the comparison of Diesel buses and BEBs GHG emission figures at
different Traffic conditions based on the Lagos BRT route is presented. The deployment
of BEBs along this route is expected to reduce GHG emission by 190gCO2/km (≈ 41%)
for a free traffic situation, 281gCO2/km (≈ 42%) for a move traffic situation, 329gCO2/km
(≈ 41%) for an appropriate traffic situation, 377gCO2/km (≈ 40%) for inappropriate traffic
situation, and 514gCO2/km (≈ 41%) for a critical traffic situation. Also, Figure 5-8
illustrates the comparison of the Diesel bus and BEBs GHG emissions under different
Traffic Conditions.
Table 5-5 BEBs GHG emission saving figures at different Traffic Conditions
No of Traffic condition/Emission savings (tCO2)/Year
BEBs Free move Appro Inappro Critical
100 1335 1974 2311 2648 3611
200 2669 3948 4622 5296 7221
300 4004 5922 6933 7945 10832
400 5339 7895 9244 10593 14442
500 6673 9869 11555 13241 18053

emissions savings
1000 13346 19738 23110 26482 36105
A transit bus travels about 43,647miles (70,243km) in a year [352]. Hence, the
deployment of 100 BEBs in the Lagos transit route could save up to 33611 tCO2 per year
in a critical traffic situation.
Free move Appro Inappro Critical

EMISSION SAVINGS IN TCO2/YEAR
40000
35000
30000
25000
20000
15000
10000
5000
0
0 200 400 600 800 1000 1200
NO OF BEB
Figure 5-9 Comparison of (100 to 1000 BEBs) Diesel bus and BEBs GHG emission (tCO2
per year) at different Traffic Conditions
As a result, the emission savings figure over a year at various BEB adoption rates (i.e.
100 to 1000 units) is evaluated using the differences between the diesel bus emission and
BEB emission values for an average anticipated km journey in one year. Table 5-5 has
these estimated values, which are depicted in Figure 5-9 using the Lagos BRT as a case
study.

Assuming that the Lagos BRT system operates an average of 200 buses concurrently
on a daily schedule. If these 200 buses are replaced with BEBs, the impact will reduce

emissions savings
GHG emissions by 2669 tCO2 to 7221 tCO2 per year. Realistically, the Lagos BRT
network could operate 400 concurrent buses to satisfy busy daily demands. Using 400
BEBs will reduce passenger waiting time [376] and reduce GHG emissions by 5339 tCO2
to 14442 tCO2 per year. Moreover, taking into account the future development of the
BRT system in Lagos as well as the adoption of an electrified transportation system that
uses around 1000 BEBs at the same time, GHG emissions could be reduced by 13346
tCO2 to 36105 tCO2 per year.
Additionally, electrified BRT that utilises BEB can increase bus travel speed and
service while maintaining reasonable fares. The non-electric vehicle could be zone-
restricted to promote mode transfer and collect revenue to finance the electrified BRT.
These emission savings could also be accomplished by establishing emission and pricing
zones along public transportation corridors. In [2], the author estimates that it takes about
100 electric cars to accomplish the same environmental relief as can be gained from a
single 18m battery-electric bus. At current occupancy rates for cars, passengers' full
busload can take more than 40 cars off the road [353]. When the BEB is operated at a
higher travel speed, the electrified BRT system could significantly attract private vehicle
users to switch modes of transport. Some private vehicle users would switch to the
proposed electrified-BRT system because the new buses with smooth drive, low noise,
and air conditioning are significantly more comfortable than the existing diesel buses.
In developing countries, the authors in [54] analysed Malaysia's policies encouraging

public transportation usage and discovered that, in addition to journey duration, age,
gender, car ownership, travel cost, and household income all have a role in influencing
an individual's mode choice. However, the essential measures favouring public
transportation are shorter trip times between homes and public transportation stations and
fare subsidies. In general, the proportions of private car users who switch to BRT vary
significantly depending on various circumstances. In summary, apart from traffic
management, shifting from individual passenger vehicles to public transit fleet is
considered a tool to reduce the significant quantity of GHG emissions and reduce traffic
congestion in cities [354], [355],[356].
In summary, this research compares BEB energy consumption and GHG emission
savings under various traffic circumstances. This study looks at how the energy

emissions savings
consumption of electric buses varies depending on route features, speed profile, and
traffic situations. The simulated energy consumption study reflects the impact of roadway
level of service on the average traffic speed to replicate diverse real-world traffic
situations. Five different traffic scenarios were presented to examine the parameters
influencing BEB energy consumption in real-world traffic and estimate the GHG
emission savings figure.
The result indicates that the deployment of BEBs in the Lagos, Nigeria BRT route
could significantly save about 13tCO2 to 36tCO2 for each BEB deployed per year.
Generally, the impact of electrified full BRT in Lagos, Nigeria, would reduce travel time
and reduce GHG emissions.
5.4 Chapter Summary

This chapter research work investigates the use of multi-terminal inductive chargers as
BEB’s charging infrastructure for the Lagos BRT system. The result shows that
positioning the charging infrastructure at an equal distance reduces cost.
This research also evaluates the potential GHG emission savings for adopting BEB in
the Lagos BRT. The results show that deploying BEBs along the Lagos, Nigeria BRT
route might reduce CO2 emissions by 13tCO2 to 36tCO2 per BEB deployed per year.

Chapter 6: Conclusions and future work
6 CONCLUSIONS AND FUTURE

WORK
6.1 Thesis Conclusions

The growing population of cities around the world is producing traffic congestion,
increasing automotive emissions, and posing a number of health risks. For that purpose,
developed countries are reforming and increasing their investments in public
transportation networks, such as public rail, trams, subways, and public transit buses. The
growing urban population in developing countries is predicted to boost public and freight
transportation demand in the near future. On the other hand, many developing countries
lack the ability to prioritise investment in public transportation systems such as the metro
and rail systems, although they are facing similar issues. While it varies between
countries, incumbent systems in such circumstances are already challenged (e.g., frequent
power outages, ageing infrastructure). This, along with a lack of investment, means that
they will not be able to shift at the same rate as developed countries in this regard. As a
result, in the case of developing countries, less investment-intensive 'infrastructure lite'
alternatives are more likely to be accessible. On this premise, implementing the bus rapid
transit system through staged electrification of existing bus fleets is predicted to be a
particularly relevant area for decarbonising the public transportation system in these low-
income nations.
It is also projected that, if nothing changes, the transportation sector's GHG emissions
will rise to higher levels. As a result of the growth in GHG emissions, creative techniques
continue to be required to accommodate rising transportation demands while addressing
environmental issues. The implementation of electric transportation networks is expected

to be a potential solution that will assist in significantly reducing pollution and help
support the delivery of GHG targets.
In this regard, the research work of this thesis presents a novel method of charging the
transit BEB with the goal of meeting the need in developing societies where the grid
cannot readily support overnight charging of large onboard BEB batteries, as is common
in developed countries (such as Europe and China), by developing a novel charging
method that can support cheap BEB (i.e. BEB with small onboard BEB battery). The
common practice in transit systems is using the battery-electric bus (BEB) with a battery
capacity of 200-500kWh. The required charging duration for this type of BEB is usually
long, and the frequently adopted charging strategy is overnight charging. This strategy is
considered to avoid disruption of the operation schedule of the transit network, and it fits
with a reduced bus service overnight and with lower electricity system utilisation (off-
peak). However, overnight charging is described as a strategy requiring the transit system
to have many extra fleet buses or battery swapping systems that bring demands for
substantial additional investment—for example, the 40ft K9 BYD's iron-phosphate
batteries BEB cost about $600k. A single BEB has a battery capacity of 324kWh, and the
maximum range is estimated to be 250km (155miles), and it requires about 2-3hour
charging duration using an 80kW charger. The implication of using this conventional 40ft
bus in a transit network includes enormous investment, especially in a situation where
there are inadequate funds like in most developing countries.
This research work offers solutions that supplement BEB charging in a transit network
without disrupting the operational schedule and minimises the initial investment cost by
scaling the BEB battery capacity for the need of the transit system. This is distinct from
other authors' work that focuses on overnight charging and the impact of the charging on
the power system. This work also promotes the design of a specific BEB that precisely
meets the transit network's need by specifying the battery size. Furthermore, promoting
the BEB design can receive charge through inductive and conductive charging methods.
This research is also applicable to the optimal design of the charging facility that
integrates both conductive and inductive charging facilities into the transit network
system. In the power network industries, it enhances the management of the transit
network demand. It could be used as a tool by transit operators to design capability for
addressing BEB scheduling on day-ahead timescales.

In Chapter 2, an overview of the most recent battery-electric bus energy consumption

modelling and simulation was provided. The design, implementation, and optimisation of
operational schedules for various infrastructure options are examined in this literature
review. This also included a review of various methods and technologies for promoting
the electrification of public transportation networks to reduce carbon emissions and
congestion. . The literature study points out some shortcomings in predicting energy
demand for transit system modelling, such as ignoring the variation of detailed real-world
operational variables such as route traffic. The work of this thesis was guided by the
existing literature in BEB and transport modelling to purposefully address the BRT
transport problems to be considered. It builds upon this to address some of the gaps
identified in existing published research. For example, integrating five traffic flow
characteristics into the modelling activities improves the overall BRT model's
representation.
Prior attempts to design charging infrastructures for BEB fleets in a transit system have
not completely accounted for (i) a multi-terminal model based on charging infrastructure
that reduces the size of the electric vehicle's onboard battery (ii) a hybrid approach that
uses both conductive and inductive charging technologies to reduce operational downtime
induced by BEB charging. Point (i) being applicable in the developing country case as
mentioned previously and (ii) as a supporting ‘opportunistic’ addition that can provide a
trade-off between centralised and distributed charging infrastructure. As a result, and
guided by the literature, this thesis developed a priority charging model and investigated
the trade-offs between various charging methodologies.
Chapter 3 tackles some of the intricacies of the BRT system electrification modelling and
develops the mathematical model that coordinates the routing and charging for BEBs. A
mathematical optimisation approach for determining the critical design variables of
conductive and inductive charger allocation is described. This chapter provides an in-
depth look at the various modelling stages as well as the implementation infrastructure
optimisation algorithm. The design of the transit fleet routing model, the mathematical
model of BEB energy consumption, and the mathematical modelling of charging
infrastructure for BRT transit fleets are all covered in detail in this chapter. The overall

simulation modelling, arising from the different parts of the framework of the governing
model, will be presented and utilised in subsequent chapters based on the various
mathematical models and features developed here.
Chapter 4 begins to consider some of the numerical case studies in relation to the BEB
transport modelling of this thesis. Based on traffic conditions and speed profile, the
average power consumption of BEBs is determined. A novel charging optimisation tool,
developed as part of the work of this thesis, is utilised to show how different charging
strategies may favourably impact key transit system elements. This includes trade-offs
between alternative charger designs, charger locations, battery sizes, and cost. In this
thesis, a particular emphasis is placed on the developing country case, and the case study
of Lagos BRT in Nigeria is explored in-depth. Current data acquired from trustworthy
sources such as the Lagos Metropolitan Area Transport Authority (LAMATA) and
PRIMERO Transport Services Limited are used to characterise the study problems.
The following list of features provides the main conclusions arising from the investigation
of the various scenarios and case studies that were simulated:
I. BEB energy consumption simulation: - the case studies result indicates that
including the different classified traffic scenarios (i.e., free, move, appropriate,
inappropriate and critical traffic) provides different energy consumption figures
due to BEB speed and trip time variations. Other published work failing to account
for this added value is demonstrated through this activity. The numerical studies
also reveal that traffic conditions affect the overall BEB energy consumption
figure, which can be as much as 180% of its expected value in critical traffic
situations. This percentage increase is usually due to lower speed values and
longer trip duration in high traffic conditions.
II. Transit fleet electrification simulation studies - Scenario I (Simulation of fixed

terminal charger capacity): - a comparison of three types of chargers is classed in
this scenario based on their power capacities (400kW, 200kW, and 50kW,
respectively). The findings of the studies conducted demonstrate that BEB
onboard batteries can be recharged quickly with a higher capacity charger. While

this is not fully unexpected, it does confirm that fewer charger units are needed,
and smaller BEB onboard batteries can be used even when external BEB and route
influences are included. Thus, the investment cost is cheaper overall as compared
to the use of low-capacity chargers that require more charging points and BEBs
with a large onboard battery.
III. Transit fleet electrification simulation studies - Scenario II (Simulation of priority

charging strategy): - this second scenario is based on a novel priority charging
strategy that allocates high priority, medium priority, and low priority chargers at
each BEB terminal based on the state of charge on arrival. The high priority
chargers in this scenario are set to a maximum capacity of 400kW. Chargers with
a maximum capacity of 200kW are termed medium-capacity, while those with a
capacity of 50kW are considered low-capacity. The priority charging technique
presented in this thesis minimises grid demand while simultaneously charging
several BEBs. As a result, it can lessen the impact of BEBs' charging demand on
the distribution network and the wider grid and potentially support DSM (mitigate
the grid's valley and peak effect). Again it is highlighted that such features are
attractive to many network operators in developing countries where supply and
demand imbalance outages are much more common than in European networks.
In addition, the cost of constructing the proposed charging infrastructures (priority
chargers) is less than the cost of constructing 400kWh capacity chargers. Finally,
the priority charging system described in this thesis forms the basis of a low-cost
energy management system (EMS) that can support the minimisation of energy
consumption in the short and long run. Furthermore, it lowers the cost of power
infrastructure because the concept spreads out energy consumption over the
transportation system's daily timetable.
IV. Transit fleet electrification simulation studies - Scenario III (Simulation of the
multi-terminal integrated charging model): - This scenario considers a novel
priority charging technique incorporating conductive and inductive chargers. It
can be inferred that scenario (III), which lowers the optimal cost of bus transit
system electrification by a significant amount, also illustrates that investing in

charging infrastructure is preferable to increasing onboard BEB battery capacity.

Furthermore, the more BEBs there are, the less expensive it gets because charging
infrastructure, particularly on-the-fly inductive chargers, can be shared. The
findings of some of the work on novel priority charging modelling and simulation
have been published in the IEEE PES/IAS Power Africa, 2020.
V. The transit electrification simulation for this scenario IV is based on a bus transit
system using inductive charging infrastructure. The simulation approach
optimises charging station infrastructure and improves operational service
reliability by displaying the cost-benefit trade-off. According to the optimisation
model, the optimal cost for bus transit infrastructure is £52,155,184.98. In this
circumstance, 24kWh of onboard battery capacity is optimal. As a result, the BEB
will be less expensive.
VI. This thesis proposes the design of a multi-terminal inductive base model of BEB's
charging infrastructure for the Lagos Nigeria transport system. The solution is
obtained using a PSO algorithm that aims to discover the minimum possible size
of the BEB onboard battery, the length of the inductive cable and the location of
inductive power transmitters to meet the required route demands in three
possible scenarios. The first scenario considers an inductive charging system for
the existing Lagos bus rapid transit system, as illustrated in Figure 4-23, with the
identified significant bus stops (i.e. bus terminals) considered the primary location
of the inductive chargers. The second scenario investigates the optimal
configuration of the inductive chargers at six equal-distance locations from each
other along the Lagos BRT transit network route Figure 5-1. The third scenario
investigates the possibility of having four equal-distance locations for the
inductive chargers Figure 5-2.
The simulation result for the first scenario yields an optimal solution, indicating
that a 26kWh onboard battery would be sufficient to meet the transit route's
requirement. The second scenario reveals that for each BEB running on the
proposed inductive transit route, a smaller onboard battery with a capacity of

16kWh should suffice. In the third scenario, the optimal cost is the lowest as a
result of using four transmitters at equal distances from each other.
When comparing the original location Figure 4-15 (scenario I) to the equal
distance scenario II, the result shows that significant savings can be achieved if
the inductive chargers are placed at equal distances from each other. The option
in Figure 5-2 (scenario III) illustrates that having four terminals in the Lagos
transit system can provide an appropriate trade-off between the inductive charger
locations and the size of the onboard battery. The placement of the inductive
charger at major stops maximises the utilisation of the inductive charging
infrastructure. The cost of charging infrastructure is reduced when it is placed at
an equal distance from each other, as seen in the case of Lagos infrastructure,
because the battery size required is smaller. Moreover, the state charge and
sensitivity analysis carried out in this thesis also support the simulation result. The
allocation of inductive charging infrastructure provided in this thesis is a cost-
cutting method that saves short-and long-term costs. The state of charge and
sensitivity analysis were performed to back up the simulation's results. The
findings of some of the work on inductive modelling and simulation have been
published at the IEEE International Smart Cities Conference 2020.
VII. A comparative case study of carbon emissions savings of the electrified BRT
system is conducted based on the Lagos BRT case studies. Estimates are provided
for the potential reduction in GHG emissions arising from moving a BEB
delivered version of the BRT. In varied traffic situations, the consumption of
diesel and electric buses was compared with the associated GHG emission figure.
The results show that deploying BEBs along the Lagos, Nigeria BRT route might
save 13 to 36 tons of CO2 per BEB deployed per year. Furthermore, assuming that
Lagos BRT operates an average of 400 BEBs every schedule, GHG emissions are
predicted to be reduced by 5339 tCO2 per year to 14442 tCO2 per year.
The benefit of an electrified transportation sector would be for EVs to have a

considerable share of road transportation demand. Although, this depends on overcoming
considerable obstacles. The most significant include the expensive cost, primarily due to
the batteries; the limited range; the lengthy battery recharge times; and the lack of public

charging infrastructure. This thesis developed a strategy for electrifying public

transportation systems that offers excellent prospects for solving transportation
electrification hurdles.
First, the small onboard battery allows for significant cost savings while investing in
charging infrastructure. The replacement of existing diesel buses with BEBs can also
result in considerable reductions in GHG emissions. This, together with a future
substantially decarbonised electricity supply network, opens up the possibility of a nearly
completely decarbonised road transportation sector. Second, the hub-based charging
proposed in this work can encourage mini-grids and energy sources that enable energy
diversification, ensuring energy supply stability and the widespread use of carbon-free
energy sources. Third, by proposing the necessary infrastructure for effective transit BEB
charging. Although, a considerable improvement would be required to support the
additional power demand for electrifying transportation.
In general, efforts are being undertaken to solve Nigeria's current energy shortfall. The
separation of distribution industry is a key endeavour to improve competition by dividing
distribution businesses into network and supply businesses. As part of sweeping power
sector reforms in 2013, the Nigerian government privatised 11 electricity distribution
firms and six generating companies to boost capacity, extend access to electricity, and
improve transmission. Additionally, the Renewable Energy Master Plan, which was
introduced in 2011, aims to increase the proportion of renewable energy in the nation's
energy mix by at least 13% by 2015, 23% by 2025, and 36% by 2030. Although the
country's present generating capacity is estimated to be 5 GW, Nigeria is thought to have
a potential solar energy output of over 427 GW. A Power Purchase Agreement worth $2.5
billion was signed in 2016 between the nation and 14 independent power producers for
the construction of solar power plants around the nation, which are projected to increase
the grid's capacity by roughly 1.1 GW. These projects are, however, at a standstill because
of some problems, including pricing structures and uncertainty about the ability of the
present transmission system to handle the extra electricity output. Smaller micro-grid
initiatives have the most significant promise, according to industry experts.

Hydroelectric energy accounts for around 20% of the country's installed capacity.
According to studies, there is a possibility for 11,500 MW of capacity in significant
hydropower plants and up to 730 MW in small hydropower projects.
Onshore wind power generation in Nigeria has enormous potential. A 100 MW wind
power project, is currently in the works, and offshore wind resources are being assessed
and mapped. Off-grid solar power systems are increasingly being used to replace more
expensive diesel generators around the country, both for commercial and industrial uses.
Hence, emerging nations' increasing interest in low-carbon solutions, such as

investment in renewable energy resources and EV adoption. The conversion of the Lagos
BRT system to an electric system is crucial for boosting mass passenger transportation in
the city while also developing low-emission transportation alternatives. Furthermore,
Nigerian grids are deficient, with voltage-sensitive to the connection on the new load.
The existing capacity is insufficient to meet home and industrial demand. The total
available capacity in 2019 is 3.7GW, while peak demand is 8.25GW [98]. According to
a recent National Bureau of Statistics estimate, Nigerians only get 6.8 hours of energy
each day from the national grid [99].
In this scenario, the findings of this research are critical for addressing charging
infrastructures that will support transport electrification with little to no reliance on the
existing power system, as well as providing guidance for the deployment of energy
solutions that will meet transit energy demand.
6.2 Future Work

1. The work of this thesis explored the electrification of the rapid bus system as an
alternative to individual cars; shifting from individual cars to transit buses will
reduce congestion and GHG emission and improve safety. This work makes
extensive use of a novel method developed for allocating charging infrastructures
for the transit system by integrating both conductive and inductive chargers.
As identified in Chapter 1, this thesis focused on applications in developing

countries for the decarbonisation of transport where the security of the grid supply
is already challenged. Hence, there is a need to propose solutions to meet the

energy demand, in this case of the bus transit systems, that account for these grid
issues while encouraging the integration of various sources of energy supply
(most especially renewable energy sources). In the African case, the local energy
system is anticipated to comprise increasing solar energy sources, the energy
storage system (ESS), the existing grid, and backup diesel generators.
This proposed bus transit energy system is hoping to consider solar energy
system as the primary source of energy; this solar renewable energy system is
characterised by intermittence that necessitates the integration of the ESS.
Moreover, the grid in developing countries is considered weak and not
dependable, especially for an effective public transit system. Therefore, the transit
energy system only considers buying from the grid when the primary sources (i.e.
the solar and ESS cannot meet the system energy demand) and selling to the grid
when there is a surplus generated energy. The diesel generator is only considered
to minimise the system's Co2 emission in an emergency need.
The proposed operation of the transit energy system that may be considered for
future study is illustrated in Figure 6-1. Connected to the solar energy source is
the PV converter that tracks the maximum power point of the generated solar
energy. The central controller is considered to perform the function of both the
power converter and energy management system that control various chargers.
An intelligent bidirectional hybrid power converter controls the ESS, the
interconnected grid and the diesel generator, as shown in Figure 6-1.
Figure 6-1 Proposed Terminal Based Transit Energy System

In addition, the modelling of this future proposed work is expected to extend to

the multi-terminal allocation of the transit energy system, as illustrated in Figure
6-2.
Figure 6-2 Proposed Multi-terminal Based Transit Energy System
The proposed renewable resource is expected to cater charging demand of the

charging model proposed in this work. This solution is considered because the
security of grid supply is already challenged. Besides supporting challenged grids,
it reduces pollution from both transport and power generation points of view.
Hence, this introduces a mathematical problem that needs to be formulated in the
form of an optimisation model for sizing various components of the proposed
transit energy system to minimise the cost while optimally sizing the system
components.
As a result, it is envisaged that industry, academia, and the government will
examine bus transit electrification as a feasible option for more sustainable
transportation. Governments and funding organisations are also urged to
recognise and promote the potential benefits of an electrified bus transit system.
2. The key determinant of BEB's operational strategy and cost analysis is energy
consumption. The BEB energy consumption figure might be a crucial factor in
calculating the fuel cost and greenhouse gas emissions for fleet buses, which could
have significant economic and ecological implications. An accurate estimation of
the BEB energy consumption figure is also essential for its development,
deployment, and necessary charging infrastructure. One of the critical variables

affecting the BEB energy consumption figure is the speed profile, which should
be available to accurately assess the BEB energy consumption figure. This speed
profile has been considered in this thesis, along with other parameters such as
route topography, weather, traffic, and passenger occupancy rate.
In this thesis, the passenger rate considered is a full bus load, which is typically
the case for the Lagos BRT considered in this thesis. However, it would be
beneficial if future research work could consider scenarios of more representative
use cases, such as variations in bus loading, traffic congestion, and charger
capacity along the route length, as well as over the course of the day and also
annual variations.

Chapter 7: References
7 REFERENCES
[1] I. Manisalidis, E. Stavropoulou, A. Stavropoulos, and E. Bezirtzoglou,

‘Environmental and Health Impacts of Air Pollution: A Review’, Frontiers in
Public Health. 2020.
[2] W. Health Organization, ‘Ambient air pollution: a global assessment of exposure

and burden of disease’, Clean Air J., 2016.
[3] D. Kim, Z. Chen, L.-F. Zhou, and S.-X. Huang, ‘Air pollutants and early origins
of respiratory diseases’, Chronic Dis. Transl. Med., vol. 4, no. 2, pp. 75–94, 2018.
[4] WHO, ‘How air pollution is destroying our health’, World Health Organization,
2018. .
[5] R. Hannah and R. Max, ‘Air Pollution (by “Our World in Data”)’, ‘Indoor Air
Pollution’. Published online at OurWorldInData.org, 2017. .
[6] J. Lin, S. Lu, X. He, and F. Wang, ‘Analyzing the impact of three-dimensional
building structure on CO2 emissions based on random forest regression’, Energy,
vol. 236, p. 121502, Dec. 2021.
[7] International Energy Agency (IEA), ‘Cities are at the frontline of the energy
transitionNo Title’, 2016. [Online]. Available: https://www.iea.org/news/cities-
are-at-the-frontline-of-the-energy-transition. [Accessed: 02-Aug-2020].
[8] World Health Organization, ‘WHO’s Urban Ambient Air Pollution database ‐
Update 2016’, Who, 2016.
[9] J. M. Baldasano, E. Valera, and P. Jiménez, ‘Air quality data from large cities’,
Sci. Total Environ., 2003.
[10] C. H. Tan, M. Y. Ong, S. M. Nomanbhay, A. H. Shamsuddin, and P. L. Show,

‘The Influence of COVID-19 on Global CO2 Emissions and Climate Change: A
Perspective from Malaysia’, Sustain. 2021, Vol. 13, Page 8461, vol. 13, no. 15, p.

8461, Jul. 2021.
[11] D. Burchart-Korol and P. Folęga, ‘Impact of road transport means on climate

change and human health in poland’, Promet - Traffic - Traffico, vol. 31, no. 2, pp.
195–204, Mar. 2019.
[12] L. Carter and J. Varghese, ‘Electric Bus Technology’, Transp. Res. Rep., no. June,
p. 57, 2017.
[13] M. Glotz-Richter and H. Koch, ‘Electrification of Public Transport in Cities

(Horizon 2020 ELIPTIC Project)’, in Transportation Research Procedia, 2016.
[14] M. Mahmoud, R. Garnett, M. Ferguson, and P. Kanaroglou, Electric buses: A

review of alternative powertrains, vol. 62. Pergamon, 2016, pp. 673–684.
[15] R. Faria et al., ‘Real-world performance of battery electric buses and their life-
cycle benefits with respect to energy consumption and carbon dioxide emissions’,
Energy, 2013.
[16] Y. Ligen, H. Vrubel, and H. H. Girault, ‘Mobility from renewable electricity:

Infrastructure comparison for battery and hydrogen fuel cell vehicles’, World
Electr. Veh. J., 2018.
[17] H. Khani, N. A. El-Taweel, and H. E. Z. Farag, ‘Supervisory Scheduling of

Storage-Based Hydrogen Fueling Stations for Transportation Sector and
Distributed Operating Reserve in Electricity Markets’, IEEE Trans. Ind.
Informatics, 2020.
[18] ERTRAC; EPoSS; and ETIP SNET, ‘European Roadmap Electrification of Road
Transport Status: final for publication’, no. June, 2017.
[19] C. Cluzel, A. H. Morley, and J. Murray, ‘Transport Energy Infrastructure Roadmap

to 2050 Prepared for the LowCVP by Element Energy Ltd ELECTRICITY
ROADMAP’, no. June, 2015.
[20] A. Asaolu, S. Galloway, and C. Edmunds, ‘Allocation of the inductive charging

system for bus rapid transit network’, in 2020 IEEE International Smart Cities
Conference, ISC2 2020, 2020.
[21] Y. Li, C. Zhan, M. de Jong, and Z. Lukszo, ‘Business innovation and government

regulation for the promotion of electric vehicle use: lessons from Shenzhen,
China’, J. Clean. Prod., 2016.
[22] C. Randall, ‘Daimler sets up electric bus charging station in Mannheim’,

media.daimler.com, 2019. [Online]. Available:
https://www.electrive.com/2019/08/05/daimler-sets-up-electric-bus-charging-
station-in-mannheim/. [Accessed: 15-Jan-2020].
[23] T. K. Wellik, J. R. Griffin, K. M. Kockelman, and M. Mohamed, ‘Utility-transit

nexus: Leveraging intelligently charged electrified transit to support a renewable
energy grid’, Renew. Sustain. Energy Rev., vol. 139, p. 110657, Apr. 2021.
[24] K. Gwilliam, ‘Cities on the move - Ten years after’, Res. Transp. Econ., 2013.
[25] ‘Cities on the move: a World Bank urban transport strategy review’, Choice Rev.
Online, 2003.
[26] A. P. Ajayi, ‘Assessment of the Influence of Lagos Bus Rapid Transit Scheme
(BRT-Lite) on Road Traffic Crashes (RTC) on Lagos Mainland-Island Corridor’,
Open Transp. J., vol. 11, no. 1, pp. 102–109, 2017.
[27] World Health Organization, ‘Global Status Report on Road Safety 2013:
Supporting a Decade ofAction’, 2015.
[28] A. P. Ajayi, ‘Analytical study of supply chain collaboration guiding forestry

products supply in a developing economy’, Int. J. Bus. Perform. Supply Chain
Model., 2016.
[29] B. Otunola, S. Kriticos, and O. Harman, ‘The BRT and the danfo : A case study of
Lagos ’ transport reforms from 1999-2019’, IGC, Cities that Work. Case Study.,
2019.
[30] A. A. Adeyanju and K. Manohar, ‘Effects of Vehicular Emission on

Environmental Pollution in Lagos’, vol. 5, no. 4, pp. 34–51, 2017.
[31] D. Mobereola, ‘Africa ’ s first BRT scheme’, SSATP Discuss. Pap. No. 9 Urban
Transp. Ser. Available http//www.ssatp.org/sites/ssatp/files/publications/SSATP-
DiscussionPapers/DP09-Lagos-BRT.pdf, no. 9, p. 54, 2009.
[32] ‘Decarbonising transport’, Traffic Eng. Control, vol. 51, no. 7, pp. 246–248, 2020.

[33] A. Asaolu, S. Galloway, and C. Edmunds, ‘Optimal design of conductive and

inductive charging system for bus rapid transit network’, in 2020 IEEE PES/IAS
PowerAfrica, PowerAfrica 2020, 2020.
[34] I. Wagner, ‘Number of licensed drivers in the United States from 1990 to 2018’,
2020. [Online]. Available: https://www.statista.com/statistics/191653/number-of-
licensed-drivers-in-the-us-since-1988/. [Accessed: 17-Jan-2021].
[35] E. YURDAY, ‘Number of Cars in the UK 2020’, 2020. [Online]. Available:

https://www.nimblefins.co.uk/number-cars-great-britain. [Accessed: 17-Jan-
2020].
[36] B. Zaccheaus, ‘LAGOS AND MOTOR VEHICLE ADMINISTRATION’, 2017.

[Online]. Available: https://lagosstate.gov.ng/blog/2017/07/05/lagos-and-motor-
vehicle-administration/. [Accessed: 17-Jan-2020].
[37] K. Lucas, G. Stokes, J. Bastiaanssen, and J. Burkinshaw, ‘Inequalities in Mobility

and Access in the UK Transport System. Future of Mobility: Evidence Review’,
p. 78, 2019.
[38] A. May, Transport Planning and Traffic Engineering. 2007.
[39] G. Santos, H. Behrendt, and A. Teytelboym, ‘Part II: Policy instruments for
sustainable road transport’, Res. Transp. Econ., 2010.
[40] J. Wilkins and S. Weekes, ‘Transport Statistics Great Britain 2017 Prepared for
publication by ’:, no. November, pp. 1–288, 2017.
[41] M. Hasselwander, J. F. Bigotte, A. P. Antunes, and R. G. Sigua, ‘Towards

sustainable transport in developing countries: Preliminary findings on the demand
for mobility-as-a-service (MaaS) in Metro Manila’, Transp. Res. Part A Policy
Pract., vol. 155, pp. 501–518, Jan. 2022.
[42] N. Unidas, ‘Traffic congestion: the problem and how to deal with it’.
[43] A. Koźlak and D. Wach, ‘Causes of traffic congestion in urban areas. Case of
Poland’, SHS Web Conf., 2018.
[44] N. A. El-Taweel, A. Zidan, and H. E. Z. Farag, ‘Novel Electric Bus Energy

Consumption Model Based on Probabilistic Synthetic Speed Profile Integrated

With HVAC’, IEEE Trans. Intell. Transp. Syst., 2020.
[45] A. H. Ahmed, ‘Adaptive intelligent traffic control systems for congestion

management’, gjastonline.com.
[46] J. De Vos and F. Witlox, ‘Transportation policy as spatial planning tool; Reducing
urban sprawl by increasing travel costs and clustering infrastructure and public
transportation’, J. Transp. Geogr., 2013.
[47] A. de Nazelle, O. Bode, and J. P. Orjuela, ‘Comparison of air pollution exposures

in active vs. passive travel modes in European cities: A quantitative review’,
Environment International. 2017.
[48] P. Romilly, ‘Substitution of bus for car travel in urban Britain: An economic
evaluation of bus and car exhaust emission and other costs’, Transp. Res. Part D
Transp. Environ., 1999.
[49] N. Government, ‘Future transport strategy 2056’, Nature, 2008.
[50] S. Strick, ‘Keynote speech ESAR Conference’, Berichte der Bundesanstalt fuer
Strassenwesen. Unterr. Fahrzeugtechnik, 2013.
[51] J. Damsere-Derry, G. Palk, and M. King, ‘Road accident fatality risks for
“vulnerable” versus “protected” road users in northern Ghana’, Traffic Injury
Prevention. 2017.
[52] T. Jia, Q. Li, C. Ma, and Y. Li, ‘Computing the CO2 Emissions of Taxi Trajectories
and Exploring Their Spatiotemporal Patterns in Wuhan City’, Wuhan Daxue
Xuebao (Xinxi Kexue Ban)/Geomatics Inf. Sci. Wuhan Univ., 2019.
[53] P. Batty, R. Palacin, and A. González-Gil, ‘Challenges and opportunities in

developing urban modal shift’, Travel Behav. Soc., 2015.
[54] A. Nurdden, R. A. O. K. Rahmat, and A. Ismail, ‘Effect of transportation policies

on modal shift from private car to public transport in Malaysia’, J. Appl. Sci., vol.
7, no. 7, pp. 1013–1018, Apr. 2007.
[55] F. Alrawi and Y. Hadi, ‘The application of bus rapid transit system in the city of
Baquba and its impact on reducing daily trips’, in Advances in Intelligent Systems
and Computing, 2020.

[56] R. Banick, ‘Bus Rapid Transit and the Latin American City: Successes to Date,
But Miles to Go’, Council on Hemispheric Affairs, 2009. .
[57] M. M. A. W. Ahmed and N. A. El Monem, ‘Sustainable and green transportation

for better quality of life case study greater Cairo – Egypt’,
https://doi.org/10.1080/16874048.2020.1719340, vol. 16, no. 1, pp. 17–37, Jan.
2020.
[58] F. Ahmad, M. S. Alam, S. M. Shariff, and M. Krishnamurthy, ‘A cost-efficient

approach to EV charging station integrated community microgrid: A case study of
indian power market’, IEEE Trans. Transp. Electrif., 2019.
[59] Natural Resources Canada, ‘Fuel Consumption Guide’, Gov. Canada, pp. 1–39,
2018.
[60] M. Xylia and S. Silveira, ‘On the road to fossil-free public transport: The case of
Swedish bus fleets’, Energy Policy, 2017.
[61] J. M. Greenwald and A. Kornhauser, ‘It’s up to us: Policies to improve climate

outcomes from automated vehicles’, Energy Policy, 2019.
[62] Y. Geng, Z. Ma, B. Xue, W. Ren, Z. Liu, and T. Fujita, ‘Co-benefit evaluation for
urban public transportation sector - A case of Shenyang, China’, J. Clean. Prod.,
2013.
[63] L. Wright and K. Fjellstrom, ‘Mass Transit Options’, Gesellschaft für Tech.
Zusammenarbeit, pp. 1–38, 2010.
[64] A. Delgocio, ‘Design of Electric Vehicle Smort Chorging Strategies to Moximize

the Self-Consumption of o Hybrid Power Ploint’, orbit.dtu.dk.
[65] E. Mulenga, M. H. J. Bollen, and N. Etherden, ‘Adapted Stochastic PV Hosting

Capacity Approach for Electric Vehicle Charging Considering Undervoltage’,
Electr. 2021, Vol. 2, Pages 387-402, vol. 2, no. 3, pp. 387–402, Sep. 2021.
[66] S. Naumov, D. R. Keith, and J. D. Sterman, ‘Accelerating vehicle fleet turnover to

achieve sustainable mobility goals’, J. Oper. Manag., 2022.
[67] N. V. Emodi, C. C. Emodi, G. P. Murthy, and A. S. A. Emodi, ‘Energy policy for

low carbon development in Nigeria: A LEAP model application’, Renewable and

Sustainable Energy Reviews, vol. 68. pp. 247–261, 2017.
[68] J. D. D. Niyonteze, F. Zou, G. Norense Osarumwense Asemota, S. Bimenyimana,

and G. Shyirambere, ‘Key technology development needs and applicability
analysis of renewable energy hybrid technologies in off-grid areas for the Rwanda
power sector’, Heliyon, vol. 6, no. 1, p. e03300, Jan. 2020.
[69] ‘Opportunities for Africa in electric vehicle market - African Business’. [Online].
Available: https://african.business/2021/09/technology-information/opportunities-
for-africa-in-electric-vehicle-market/. [Accessed: 28-Dec-2021].
[70] M. Boulakhbar et al., ‘Towards a large-scale integration of renewable energies in

Morocco’, J. Energy Storage, vol. 32, p. 101806, Dec. 2020.
[71] D. McPhail, ‘Evaluation of ground energy storage assisted electric vehicle DC fast
charger for demand charge reduction and providing demand response’, Renew.
Energy, 2014.
[72] M. O. Badawy and Y. Sozer, ‘Power Flow Management of a Grid Tied PV-Battery
System for Electric Vehicles Charging’, IEEE Trans. Ind. Appl., 2017.
[73] S. Negarestani, M. Fotuhi-Firuzabad, M. Rastegar, and A. Rajabi-Ghahnavieh,

‘Optimal sizing of storage system in a fast charging station for plug-in hybrid
electric vehicles’, IEEE Trans. Transp. Electrif., 2016.
[74] H. Ding, Z. Hu, and Y. Song, ‘Value of the energy storage system in an electric
bus fast charging station’, Appl. Energy, 2015.
[75] I. S. Bayram, G. Michailidis, M. Devetsikiotis, S. Bhattacharya, A. Chakrabortty,

and F. Granelli, ‘Local energy storage sizing in plug-in hybrid electric vehicle
charging stations under blocking probability constraints’, in 2011 IEEE
International Conference on Smart Grid Communications, SmartGridComm 2011,
2011.
[76] O. Rahbari et al., ‘An optimal versatile control approach for plug-in electric
vehicles to integrate renewable energy sources and smart grids’, Energy, 2017.
[77] S. Shojaabadi, S. Abapour, M. Abapour, and A. Nahavandi, ‘Optimal planning of

plug-in hybrid electric vehicle charging station in distribution network considering

demand response programs and uncertainties’, IET Gener. Transm. Distrib., 2016.
[78] C. Capasso, D. Iannuzzi, and O. Veneri, ‘DC charging station for electric and plug-
in vehicles’, in Energy Procedia, 2014.
[79] V. I. Herrera, A. Milo, H. Gaztanaga, A. Gonzalez-Garrido, H. Camblong, and A.

Sierra, ‘Design and Experimental Comparison of Energy Management Strategies
for Hybrid Electric Buses Based on Test-Bench Simulation’, in IEEE Transactions
on Industry Applications, 2019.
[80] T. Zhang, X. Chen, Z. Yu, X. Zhu, and D. Shi, ‘A Monte Carlo Simulation
Approach to Evaluate Service Capacities of EV Charging and Battery Swapping
Stations’, IEEE Trans. Ind. Informatics, 2018.
[81] N. A. El-Taweel, M. Mohamed, and H. E. Farag, ‘Optimal design of charging

stations for electrified transit networks’, in 2017 IEEE Transportation and
Electrification Conference and Expo, ITEC 2017, 2017.
[82] M. Mohamed, H. Farag, N. El-Taweel, and M. Ferguson, ‘Simulation of electric

buses on a full transit network: Operational feasibility and grid impact analysis’,
Electr. Power Syst. Res., 2017.
[83] ‘Figure of the week: Electricity access in Africa’. [Online]. Available:

https://www.brookings.edu/blog/africa-in-focus/2019/03/29/figure-of-the-week-
electricity-access-in-africa/. [Accessed: 30-Dec-2021].
[84] Cable, ‘Electric Cars To Be Introduced Into Nigerian Market By 2018’, 2016.
[Online]. Available: https://nipc.gov.ng/2017/08/25/electric-cars-introduced-
nigerian-market-2018/.
[85] H. Ranjbar and M. Sharifzadeh, ‘Electrification of Transportation’, in Industry 4.0

Vision for Energy and Materials, John Wiley & Sons, Ltd, 2022, pp. 269–296.
[86] F. Agenda, ‘The Electrification of Transport’, 2018. [Online]. Available:

https://www.futureagenda.org/news/the-electrification-of-transport-key-
questions. [Accessed: 25-Jun-2019].
[87] E. T. Asimeng and D. Heinrichs, ‘Why do paratransit operators resist participation

in bus rapid transit? Case evidence from Bogota, Mexico City, Johannesburg and

Lagos’, https://doi.org/10.1080/01441647.2020.1818872, vol. 41, no. 1, pp. 115–

135, 2020.
[88] P. Finn, C. Fitzpatrick, D. C.- Energy, and undefined 2012, ‘Demand side
management of electric car charging: Benefits for consumer and grid’, Elsevier.
[89] P. Shinde, K. S.-2016 F. I. C. on, and undefined 2016, ‘Optimal Electric Vehicle
charging schedule for demand side management’, ieeexplore.ieee.org.
[90] C. Pang, M. Kezunovic, … M. E.-I. E. V., and undefined 2012, ‘Demand side
management by using electric vehicles as distributed energy resources’,
ieeexplore.ieee.org.
[91] M. Tushar, … A. Z.-I. T. on, and undefined 2017, ‘Demand-side management by

regulating charging and discharging of the EV, ESS, and utilizing renewable
energy’, ieeexplore.ieee.org.
[92] H. Alhelou, M. G.-2016 24th I. C. on, and undefined 2016, ‘Hierarchical plug-in
EV control based on primary frequency response in interconnected smart grid’,
[93] X. Zhu, M. Xia, H. C.-A. Energy, and undefined 2018, ‘Coordinated sectional
droop charging control for EV aggregator enhancing frequency stability of
microgrid with high penetration of renewable energy sources’, Elsevier.
[94] U. Datta, A. Kalam, J. Shi, J. L.-2019 14th I. C. on, and undefined 2019, ‘Electric
vehicle charging station for providing primary frequency control in microgrid’,
[95] H. J. Vermaak and K. Kusakana, ‘Design of a photovoltaic-wind charging station

for small electric Tuk-tuk in D.R.Congo’, Renew. Energy, 2014.
[96] O. M. Mueller and E. K. Mueller, ‘Off-grid, low-cost, electrical sun-car system for
developing countries’, in Proceedings of the 4th IEEE Global Humanitarian
Technology Conference, GHTC 2014, 2014.
[97] R. Geometry and G. Analysis, ‘Mini-Grids in Nigeria - The Three-Pronged

Approach of Investment Opportunities, Cost Reduction and Temporary Political
Gains’.

[98] A. E. Airoboman, F. O. Adunola, H. A. Salihu, I. S. Fahim, and M. A. Aderibigbe,

‘Electricity pricing in an emerging deregulated market: A Case of Nigeria and
Egypt.’, 6th IEEE Int. Energy Conf. ENERGYCon 2020, pp. 797–801, Sep. 2020.
[99] A. Adikpe, A. Yaro, A. Tekanyi, M. Almustapha, E. Agbon, and O. Ayofe, ‘A

Review on Technology-Based Contact Tracing Solutions and Its Application in
Developing Countries’, Jordan J. Electr. Eng., vol. 8, no. 1, p. 48, 2022.
[100] N. A. El-Taweel, H. E. Z. Farag, and M. Mohamed, ‘Integrated Utility-Transit

Model for Optimal Configuration of Battery Electric Bus Systems’, IEEE Syst. J.,
2020.
[101] A. Asaolu, S. Galloway, and C. Edmunds, ‘Allocation of the inductive charging

system for bus rapid transit network’, 2020.
[102] M. Mohamed, M. Ferguson, and P. Kanaroglou, ‘What hinders adoption of the

electric bus in Canadian transit? Perspectives of transit providers’, Transp. Res.
Part D Transp. Environ., 2018.
[103] D. Silva Ardila, ‘Global policies for moving cities: the role of think tanks in the
proliferation of Bus Rapid Transit systems in Latin America and worldwide’,
Policy Soc., 2020.
[104] M. Jalalian, S. Gholami, and R. Ramezanian, ‘Analyzing the trade-off between

CO2 emissions and passenger service level in the airline industry: Mathematical
modeling and constructive heuristic’, J. Clean. Prod., 2019.
[105] N. A. El-Taweel, H. Khani, and H. E. Z. Farag, ‘Analytical Size Estimation

Methodologies for Electrified Transportation Fueling Infrastructures Using Public-
Domain Market Data’, IEEE Trans. Transp. Electrif., 2019.
[106] L. Sieber, C. Ruch, S. Hörl, K. W. Axhausen, and E. Frazzoli, ‘Improved public

transportation in rural areas with self-driving cars: A study on the operation of
Swiss train lines’, Transp. Res. Part A Policy Pract., vol. 134, pp. 35–51, Apr.
2020.
[107] L. Mejia-Dorantes, A. Paez, and J. M. Vassallo, ‘Transportation infrastructure

impacts on firm location: the effect of a new metro line in the suburbs of Madrid’,
J. Transp. Geogr., vol. 22, pp. 236–250, May 2012.

[108] A. Weinstock, W. Hook, M. Replogle, and R. Cruz, ‘Recapturing Global

Leadership in Bus Rapid Transit’, Itdp, 2011.
[109] Institute for Transportation and Development Policy (ITDP), ‘Public Transport’,
2011. .
[110] P. Abelson, ‘Cost benefit analysis of proposed major rail development in lagos,
nigeria: Foreign summaries’, Transp. Rev., 1995.
[111] ‘Africa’s rail sector poised for growth | International Railway Journal’. [Online].
Available: https://www.railjournal.com/in_depth/africas-rail-sector-poised-for-
growth. [Accessed: 31-Dec-2021].
[112] Africa Development Bank., ‘Rail infrastructure in africa: financing policy options.
Abidjan, Côte d’Ivoire: AFrica Development Bank: Transport, Urban
Development & ICT Department.’, 2015.
[113] H. Suzuki, R. Cervero, and K. Iuchi, Transforming Cities with Transit. 2013.
[114] Planning and design for sustainable urban mobility: Global report on human
settlements 2013. 2013.
[115] Colin Hughes & Xianyuan Zhu, ‘Guangzhou , China Bus Rapid Transit -Emission
Impact Analysis’, no. May, 2011.
[116] W. H. et al., ‘The impact of a Bus Rapid Transit system on commuters’ exposure
to Benzene, CO, PM2.5 and PM10 in Mexico City’, Atmospheric Environment.
2008.
[117] S. A. R. Shah et al., ‘Performance evaluation of bus rapid transit system: A

comparative analysis of alternative approaches for energy efficient eco-friendly
public transport system’, Energies, 2020.
[118] ‘Dedicated Bus Lanes, Without the Extra Lane - Bloomberg’. [Online]. Available:
https://www.bloomberg.com/news/articles/2018-05-31/dedicated-bus-lanes-
without-the-extra-lane. [Accessed: 31-Dec-2021].
[119] S. Jiang, H. Xue, and Z.-X. Li, ‘Intelligent Systems and Applications’, Intell. Syst.
Appl., vol. 2, pp. 56–63, 2010.
[120] M. Poku-Boansi and G. Marsden, ‘Bus rapid transit systems as a governance

reform project’, J. Transp. Geogr., 2018.
[121] E. Maeso-González and P. Pérez-Cerón, ‘State of art of bus rapid transit

transportation’, Eur. Transp. Res. Rev., 2014.
[122] S. Elshaikh, I. Kapoor, and U. Change, ‘Bus Rapid Transit as a Sustainable Mode
of Public Transport for Khartoum’, 2021.
[123] J. C. Echeverry, A. M. Ibanez, A. Moya, and L. C. Hillon, ‘The Economics of

TransMilenio, a Mass Transit System for Bogota’, Economía, 2005.
[124] INE, ‘The Benefits and Costs of a Bus Rapid Transit System in Mexico City’, Fuel,
2008.
[125] E. Vaz and C. Venter, ‘the Effectiveness of Bus Rapid Transit As Part of a Poverty-
Reduction Strategy: Some Early Impacts in Johannesburg’, Abstr. 31st South.
African Transp. Conf. (SATC 2012), 2012.
[126] P. Alpkokin and M. Ergun, ‘Istanbul Metrobüs: first intercontinental bus rapid
transit’, J. Transp. Geogr., 2012.
[127] J. P. Bocarejo, J. M. Velasquez, C. A. Díaz, and L. E. Tafur, ‘TRR 2317’, Transp.

Res. Rec. J. Transp. Res. Board Transp. Res. Board Natl. Acad., 2012.
[128] M. A. Yazici, H. S. Levinson, M. Ilicali, N. Camkesen, and C. Kamga, ‘A bus

rapid transit line case study: Istanbul’s metrobüs system’, J. Public Transp., 2013.
[129] A. Carrigan, R. King, J. M. Velasquez, M. Raifman, and N. Duduta, ‘Social,

Environmental and Economic Impacts of Brt Systems’, p. 151, 2013.
[130] World Health Organization, ‘WHO/Europe | HEAT | Health economic assessment

tool’, WHO Website, 2015. .
[131] D. Stead and D. Pojani, ‘The urban transport crisis in emerging economies: A
comparative overview’, Urban B. Ser., pp. 283–295, 2017.
[132] L. C.-J. of transport geography and undefined 2007, ‘Transport and climate
change: a review’, Elsevier.
[133] K. Olagunju, ‘Evaluating Traffic Congestion in Developing Countries – a case

study of Nigeria .’, 2015 Chart. Inst. Logist. Transp. Africa forum held Mt. Meru

Hotel. Arusha, Tanzania, 2015.
[134] S. Bharadwaj, S. Ballare, Rohit, and M. K. Chandel, ‘Impact of congestion on

greenhouse gas emissions for road transport in Mumbai metropolitan region’, in
Transportation Research Procedia, 2017, vol. 25, pp. 3538–3551.
[135] V. Jain, A. Sharma, and L. Subramanian, ‘Road Traffic Congestion in the

Developing World’, Proc. 2nd ACM Symp. Comput. Dev. - ACM DEV ’12.
[136] J. F. K. Akinbami and S. O. Fadare, ‘Strategies for sustainable urban and transport
development in Nigeria’, Transp. Policy, 1997.
[137] M. A. Mosaberpanah and S. D. Khales, ‘The role of transportation in sustainable

development’, in ICSDEC 2012: Developing the Frontier of Sustainable Design,
Engineering, and Construction - Proceedings of the 2012 International
Conference on Sustainable Design and Construction, 2013.
[138] W. R. Black, ‘North American transportation: Perspectives on research needs and

sustainable transportation’, Journal of Transport Geography. 1997.
[139] World Bank, ‘Safe, Clean, and Affordable… Transport for Development’, World
Bank Gr., p. 113, 2013.
[140] A. Abdelwahed, P. L. van den Berg, and T. Brandt, ‘Enabling sustainable public
transport in smart cities through real-time decision support’, in 40th International
Conference on Information Systems, ICIS 2019, 2020.
[141] S. Borén, L. Nurhadi, and H. Ny, ‘Preferences of Electric Buses in public

Transport; Conclusions from Real Life Testing in Eight Swedish Municipalities.’,
Int. J. Environ. Ecol. Eng., 2016.
[142] A. García-Olivares, J. Solé, R. Samsó, and J. Ballabrera-Poy, ‘Sustainable

European transport system in a 100% renewable economy’, Sustain., 2020.
[143] M. A. Bhuva, ‘Bus Rapid Transit System: Enhancing and Sustaining Mobility in
Urban Traffic.’, Daw. J. Contemp. Res. Manag., 2015.
[144] D. A. Hensher and T. F. Golob, ‘Bus rapid transit systems: A comparative

assessment’, Transportation (Amst)., 2008.
[145] M. Khan, M. Babar, S. H. Ahmed, S. C. Shah, and K. Han, ‘Smart city designing

and planning based on big data analytics’, Sustain. Cities Soc., 2017.
[146] J. A. Bunce, ‘A segregated urban busway system’, Transp. Plan. Technol., 1973.
[147] M. Garcia and K. Yamamoto, ‘Busways and bus lanes in Brazil and Japan’, 2002.
[148] R. Guevara, ‘Exploring BRT in Lima, Peru’, Multiple Cites, 2014. [Online].
Available: https://www.multiplecities.org/home/2017/12/28/exploring-brt-in-
lima-peru. [Accessed: 13-Oct-2020].
[149] R. Iles, ‘Public Transport Stations and Terminals’, in Public Transport in

Developing Countries, Emerald Group Publishing Limited, 2005, pp. 91–109.
[150] S. C. Wirasinghe, G. J. Quain, U. Vandebona, and J. M. S. J. Bandara, ‘Optimal

Terminus Location for a Rail Line with Many to Many Travel Demand’, Transp.
Traffic Theory 21 st Century , pp. 75–97, Jun. 2002.
[151] ‘Oshodi Transport Interchange – Planet Projects’. [Online]. Available:

https://planetprojectsltd.com/projects/oshodi-transport-interchange/. [Accessed:
03-Jan-2022].
[152] S. C. Wirasinghe and U. Vandebona, ‘Route layout analysis for express buses’,
Transp. Res. Part C Emerg. Technol., vol. 19, no. 2, pp. 374–385, Apr. 2011.
[153] R. Fernandez and R. Fernández, ‘Design of bus stop priorities’, Traffic Eng.
Control, vol. 40, no. 6, pp. 335–340, 1999.
[154] ‘Lagos’ Bus Rapid Transit System: Decongesting and Depolluting Mega-Cities’.
[Online]. Available: https://blogs.worldbank.org/transport/lagos-bus-rapid-transit-
system-decongesting-and-depolluting-mega-cities-0. [Accessed: 02-Jan-2022].
[155] B. Zeng and Y. Luo, ‘Potential of harnessing operational flexibility from public
transport hubs to improve reliability and economic performance of urban multi-
energy systems: A holistic assessment framework’, Appl. Energy, vol. 322, p.
119488, Sep. 2022.
[156] Okeowo Yinka, ‘Primero withdraws Lagos BRT services, gives reasons –
TechEconomy.ng’, 28-Oct-2020. [Online]. Available:
https://techeconomy.ng/2020/10/again-primero-withdraws-lagos-brt-services-
gives-reasons/. [Accessed: 03-Jan-2022].

[157] ‘Raised Boarding Platforms Will Make 14th Street Busway Even Better –
Streetsblog New York City’. [Online]. Available:
https://nyc.streetsblog.org/2019/11/06/raised-boarding-platforms-will-make-
14th-street-busway-even-better/. [Accessed: 03-Jan-2022].
[158] ‘Bogota’s Transmilenio: Prepaid boarding & passenger information’. [Online].

Available: https://www.pinterest.com/pin/2040762312650932/. [Accessed: 24-
Aug-2021].
[159] ITDP, ‘Climate and Transport Policy’, Institute for Transportation and
Development Policy, 2013. [Online]. Available: https://www.itdp.org/what-we-
do/climate-and-transport-policy/. [Accessed: 14-Oct-2020].
[160] N. Kholod and M. Evans, ‘Reducing black carbon emissions from diesel vehicles
in Russia: An assessment and policy recommendations’, Environ. Sci. Policy, vol.
56, pp. 1–8, Feb. 2016.
[161] B. Zhou et al., ‘Real-world performance of battery electric buses and their life-
cycle benefits with respect to energy consumption and carbon dioxide emissions’,
Energy, vol. 96, pp. 603–613, Feb. 2016.
[162] M. Mahmoud, R. Garnett, M. Ferguson, and P. Kanaroglou, ‘Electric buses: A

review of alternative powertrains’, Renewable and Sustainable Energy Reviews,
vol. 62. Elsevier Ltd, pp. 673–684, 2016.
[163] and J. van de S. Arnout de Pee, Hauke Engel, Marte Guldemond, Arjan Keizer,
‘The European electric bus market is charging ahead, but how will it develop’,
Transport Research Report, 2018. [Online]. Available:
https://www.mckinsey.com/industries/oil-and-gas/our-insights/the-european-
electric-bus-market-is-charging-ahead-but-how-will-it-develop. [Accessed: 12-
Oct-2020].
[164] S. Deuten, J. J. Gómez Vilchez, and C. Thiel, ‘Analysis and testing of electric car
incentive scenarios in the Netherlands and Norway’, Technol. Forecast. Soc.
Change, vol. 151, 2020.
[165] F. Thorpe, ‘Global Battery Electric Bus Market’. [Online]. Available:

http://www.cte.tv/wp-content/uploads/2016/12/5_Thorpe.pdf. [Accessed: 25-

Aug-2021].
[166] ‘Electric bus, main fleets and projects around the world’. [Online]. Available:
https://www.sustainable-bus.com/electric-bus/electric-bus-public-transport-main-
fleets-projects-around-world/. [Accessed: 25-Aug-2021].
[167] BYD ADL, ‘BYD Alexander Dennis Electric Buses’. [Online]. Available:
https://www.evbus.co.uk/products/. [Accessed: 15-Oct-2020].
[168] ‘Bus: C6M - BYD USA’. [Online]. Available: https://en.byd.com/bus/bus-c6m/.

[Accessed: 25-Aug-2021].
[169] ‘Bus: C9M - BYD USA’. [Online]. Available: https://en.byd.com/bus/bus-c9m/.

[170] ‘Bus: K11M - BYD USA’. [Online]. Available: https://en.byd.com/bus/k11m/.

[171] P. Katalog, ‘EMISSIONS POWERTRAINS’, 2020.
[172] P. Di Justo, ‘Volvo 7900 Electric’, Wired, pp. 2002–2003, 2002.
[173] ‘Electric bus, main fleets and projects around the world’, Sustainable Bus, 2020.
[Online]. Available: https://www.sustainable-bus.com/electric-bus/electric-bus-
public-transport-main-fleets-projects-around-world/. [Accessed: 15-Oct-2020].
[174] C. Yang, W. Lou, J. Yao, and S. Xie, ‘On Charging Scheduling Optimization for
a Wirelessly Charged Electric Bus System’, IEEE Trans. Intell. Transp. Syst.,
2018.
[175] H. Lee, H. K. School, and A. Clark, ‘Charging the Future: Challenges and
Opportunities for Electric Vehicle Adoption Faculty Research Working Paper
Series’, 2018.
[176] Y. He, Z. Liu, and Z. Song, ‘Optimal charging scheduling and management for a
fast-charging battery electric bus system’, Transp. Res. Part E Logist. Transp.
Rev., vol. 142, p. 102056, Oct. 2020.
[177] U. Guida and S. Leonard, ‘ZeEUS: Zero Emission Urban Bus System’, in 2014
IEEE International Electric Vehicle Conference, IEVC 2014, 2014.

[178] S. Sawilla and O. Schütt, ‘Hands on Sustainable Mobility Technical Overview of

Inductive Charging for Electric Buses in Europe’, 2019.
[179] J. Dong, C. Liu, and Z. Lin, ‘Charging infrastructure planning for promoting
battery electric vehicles: An activity-based approach using multiday travel data’,
Transp. Res. Part C Emerg. Technol., 2014.
[180] J. A. Manzolli, ‘DECISION SUPPORT FOR PLANNING A BUS RAPID

TRANSIT CHARGING’, no. March, 2020.
[181] A. Houbbadi, R. Trigui, S. Pelissier, E. Redondo-Iglesias, and T. Bouton, ‘Optimal

Scheduling to Manage an Electric Bus Fleet Overnight Charging’, Energies, vol.
12, no. 14, p. 2727, 2019.
[182] Z. Liu, Z. Song, and Y. He, ‘Economic Analysis of On-Route Fast Charging for
Battery Electric Buses: Case Study in Utah’, Transp. Res. Rec., 2019.
[183] Y. He, Z. Song, and Z. Liu, ‘Fast-charging station deployment for battery electric
bus systems considering electricity demand charges’, Sustain. Cities Soc., 2019.
[184] J. A. Manzolli, ‘Decision Support for Planning a Bus Rapid Transit Charging
Infrastructure’, 2020.
[185] Daimler, ‘THE MERCEDES-BENZ ECITARO’, 2018. [Online]. Available:

https://www.akasol.com/en/reference-ecitaro. [Accessed: 30-Dec-2020].
[186] S. S. and O. Sawilla, ‘Wireless Opportunity Charging buses in Madrid’, 2021.

[Online]. Available: https://ipt-technology.com/case-opportunity-charging-
madrid/. [Accessed: 29-Mar-2021].
[187] ‘Largest Full Operation Fast Charging Network For An E Bus Fleet In Europe’.
[Online]. Available: https://www.heliox-energy.com/case-studies/largest-full-
operation-fast-charging-network-for-an-e-bus-fleet-in-europe. [Accessed: 31-
Dec-2021].
[188] Y. J. Jang, Y. D. Ko, and S. Jeong, ‘Optimal design of the wireless charging electric
vehicle’, in 2012 IEEE International Electric Vehicle Conference, IEVC 2012,
2012.
[189] Long Branch Mike, ‘BERLIN PLANS HYBRID TROLLEYBUS NETWORK

(URBANTRANSPORT)’, 2021. [Online]. Available:

https://www.londonreconnections.com/2020/berlin-plans-hybrid-trolleybus-
network-urbantransport/. [Accessed: 29-Mar-2021].
[190] T. Gnann, S. Funke, N. Jakobsson, P. Plötz, F. Sprei, and A. Bennehag, ‘Fast

charging infrastructure for electric vehicles: Today’s situation and future needs’,
Transp. Res. Part D Transp. Environ., vol. 62, no. March, pp. 314–329, 2018.
[191] M. Rogge, S. Wollny, and D. U. Sauer, ‘Fast charging battery buses for the
electrification of urban public transport-A feasibility study focusing on charging
infrastructure and energy storage requirements’, Energies, 2015.
[192] R. Sclar, C. Gorguinpour, S. Castellanos, and X. Li, ‘Barriers to Adopting Electric

Buses’, WRI World Resour. Inst., p. 60, 2019.
[193] M. Adhikari, L. P. Ghimire, Y. Kim, P. Aryal, and S. B. Khadka, ‘Identification

and analysis of barriers against electric vehicle use’, Sustain., vol. 12, no. 12, pp.
1–20, 2020.
[194] I. Borok, J. Agandu, and M. M. Morgan, ‘Energy Security in Nigeria: Challenges

and Way Forward’, Int. J. Eng. Sci. Invent. ISSN (Online, vol. 2, no. 11, pp. 2319–
6734, 2013.
[195] M. Xylia, S. Leduc, P. Patrizio, F. Kraxner, and S. Silveira, ‘Locating charging

infrastructure for electric buses in Stockholm’, Transp. Res. Part C Emerg.
Technol., 2017.
[196] X. Qu, L. Zhen, R. J. Howlett, and L. C. Jain, Smart Transportation Systems 2020.
2020.
[197] M. Pihlatie, S. Kukkonen, T. Halmeaho, V. Karvonen, and N. O. Nylund, ‘Fully

electric city buses - The viable option’, in 2014 IEEE International Electric
Vehicle Conference, IEVC 2014, 2015.
[198] Bloomberg New Energy Finance, ‘Electric Buses in Cities: Driving Towards
Cleaner Air and Lower CO2’, p. 63, 2018.
[199] Sustainable Bus, ‘Electric bus range, focus on electricity consumption. A sum-up’,
2020. [Online]. Available: https://www.sustainable-bus.com/news/electric-bus-

range-focus-on-electricity-consumption-a-sum-up/. [Accessed: 30-Mar-2020].
[200] M. Weiss, K. C. Cloos, and E. Helmers, ‘Energy efficiency trade-offs in small to

large electric vehicles’, Environ. Sci. Eur., vol. 32, no. 1, Dec. 2020.
[201] Low Carbon Vehicle Partnership, ‘THE LOW EMISSION Interactive version’,
2016.
[202] T. Paul and H. Yamada, ‘Operation and charging scheduling of electric buses in a
city bus route network’, in 2014 17th IEEE International Conference on Intelligent
Transportation Systems, ITSC 2014, 2014.
[203] Y. Wang, Y. Huang, J. Xu, and N. Barclay, ‘Optimal recharging scheduling for
urban electric buses: A case study in Davis’, Transp. Res. Part E Logist. Transp.
Rev., 2017.
[204] J. Lee, B. J. Lee, G. L. Park, and Y. C. Kim, ‘Web service-based tour-and-charging

scheduler framework for rent-a-car systems employing electric vehicles’, Int. J.
Control Autom., 2013.
[205] T. Lan, J. Hu, Q. Kang, C. Si, L. Wang, and Q. Wu, ‘Optimal control of an electric
vehicle’s charging schedule under electricity markets’, Neural Comput. Appl.,
2013.
[206] A. Aabrandt et al., ‘Prediction and optimization methods for electric vehicle
charging schedules in the EDISON project’, in 2012 IEEE PES Innovative Smart
Grid Technologies, ISGT 2012, 2012.
[207] Y. He, B. Venkatesh, and L. Guan, ‘Optimal scheduling for charging and
discharging of electric vehicles’, IEEE Trans. Smart Grid, 2012.
[208] A. Kunith, D. Goehlich, and R. Mendelevitch, ‘Planning and optimization of a fast-

charging infrastructure for electric urban bus systems’, Int. Conf. Traffic Transp.
Eng., 2015.
[209] A. Kunith, R. Mendelevitch, and D. Goehlich, ‘Electrification of a city bus

network—An optimization model for cost-effective placing of charging
infrastructure and battery sizing of fast-charging electric bus systems’, Int. J.
Sustain. Transp., 2017.

[210] S. Y. R. Hui, W. Zhong, and C. K. Lee, ‘A critical review of recent progress in

mid-range wireless power transfer’, IEEE Transactions on Power Electronics.
2014.
[211] H. Z. Z. Beh, G. A. Covic, and J. T. Boys, ‘Wireless fleet charging system for
electric bicycles’, IEEE J. Emerg. Sel. Top. Power Electron., 2015.
[212] S. Ahn, J. Y. Lee, D. H. Cho, and J. Kim, ‘Magnetic field design for low emf and
high efficiency wireless power transfer system in on-line electric vehicles’, in
Proceedings of the 21st CIRP Design Conference, Korea 2011: Interdisciplinary
Design, 2011.
[213] Y. Yang, M. El Baghdadi, Y. Lan, Y. Benomar, J. Van Mierlo, and O. Hegazy,

‘Design methodology, modeling, and comparative study of wireless power transfer
systems for electric vehicles’, Energies, 2018.
[214] Z. Bi, L. Song, R. De Kleine, C. C. Mi, and G. A. Keoleian, ‘Plug-in vs. wireless
charging: Life cycle energy and greenhouse gas emissions for an electric bus
system’, Appl. Energy, 2015.
[215] Z. Wang, X. Wei, and H. Dai, ‘Design and control of a 3 kW wireless power
transfer system for electric vehicles’, Energies, 2016.
[216] F. Musavi and W. Eberle, ‘Overview of wireless power transfer technologies for
electric vehicle battery charging’, IET Power Electron., 2014.
[217] Y. D. Ko and Y. J. Jang, ‘The optimal system design of the online electric vehicle
utilizing wireless power transmission technology’, IEEE Trans. Intell. Transp.
Syst., vol. 14, no. 3, pp. 1255–1265, 2013.
[218] Y. Kongjeen and K. Bhumkittipich, ‘Impact of Plug-in Electric Vehicles Integrated

into Power Distribution System Based on Voltage-Dependent Power Flow
Analysis’, Energies 2018, Vol. 11, Page 1571, vol. 11, no. 6, p. 1571, Jun. 2018.
[219] H. Shareef, M. M. Islam, and A. Mohamed, ‘A review of the stage-of-the-art

charging technologies, placement methodologies, and impacts of electric vehicles’,
Renewable and Sustainable Energy Reviews. 2016.
[220] A. Ahmad, M. S. Alam, and R. Chabaan, ‘A Comprehensive Review of Wireless

Charging Technologies for Electric Vehicles’, IEEE Trans. Transp. Electrif., 2017.
[221] M. Yilmaz and P. T. Krein, ‘Review of charging power levels and infrastructure
for plug-in electric and hybrid vehicles’, in 2012 IEEE International Electric
Vehicle Conference, IEVC 2012, 2012.
[222] R. Collin, Y. Miao, A. Yokochi, P. Enjeti, and A. Von Jouanne, ‘Advanced electric
vehicle fast-charging technologies’, Energies, 2019.
[223] M. Al-Saadi et al., ‘Impact on the Power Grid Caused via Ultra-Fast Charging
Technologies of the Electric Buses Fleet’, Energies, vol. 15, no. 4, p. 1424, Feb.
2022.
[224] ‘OPPCharge. OPPChargeCommon Interface for Automated Charging of Hybrid

Electric and Electric Commercial Vehicles. 2019.’ .
[225] SIEMENS, ‘Charging systems for ebuses’, 2020. [Online]. Available:

https://new.siemens.com/my/en/markets/transportation-
logistics/electromobility/ebus-charging.html. [Accessed: 31-Dec-2020].
[226] O. L. de B. Gorosabel, ‘Assessing the Construction of Charging Stations for

Electric Buses: A Case Study for Stockholm Inner City’, p. 108, 2017.
[227] WSP Canada Inc., ‘Lima eBus Charging Infrastructure’. [Online]. Available:
https://globalelectricity.org/content/uploads/WSP5-Charging_Infrastructure.pdf.
[Accessed: 05-Oct-2020].
[228] M. R. Sarker, H. Pandžić, and M. A. Ortega-Vazquez, ‘Optimal operation and

services scheduling for an electric vehicle battery swapping station’, IEEE Trans.
Power Syst., 2015.
[229] F. Ahmad, M. S. Alam, I. S. Alsaidan, and S. M. Shariff, ‘Battery swapping station

for electric vehicles: Opportunities and challenges’, IET Smart Grid, vol. 3, no. 3,
pp. 280–286, 2020.
[230] T. Budde Christensen, P. Wells, and L. Cipcigan, ‘Can innovative business models
overcome resistance to electric vehicles? Better Place and battery electric cars in
Denmark’, Energy Policy, vol. 48, pp. 498–505, 2012.
[231] F. Adegbohun, A. von Jouanne, and K. Y. Lee, ‘Autonomous battery swapping

system and methodologies of electric vehicles’, Energies. 2019.
[232] ‘Better Place Battery Swapping model’. [Online]. Available: http://www.change-

climate.com/Transport_Land_Sea_Sustainable/SMART_Hybrid_EV_Energy_Se
rvice_Networks/BETTER_PLACE_Israeli_Service_Stations_Renault_Fluence_
Battery_Cartridges_EVs.htm. [Accessed: 06-Oct-2020].
[233] T. Budde Christensen, P. Wells, and L. Cipcigan, ‘Can innovative business models
overcome resistance to electric vehicles? Better Place and battery electric cars in
Denmark’, Energy Policy, 2012.
[234] D. Connolly, ‘Economic viability of electric roads compared to oil and batteries
for all forms of road transport’, Energy Strateg. Rev., vol. 18, pp. 235–249, 2017.
[235] J. Vetter et al., ‘Ageing mechanisms in lithium-ion batteries’, J. Power Sources,

2005.
[236] C. Zhou, K. Qian, M. Allan, and W. Zhou, ‘Modeling of the cost of EV battery
wear due to V2G application in power systems’, IEEE Trans. Energy Convers.,
vol. 26, no. 4, pp. 1041–1050, Dec. 2011.
[237] F. Burmeister, L. Schnieder, and T. Kurczveil, ‘Simulation Based Studies on the

Integration of Battery-Electric Vehicles in Regional Bus Services’, in IEEE
Conference on Intelligent Transportation Systems, Proceedings, ITSC, 2015.
[238] G. De Filippo, V. Marano, and R. Sioshansi, ‘Simulation of an electric

transportation system at The Ohio State University’, Appl. Energy, 2014.
[239] T. Kurzcvezil, L. Schnieder, and F. Burmeister, ‘Optimized energy management

of inductively charged electric buses reflecting operational constraints and traffic
conditions’, in 2015 International Conference on Models and Technologies for
Intelligent Transportation Systems, MT-ITS 2015, 2015.
[240] G. Domingues-olavarría, Analysis of Electromobility Systems. .
[241] X. Qi, G. Wu, K. Boriboonsomsin, and M. J. Barth, ‘Data-driven decomposition

analysis and estimation of link-level electric vehicle energy consumption under
real-world traffic conditions’, Transp. Res. Part D Transp. Environ., 2018.
[242] N. A. El-Taweel and H. E. Z. Farag, ‘Incorporation of Battery Electric Buses in

the Operation of Intercity Bus Services’, in ITEC 2019 - 2019 IEEE Transportation
Electrification Conference and Expo, 2019.
[243] M. Gallet, T. Massier, and T. Hamacher, ‘Estimation of the energy demand of

electric buses based on real-world data for large-scale public transport networks’,
Appl. Energy, vol. 230, no. January, pp. 344–356, 2018.
[244] A. Lajunen, ‘Lifecycle costs and charging requirements of electric buses with
different charging methods’, J. Clean. Prod., 2018.
[245] Y. J. Jang, E. S. Suh, and J. W. Kim, ‘System architecture and mathematical

models of electric transit bus system utilizing wireless power transfer technology’,
IEEE Syst. J., 2016.
[246] Y. Gao, S. Guo, J. Ren, Z. Zhao, A. Ehsan, and Y. Zheng, ‘An electric bus power
consumption model and optimization of charging scheduling concerning multi-
external factors’, Energies, 2018.
[247] G. E. Blomgren, ‘The Development and Future of Lithium Ion Batteries’, J.

Electrochem. Soc., 2017.
[248] E. S. Pampal, E. Stojanovska, B. Simon, and A. Kilic, ‘A review of nanofibrous

structures in lithium ion batteries’, Journal of Power Sources. 2015.
[249] L. Li, S. You, and C. Yang, ‘Multi-Objective Stochastic MPC-Based System

Control Architecture for Plug-In Hybrid Electric Buses’, IEEE Trans. Ind.
Electron., 2016.
[250] M. T. Sebastiani, R. Luders, and K. V. O. Fonseca, ‘Evaluating Electric Bus

Operation for a Real-World BRT Public Transportation Using Simulation
Optimization’, IEEE Trans. Intell. Transp. Syst., 2016.
[251] Z. Chen, L. Li, B. Yan, C. Yang, C. Marina Martinez, and D. Cao, ‘Multimode
Energy Management for Plug-In Hybrid Electric Buses Based on Driving Cycles
Prediction’, IEEE Trans. Intell. Transp. Syst., 2016.
[252] G. Correa, P. Muñoz, T. Falaguerra, and C. R. Rodriguez, ‘Performance

comparison of conventional, hybrid, hydrogen and electric urban buses using well
to wheel analysis’, Energy, 2017.

[253] E. Tazelaar, J. Bruinsma, B. Veenhuizen, and P. van den Bosch, ‘Driving cycle
characterization and generation, for design and control of fuel cell buses’, World
Electr. Veh. J., 2009.
[254] D. Krajzewicz, J. Erdmann, M. Behrisch, and L. Bieker, ‘Recent Development and

Applications of {SUMO - Simulation of Urban MObility}’, Int. J. Adv. Syst.
Meas., 2012.
[255] R. Maia, M. Silva, R. Araujo, and U. Nunes, ‘Electric vehicle simulator for energy
consumption studies in electric mobility systems’, in 2011 IEEE Forum on
Integrated and Sustainable Transportation Systems, FISTS 2011, 2011.
[256] M. Yaich, M. R. Hachicha, and M. Ghariani, ‘Modeling and simulation of electric

and hybrid vehicles for recreational vehicle’, 16th Int. Conf. Sci. Tech. Autom.
Control Comput. Eng. STA 2015, vol. 95, no. 4, pp. 181–187, 2016.
[257] T. Markel et al., ‘ADVISOR: A systems analysis tool for advanced vehicle
modeling’, in Journal of Power Sources, 2002.
[258] T. Halmeaho, P. Rahkola, A. P. Pellikka, K. Tammi, and S. Ruotsalainen, ‘Electric

City Bus Energy Flow Model and Its Validation by Dynamometer Test’, 2015
IEEE Veh. Power Propuls. Conf. VPPC 2015 - Proc., pp. 3–8, 2015.
[259] H. S. Cho and H. S. Chang, ‘Electric power consumption analysis model based on
user activity for power saving’, Proc. - 2013 IEEE Int. Work. Intell. Energy Syst.
IWIES 2013, pp. 95–100, 2013.
[260] A. Kontou and J. Miles, ‘Electric Buses: Lessons to be Learnt from the Milton
Keynes Demonstration Project’, in Procedia Engineering, 2015.
[261] C. L. Guo, L. Wu, D. Wang, W. B. Qi, and X. N. Xiao, ‘Impact of electric vehicle
charging on power grid’, in 2011 International Conference on Electrical and
Control Engineering, ICECE 2011 - Proceedings, 2011.
[262] R. C. Green, L. Wang, and M. Alam, ‘The impact of plug-in hybrid electric
vehicles on distribution networks: A review and outlook’, Renewable and
Sustainable Energy Reviews. 2011.
[263] R. C. Leou, C. L. Su, and C. N. Lu, ‘Stochastic analyses of electric vehicle charging

impacts on distribution network’, IEEE Trans. Power Syst., 2014.
[264] K. Clement-Nyns, E. Haesen, and J. Driesen, ‘The impact of Charging plug-in

hybrid electric vehicles on a residential distribution grid’, IEEE Trans. Power
Syst., 2010.
[265] J. Taylor, A. Maitra, M. Alexander, D. Brooks, and M. Duvall, ‘Evaluations of

plug-in electric vehicle distribution system impacts’, in IEEE PES General
Meeting, PES 2010, 2010.
[266] S. Ruayariyasub, S. Sirisumrannukul, and S. Wangsatitwong, ‘Impact of

Residential and Public Charging Stations of Electric Vehicles on Distribution
System by Stochastic Simulation’, Adv. Mater. Res., 2015.
[267] K. Yunus and A. B. B. Ab, ‘Distribution Grid Impact of Plug-In Electric Vehicles
Charging at Fast Charging Stations Using Stochastic Charging Model Keywords
Need for Fast Charging’, Power Electron. Appl. (EPE 2011), Proc. 2011-14th Eur.
Conf., 2011.
[268] K. Clement, E. Haesen, and J. Driesen, ‘Coordinated charging of multiple plug-in

hybrid electric vehicles in residential distribution grids’, in 2009 IEEE/PES Power
Systems Conference and Exposition, PSCE 2009, 2009.
[269] H. LIU, P. ZENG, J. GUO, H. WU, and S. GE, ‘An optimization strategy of
controlled electric vehicle charging considering demand side response and regional
wind and photovoltaic’, J. Mod. Power Syst. Clean Energy, 2015.
[270] S. Huang, ‘Statistical Analysis of Passenger Car Use to Model the Impact of
Electric Vehicle Take-up on the Power Distribution Network Sikai Huang’, no.
October, 2014.
[271] J. De Hoog, T. Alpcan, M. Brazil, D. A. Thomas, and I. Mareels, ‘Optimal charging

of electric vehicles taking distribution network constraints into account’, IEEE
Trans. Power Syst., 2015.
[272] N. Qin, A. Gusrialdi, R. Paul Brooker, and A. T-Raissi, ‘Numerical analysis of

electric bus fast charging strategies for demand charge reduction’, Transp. Res.
Part A Policy Pract., 2016.

[273] L. E. Teoh, H. L. Khoo, S. Y. Goh, and L. M. Chong, ‘Scenario-based electric bus

operation: A case study of Putrajaya, Malaysia’, Int. J. Transp. Sci. Technol., 2018.
[274] B. R. Ke, C. Y. Chung, and Y. C. Chen, ‘Minimizing the costs of constructing an

all plug-in electric bus transportation system: A case study in Penghu’, Appl.
Energy, 2016.
[275] M. Vahedipour-Dahraie, H. Rashidizaheh-Kermani, H. Najafi, A. Anvari-

Moghaddam, and J. Guerrero, ‘Coordination of EVs Participation for Load
Frequency Control in Isolated Microgrids’, Appl. Sci., 2017.
[276] O. Hafez and K. Bhattacharya, ‘Optimal design of electric vehicle charging

stations considering various energy resources’, Renew. Energy, 2017.
[277] F. Locment and M. Sechilariu, ‘DC microgrid for future electric vehicle charging
station designed by Energetic Macroscopic Representation and Maximum Control
Structure’, in ENERGYCON 2014 - IEEE International Energy Conference, 2014.
[278] S. G. Yoon and S. G. Kang, ‘Economic microgrid planning algorithm with electric
vehicle charging demands’, Energies, 2017.
[279] F. Marra et al., ‘EV charging facilities and their application in LV feeders with
photovoltaics’, IEEE Trans. Smart Grid, vol. 4, no. 3, pp. 1533–1540, 2013.
[280] H. Zhao and A. Burke, ‘An intelligent solar powered battery buffered EV charging
station with solar electricity forecasting and EV charging load projection
functions’, in 2014 IEEE International Electric Vehicle Conference, IEVC 2014,
2014.
[281] V. Lakshminarayanan, S. Pramanick, K. Rajashekara, L. Ben-Brahim, and A.

Gastli, ‘Optimal energy management scheme for electric vehicle integration in
microgrid’, in 2017 North American Power Symposium, NAPS 2017, 2017.
[282] K. Aidoo and R. C. Briggs, ‘Underpowered: Rolling blackouts in Africa

disproportionately hurt the poor’, Afr. Stud. Rev., vol. 62, no. 3, pp. 112–131, Sep.
2019.
[283] C. Kumavat, H. Sonawane, T. Patel, X. Sakhalkar, and A. Pansare, ‘EFFECTIVE

LEARNING FROM DELHI BRTS-A CASE STUDY OF PUNE BRTS eSAT

Journals EFFECTIVE LEARNING FROM DELHI BRTS-A CASE STUDY OF

PUNE BRTS’.
[284] M. Bielli, A. Bielli, and R. Rossi, ‘Trends in models and algorithms for fleet
management’, in Procedia - Social and Behavioral Sciences, 2011.
[285] V. Guihaire and J. K. Hao, ‘Transit network design and scheduling: A global
review’, Transp. Res. Part A Policy Pract., 2008.
[286] N. Regional, P. Commission, and T. Program, ‘A Roadmap to Understanding

Transportation in Your Community’, no. August, 2007.
[287] M. Bartlomiejczyk, ‘Driving Performance Indicators of Electric Bus Driving

Technique: Naturalistic Driving Data Multicriterial Analysis’, IEEE Trans. Intell.
Transp. Syst., vol. 20, no. 4, 2019.
[288] X. He et al., ‘Energy consumption and well-to-wheels air pollutant emissions of

battery electric buses under complex operating conditions and implications on fleet
electrification’, J. Clean. Prod., 2018.
[289] X. Feng, M. Lewis, and C. Hearn, ‘Modeling and validation for zero emission
buses’, 2017 IEEE Transp. Electrif. Conf. Expo, ITEC 2017, pp. 501–506, Jul.
2017.
[290] L. Li, C. Yang, Y. Zhang, L. Zhang, and J. Song, ‘Correctional DP-Based Energy
Management Strategy of Plug-In Hybrid Electric Bus for City-Bus Route’, IEEE
Trans. Veh. Technol., vol. 64, no. 7, pp. 2792–2803, Jul. 2015.
[291] A. Ana¨ıssia and F. B. Eng, ‘Electricity consumption and battery lifespan

estimation for transit electric buses: drivetrain simulations and electrochemical
modelling’, 2018.
[292] A. Franca, ‘Electricity consumption and battery lifespan estimation for transit
electric buses: drivetrain simulations and electrochemical modelling’, Isbn, 2018.
[293] O. Sassi and A. Oulamara, ‘Electric vehicle scheduling and optimal charging
problem: complexity, exact and heuristic approaches’,
https://doi.org/10.1080/00207543.2016.1192695, vol. 55, no. 2, pp. 519–535, Jan.
2016.

[294] M. Potkány, M. Hlatká, M. Debnár, and J. Hanzl, ‘Comparison of the lifecycle cost
structure of electric and diesel buses’, Nase More, 2018.
[295] H. Y. Mak, Y. Rong, and Z. J. M. Shen, ‘Infrastructure planning for electric

vehicles with battery swapping’, Manage. Sci., 2013.
[296] W. Li, Y. Li, H. Deng, and L. Bao, ‘Planning of electric public transport system
under battery swap mode’, Sustain., 2018.
[297] J. Q. Li, ‘Battery-electric transit bus developments and operations: A review’,

International Journal of Sustainable Transportation. 2016.
[298] R. Wei, X. Liu, Y. Ou, and S. Kiavash Fayyaz, ‘Optimizing the spatio-temporal
deployment of battery electric bus system’, J. Transp. Geogr., 2018.
[299] N. A. El-Taweel, M. Mohamed, and H. E. Farag, ‘Optimal design of charging

stations for electrified transit networks’, in 2017 IEEE Transportation and
Electrification Conference and Expo, ITEC 2017, 2017.
[300] Z. Zhang, H. Pang, A. Georgiadis, and C. Cecati, ‘Wireless Power Transfer - An

Overview’, IEEE Transactions on Industrial Electronics. 2019.
[301] S. Li and C. C. Mi, ‘Wireless power transfer for electric vehicle applications’,
IEEE J. Emerg. Sel. Top. Power Electron., 2015.
[302] X. Mou and H. Sun, ‘Wireless power transfer: Survey and roadmap’, in IEEE
Vehicular Technology Conference, 2015.
[303] A. Karalis, J. D. Joannopoulos, and M. Soljačić, ‘Efficient wireless non-radiative

mid-range energy transfer’, Ann. Phys. (N. Y)., 2008.
[304] Kurzweil and A. Intelligence, ‘Dealing with “range anxiety”: wireless highways to
charge electric cars’. [Online]. Available: https://www.kurzweilai.net/dealing-
with-range-anxiety-wireless-highways-to-charge-electric-cars. [Accessed: 08-
Jun-2020].
[305] R. Selvakumaran, W. Liu, B. H. Soong, L. Ming, and S. Y. Loon, ‘Design of

inductive coil for wireless power transfer’, in IEEE/ASME International
Conference on Advanced Intelligent Mechatronics, AIM, 2009.
[306] J. C. Lin, ‘Wireless power transfer for mobile applications, and health effects’,

IEEE Antennas Propag. Mag., 2013.
[307] Lagos Metropolitan Area Transport Authority, ‘BRT (Bus Rapid Transit) Lagos’,
Lagos Metropolitan Area Transport Authority, 2015. .
[308] World Population Review, ‘Lagos Population’, World Population Review, 2015.
[Online]. Available: http://worldpopulationreview.com/world-cities/lagos-
population/.
[309] A. Oluwaseyi, ‘Assessment of Bus Rapid Transit in Efficient of Movement of

Commuters in Lagos State’, Eur. J. Humanit. Soc. Sci. Available
https//www.researchgate.net/publication/306344953, 2016.
[310] D. Amiegbebhor and P. Boluwatife, ‘The Lagos Bus Rapid Transit : Review of
Users ’ Perception’, Am. J. Humanit. ansSocial Sci. Res., vol. 02, no. 11, pp. 88–
108, 2018.
[311] O. Osemwegie, S. John, K. Okokpujie, and I. Shorinwa, ‘Development of an

electronic fare collection system using stationary tap-out devices’, in Proceedings
- 2016 International Conference on Computational Science and Computational
Intelligence, CSCI 2016, 2017.
[312] K. B. Wipke, M. R. Cuddy, and S. D. Burch, ‘ADVISOR 2.1: A user-friendly

advanced powertrain simulation using a combined backward/forward approach’,
IEEE Trans. Veh. Technol., vol. 48, no. 6, pp. 1751–1761, 1999.
[313] K. B. Wipke and M. R. Cuddy, ‘Using an Advanced Vehicle Simulator (

ADVISOR ) to Guide Hybrid Vehicle Propulsion System Development’, No.
NREL/TP-425-21615; CONF-9609227-2, Natl. Renew. Energy Lab., Golden, CO,
pp. 120–126, 1996.
[314] K. B. Wipke, M. R. Cuddy, and S. D. Burch, ‘ADVISOR 2.1: A user-friendly

advanced powertrain simulation using a combined backward/forward approach’,
IEEE Trans. Veh. Technol., vol. 48, no. 6, pp. 1751–1761, 1999.
[315] T. Markel et al., ‘ADVISOR: A systems analysis tool for advanced vehicle
modeling’, J. Power Sources, vol. 110, no. 2, pp. 255–266, Aug. 2002.
[316] V. H. Johnson, ‘Battery performance models in ADVISOR’, J. Power Sources,

vol. 110, no. 2, pp. 321–329, Aug. 2002.
[317] R. Myers, ‘Evaluating plug-in series hybrid designs for postal delivery vehicles."
MATLAB Digest Academic Edition"’. .
[318] B. Sørensen, ‘Assessing current vehicle performance and simulating the

performance of hydrogen and hybrid cars’, Int. J. Hydrogen Energy, vol. 32, no.
10–11, pp. 1597–1604, Jul. 2007.
[319] L. Wang and W. Bai, ‘Development and simulation of electric vehicle based on
ADVISOR’, J. Southeast Univ. (English Ed., vol. 22, no. 2, pp. 196–199, Jun.
2006.
[320] A. Same, A. Stipe, D. Grossman, and J. W. Park, ‘A study on optimization of

hybrid drive train using Advanced Vehicle Simulator (ADVISOR)’, J. Power
Sources, vol. 195, no. 19, pp. 6954–6963, Oct. 2010.
[321] ‘BYD ADL Enviro200EV | Alexander Dennis’. [Online]. Available:

https://www.alexander-dennis.com/products/single-deck-buses/byd-adl-
enviro200ev/. [Accessed: 24-Mar-2022].
[322] S. Varnhagen, A. Same, J. Remillard, and J. W. Park, ‘A numerical investigation

on the efficiency of range extending systems using Advanced Vehicle Simulator’,
J. Power Sources, vol. 196, no. 6, pp. 3360–3370, Mar. 2011.
[323] Atlas of Tehran metropolis, ‘Volume Capacity Ratio’, 2004. [Online]. Available:
https://atlas.tehran.ir/Default.aspx?tabid=299. [Accessed: 07-Jan-2021].
[324] S. P. Measures, ‘9 Transportation Analysis Performance’, pp. 1–71, 2019.
[325] N. R. Council and TRB, Highway capacity manual. 2000.
[326] ‘Why are Lagos workers so stressed? “Hellish traffic” | CNN Travel’. [Online].
Available: https://edition.cnn.com/travel/article/traffic-stress-lagos-
nigeria/index.html. [Accessed: 27-Mar-2022].
[327] ‘Lagos commuters lose 75% of weekly working hours to traffic’. [Online].
Available: https://businessday.ng/uncategorized/article/lagos-commuters-lose-75-
of-weekly-working-hours-to-traffic/. [Accessed: 27-Mar-2022].
[328] ‘Lagos Traffic: The Cost of Waiting in Line - Stears Business’. [Online].

Available: https://www.stearsng.com/article/lagos-traffic-the-cost-of-waiting-in-
line. [Accessed: 27-Mar-2022].
[329] ‘In Lagos, the traffic jams can add four hours to your commute - City Monitor’.
[Online]. Available: https://citymonitor.ai/transport/lagos-traffic-jams-can-add-
four-hours-your-commute-1406. [Accessed: 26-Mar-2022].
[330] A. A. Chidiebere and O. C. Felix, ‘Analytical Study of Causes, Effects and

Remedies of Traffic Congestion in Nigeria: Case Study of Lagos State’, Int. J. Eng.
Res. Adv. Technol., 2019.
[331] A. O. Atubi, ‘The Influence of Travel Time on Accessibility in Lagos Island’,

AFRREV STECH An Int. J. Sci. Technol., vol. 1, no. 2, pp. 112–127, Aug. 2014.
[332] ‘Lagos topographic map, elevation, relief’. [Online]. Available: https://en-

gb.topographic-map.com/maps/lpjw/Lagos/. [Accessed: 29-Mar-2022].
[333] B. S. G. de Almeida and V. C. Leite, ‘Particle Swarm Optimization: A Powerful

Technique for Solving Engineering Problems’, Swarm Intell. - Recent Adv. New
Perspect. Appl., Dec. 2019.
[334] A. Slowik, ‘Particle Swarm Optimization’, in The Industrial Electronics Handbook

- Five Volume Set, vol. 4, IEEE, 2011, pp. 1942–1948.
[335] S. Djemame, M. Batouche, H. Oulhadj, and P. Siarry, ‘Solving reverse emergence

with quantum PSO application to image processing’, Soft Comput., vol. 23, no. 16,
pp. 6921–6935, Aug. 2019.
[336] Y. Shi and R. Eberhart, ‘Modified particle swarm optimizer’, in Proceedings of the
IEEE Conference on Evolutionary Computation, ICEC, 1998, pp. 69–73.
[337] Z. Michalewicz and D. B. Fogel, ‘How to Solve It: Modern Heuristics’, How to
Solve It Mod. Heuristics, 2000.
[338] J. E. Fieldsend, J. E. Fieldsend, E. Q. Uk, and S. Singh, ‘A Multi-Objective

Algorithm based upon Particle Swarm Optimisation, an Efficient Data Structure
and Turbulence.’, 2002.
[339] G. T. Pulido and C. A. C. Coello, ‘A constraint-handling mechanism for particle

swarm optimization Evolutionary Multiobjective Optimization View project A

Constraint-Handling Mechanism for Particle Swarm Optimization’,

ieeexplore.ieee.org, 2004.
[340] Y. D. Ko and Y. J. Jang, ‘The optimal system design of the online electric vehicle
utilizing wireless power transmission technology’, IEEE Trans. Intell. Transp.
Syst., 2013.
[341] ‘BYD Europe’. [Online]. Available: https://bydeurope.com/pdp-bus-model-12.

[342] ‘Low Entry City Bus 12m/12,8m-3 doors version’, 2014.
[343] T. R. Tanim et al., ‘Extended cycle life implications of fast charging for lithium-
ion battery cathode’, Energy Storage Mater., vol. 41, pp. 656–666, Oct. 2021.
[344] ‘Distribution grid impact of Plug-In Electric Vehicles charging at fast charging
stations using stochastic charging model | IEEE Conference Publication | IEEE
Xplore’. .
[345] A. Ashtari, E. Bibeau, … S. S.-I. T. on, and undefined 2011, ‘PEV charging profile
prediction and analysis based on vehicle usage data’, ieeexplore.ieee.org.
[346] S. Mohrehkesh and T. Nadeem, ‘Toward a wireless charging for battery electric
vehicles at traffic intersections’, IEEE Conf. Intell. Transp. Syst. Proceedings,
ITSC, pp. 113–118, 2011.
[347] S. Piller, M. Perrin, A. J.-J. of power sources, and undefined 2001, ‘Methods for
state-of-charge determination and their applications’, Elsevier.
[348] N. Resources Canada, ‘Learn the facts: Emissions from your vehicle’.
[349] EIA, ‘Frequently Asked Questions How much carbon dioxide is produced by
burning gasoline and diesel fuel ?’, US Energy Information Administration, 2016.
.
[350] Bulb, ‘Carbon tracker’. [Online]. Available: https://bulb.co.uk/carbon-

tracker/#:~:text=The UK government report on,kg per kWh of gas. [Accessed: 20-
Jan-2021].
[351] J. Liu, Q. Yang, Y. Zhang, W. Sun, and Y. Xu, ‘2019 GOVERNMENT

GREENHOUSE GAS CONVERSION FACTORS FOR COMPANY

REPORTING Methodology Paper for Emission Factors Final Report’, Sustain.,

vol. 11, no. 1, 2019.
[352] USDOE, ‘Average Annual Vehicle Miles Traveled by Major Vehicle Category’,
US Dep. Energy Energy Effic. Renew. Energy, no. February, p. 2020, 2020.
[353] Ozebus, ‘Climate Change and Public Transport’, 2014. [Online]. Available:
http://bic.asn.au/information-for-moving-people/climate-change-and-public-
transport. [Accessed: 03-Feb-2021].
[354] S. SHAHEEN and T. LIPMAN, ‘REDUCING GREENHOUSE EMISSIONS

AND FUEL CONSUMPTION : Sustainable Approaches for Surface
Transportation’, IATSS Res., 2007.
[355] A. Bull, Traffic Congestion - The Problem and How to Deal with it? 2004.
[356] American Public Transportation Association, ‘Public Transportation Reduces

Greenhouse Gases and Conserves Energy’, p. 6, 2008.
[357] R. Eberhart and J. Kennedy, ‘New optimizer using particle swarm theory’, in
Proceedings of the International Symposium on Micro Machine and Human
Science, 1995.
[358] J. Kennedy and R. Eberhart, ‘Particle swarm optimization’, in IEEE International

Conference on Neural Networks - Conference Proceedings, 1995.
[359] J. Torres-Jiménez and J. Pavón, ‘Applications of metaheuristics in real-life

problems’, Prog. Artif. Intell. 2014 24, vol. 2, no. 4, pp. 175–176, Apr. 2014.
[360] H. Faris, I. Aljarah, M. A. Al-Betar, and S. Mirjalili, ‘Grey wolf optimizer: a

review of recent variants and applications’, Neural Comput. Appl., vol. 30, no. 2,
pp. 413–435, 2018.
[361] S. Mirjalili, J. S. Dong, A. Lewis, and L. Reviews, Nature- Inspired Optimizers. .
[362] S. Kirkpatrick, ; C D Gelatt, and ; M P Vecchi, ‘Optimization by Simulated

Annealing’, New Ser., vol. 220, no. 4598, pp. 671–680, 1983.
[363] ‘Genetic Algorithms on JSTOR’. [Online]. Available:

https://www.jstor.org/stable/24939139. [Accessed: 07-Jul-2021].

[364] R. Storn and K. Price, ‘Differential Evolution – A Simple and Efficient Heuristic
for global Optimization over Continuous Spaces’, J. Glob. Optim. 1997 114, vol.
11, no. 4, pp. 341–359, 1997.
[365] S. Mirjalili, ‘Ant Colony Optimisation’, Stud. Comput. Intell., vol. 780, pp. 33–42,
2019.
[366] J. Kennedy and R. Eberhart, ‘Particle swarm optimization’, Proc. ICNN’95 - Int.
Conf. Neural Networks, vol. 4, pp. 1942–1948.
[367] M. Clerc, ‘Discrete Particle Swarm Optimization, illustrated by the Traveling

Salesman Problem’, pp. 219–239, 2004.
[368] S. Saremi, S. Mirjalili, and A. Lewis, ‘How important is a transfer function in

discrete heuristic algorithms’, Neural Comput. Appl. 2014 263, vol. 26, no. 3, pp.
625–640, Oct. 2014.
[369] S. Mirjalili and A. Lewis, ‘S-shaped versus V-shaped transfer functions for binary
Particle Swarm Optimization’, Swarm Evol. Comput., vol. 9, pp. 1–14, Apr. 2013.
[370] C. A. Coello Coello, G. T. Pulido, and M. S. Lechuga, ‘Handling multiple

objectives with particle swarm optimization’, IEEE Trans. Evol. Comput., vol. 8,
no. 3, pp. 256–279, Jun. 2004.
[371] S. Mostaghim and J. Teich, ‘Strategies for finding good local guides in multi-
objective particle swarm optimization (MOPSO)’, 2003 IEEE Swarm Intell. Symp.
SIS 2003 - Proc., pp. 26–33, 2003.
[372] M. R. Sierra and C. A. C. Coello, ‘Improving PSO-Based Multi-objective

Optimization Using Crowding, Mutation and ∈-Dominance’, Lect. Notes Comput.
Sci., vol. 3410, pp. 505–519, 2005.
[373] X. Li, ‘A Non-dominated Sorting Particle Swarm Optimizer for Multiobjective

Optimization’, Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell.
Lect. Notes Bioinformatics), vol. 2723, pp. 37–48, 2003.
[374] M. Reyes-Sierra and C. A. Coello Coello, ‘Multi-Objective Particle Swarm

Optimizers: A Survey of the State-of-the-Art’, Int. J. Comput. Intell. Res., vol. 2,
no. 3, pp. 287–308, 2006.

[375] K. Parsopoulos, M. V.-I. T. and, and undefined 2002, ‘Particle swarm optimization
method for constrained optimization problems’, mat.uab.cat.
[376] J. C. F. Cabrera and C. A. C. Coello, ‘Handling Constraints in Particle Swarm

Optimization Using a Small Population Size’, Lect. Notes Comput. Sci. (including
Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 4827 LNAI, pp.
41–51, 2007.
[377] C. C.-C. methods in applied mechanics and and undefined 2002, ‘Theoretical and
numerical constraint-handling techniques used with evolutionary algorithms: a
survey of the state of the art’, Elsevier.
[378] R. Eberhart, Y. S.-P. of the 2001 congress on, and undefined 2001, ‘Tracking and
optimizing dynamic systems with particle swarms’, ieeexplore.ieee.org.
[379] S. Mirjalili, A. Lewis, S. M.-I. Sciences, and undefined 2015, ‘Confidence

measure: a novel metric for robust meta-heuristic optimisation algorithms’,
Elsevier.
[380] F. Luan, J. Choi, H. J.-I. transactions on Magnetics, and undefined 2012, ‘A

particle swarm optimization algorithm with novel expected fitness evaluation for
robust optimization problems’, ieeexplore.ieee.org.
[381] S. Mirjalili, A. Lewis, J. D.-S. and E. Computation, and undefined 2018,

‘Confidence-based robust optimisation using multi-objective meta-heuristics’,
Elsevier.
[382] S. Ono, Y. Yoshitake, S. N.-A. L. and Robotics, and undefined 2009, ‘Robust
optimization using multi-objective particle swarm optimization’, Springer, vol. 14,
no. 2, pp. 174–177, Nov. 2009.
[383] S. Huang, N. Tian, Y. Wang, and Z. Ji, ‘Particle swarm optimization using multi-
information characteristics of all personal-best information’, Springerplus, vol. 5,
no. 1, Dec. 2016.
[384] ‘Particle Swarm Optimization’.
[385] A. Thevenot, ‘Particle Swarm Optimization (PSO) Visually Explained | by ⭐Axel

Thevenot | Towards Data Science’. [Online]. Available:

https://towardsdatascience.com/particle-swarm-optimization-visually-explained-
46289eeb2e14. [Accessed: 03-Apr-2022].
[386] S. Mirjalili, J. Song Dong, A. Lewis, and A. S. Sadiq, ‘Particle swarm

optimization: Theory, literature review, and application in airfoil design’, in
Studies in Computational Intelligence, vol. 811, Springer Verlag, 2020, pp. 167–
184.

Chapter 8: Appendices
8 APPENDICES
8.1 Appendix A: PSO Algorithm

Particle swarm optimisation (PSO) is a stochastic optimisation technique that focuses on
swarm intelligence. It was introduced in 1995 by Eberhart and Kennedy [357], [358].
PSO simulates the social behaviour of animals such as insects, herds, birds, and fish.
These swarms work cooperatively to hunt food, and each swarm member modifies its
search behaviour in response to its own and other swarm members' learning experiences.
The core design principle of the PSO algorithm is closely linked to two studies: One is an
evolutionary algorithm; similar to evolutionary algorithms, PSO operates in a swarm
mode, which enables it to simultaneously scan a large part of the solution space for the
optimised objective function.
Meta-heuristics are high-level strategies for employing heuristics to solve a wide variety
of problems Examples include complex scheduling problems, space allocation problems,
and clustering problems [359]. In contrast to heuristics, they do not require any problem-
dependent heuristic expertise. One of their key advantages is that they avoid making bold
conclusions about the problem instead of treating it as a black box [360] [361]. They
sample the problem's search space, which is too large to explore exhaustively, in order to
find optimal solutions. Meta-heuristics are a subclass of Soft Computing approaches that
permit incomplete or inaccurate solutions. This comes at the cost of inaccuracy when

selecting the ideal solution to a given problem. The tremendous complexity of some
issues and the failure of exact approaches explain the adoption of such problem-solving
procedures that quickly find optimal or near-optimal solutions. Numerous algorithms
have been created in the field of meta-heuristics, with the majority of them being inspired
by nature. Several of the most popular is Simulated Annealing [362], which mimics the
annealing mechanism in physics. Genetic Algorithms [363] which are inspired by
evolutionary theory. Differential Evolution [364], which incorporates evolutionary
principles into differential equations, Ant Colony Optimization (ACO) [365], which is
inspired by ant swarm intelligence, and Particle Swarm Optimization (PSO) [366].
The PSO has seen many revisions in the literature recently. Importantly, this algorithm
has varied versions. Recently, the PSO algorithm models have incorporated many
transfer functions to solve binary issues [367]. These functions use the velocity values to
flip a binary bit. The larger the velocity vector, the more likely it is that the binary bit will
flip [368], [369] [367].
Numerous multi-objective versions of PSO have been proposed to address multi-

objective issues in the literature. In [370], the authors proposed the most extensively used
method, in which non-dominated solutions (Pareto optimum solutions) were stored and
developed during the optimisation process using an index. The archive's global best
solution was chosen for each iteration, as all solutions in the archive are considered
solutions to a multi-objective optimisation problem. A grid technique was applied in the
archive to increase the solutions' heterogeneity [371]. Additionally, there are methods for
increasing diversity, such as crowding distance [372]. Another often-used strategy is
ranking and boosting previously obtained non-dominated solutions using non-dominated
sorting in the PSO [373]. Additionally, there are algorithms that aggregate objectives in
the literature [374].
Numerous functions for managing constraints are available in the literature [375],
[376]. These functions are occasionally used with the goal function to penalise particles
that breach constraint violations. Some of these functions apply the same penalty to
particles regardless of the severity of the infraction. In this context, such functions are
referred to be barrier functions. Additionally, there exist functions that penalise particles
based on the gravity of their violation. These functions are crucial when addressing

problems characterised by dominating infeasible zones, as PSO iterates through a large

number of infeasible solutions [377].
In [378], the authors proposed a dynamic version of PSO for solving situations with
quickly changing objectives. The fundamental purpose is to prevent PSO from
converging on a solution in order to track changes in the objective function(s) [378].
Additionally, robust PSOs [379], [380] and robust MOPSOs [381], [382] have been
proposed to handle problems involving uncertainty in the literature.
8.1.1 PSO Basic Modelling Concepts

The PSO algorithm was inspired by the flocking behaviour of birds in nature, as
previously stated. Each particle is seen as a solution to a specific optimisation problem in
this method. It's made up of two vectors: one representing the position and the other
representing velocity. The values of all the variables in the problem are stored in the
position vector. The position vectors of the particles will be two-dimensional if the
problem has two parameters. In an n-dimensional search space, where n is the number of
variables, each particle has total freedom of movement. The particle's location is tracked
using the second vector (velocity). This vector specifies the quantity and direction of each
dimension's and particle's step size [383].
At each optimisation step, the following equation is used to update the location of
particles [384]:
⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝐗 𝐢 (𝐭 + 𝟏) = 𝐗 ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗ ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝐢 (𝐭) + 𝐕𝐢 (𝐭 + 𝟏) (8.1)
Where ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗ ⃗⃗⃗⃗⃗⃗⃗⃗⃗

𝑋𝚤 (𝑡) denotes the position of 𝑖_𝑡ℎ particle at 𝑡_𝑡ℎ iteration and 𝑉 𝚤 (𝑡) , denotes the
velocity of 𝑖_𝑡ℎ particle at 𝑡_𝑡ℎ iteration. This equation demonstrates how simple it is to
update position using the velocity vector as the primary variable. The velocity vector is
defined as follows [383]:
⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝐕𝐢 (𝐭 + 𝟏) = 𝐰𝐕 ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗ ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗ ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗ ⃗⃗⃗⃗⃗⃗⃗⃗⃗ ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝐢 (𝐭) + 𝐜𝟏 𝐫𝟏 (𝐏𝐢 (𝐭) − 𝐱 𝐢 (𝐭)) + 𝐜𝟐 𝐫𝟐 (𝐆(𝐭) − 𝐱 𝐢 (𝐭)) (8.2)
where ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝑋𝚤 (𝑡) denotes the i_th particle's position during the 𝑡_𝑡ℎ iteration, ⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝑉𝚤 (𝑡) denotes the
i_th particle's velocity during the 𝑡_𝑡ℎ iteration, w denotes the inertial weight, 𝑐1 denotes
the individual coefficient, 𝑐2 denotes the social coefficient, 𝑟1 𝑟2 denote random numbers
in the range [0, 1], ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗

𝑃𝚤 (𝑡) , denotes the best solution obtained by the 𝑖𝑡ℎ particle.

As defined in (4.7), the velocity vector is composed of three parts. The first element,
⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝑤𝑉 𝚤 (𝑡), preserves the trend of the present velocity. This factor is multiplied by a value for
the inertia parameter w. The bigger this parameter's value, the more likely it is that the
⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
prior velocity will be maintained. In the next part, 𝑐1 𝑟1 (𝑃 ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝚤 (𝑡) − 𝑥𝚤 (𝑡)), simulates the
individual intelligence of a bird by memorising and employing each particle's best answer
thus far. The vector ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
𝑃𝚤 (𝑡), is changed after each iteration in case the ith particle identifies
a more efficient solution. By adjusting 𝑐1, the effect of this variable on the final velocity
value can be enhanced or lowered. Due to the stochastic nature of PSO, this parameter is
multiplied by a random value in the range [0, 1] to generate randomised behaviour. By
and large, the second component preserves a bias towards the particle's best solution thus
⃗⃗⃗⃗⃗⃗⃗⃗⃗ − ⃗⃗⃗⃗⃗⃗⃗⃗⃗⃗
far, the so-called "personal best". The third part, 𝑐2 𝑟2 (𝐺(𝑡) 𝑥𝚤 (𝑡)), replicates the social
intelligence of a flock of birds by storing and utilising the optimal solution obtained by
all particles in ⃗⃗⃗⃗⃗⃗⃗⃗⃗

𝐺(𝑡). This means that contemplating the optimal solution for the swarm
draws all particles to a single spot. Additionally, this component's impact can be adjusted
via 𝑐2 [379], [381].
Figure 8-1 In PSO, each particle considers its prior velocity, personal best, and global
best to define its current velocity and position (Adopted from [386]).
The three Cartesian components can be used to describe the next position of a particle.
This principle is illustrated in Figure 8-1. This illustration explains how each particle

determines its current velocity and position by comparing it to its prior velocity, personal
best, and global best. The position immediately following is significantly reliant on the
random integers generated by 𝑟1 and 𝑟2 . In Figure 8-1, it is assumed that the particle
considers 50% of its current velocity, 40% of the global best, and 40% of the personal
best [385]. 𝑐1 and 𝑐2 are, in reality, multiplied by random numbers. The values of w, c1,
and c2 must be known in order to identify the zone in which a particle can move. In the
most generally used PSO variant, the value of w drops linearly from 0.9 to 0.4 in relation
to the number of iterations [386]. Additionally, the c1 and c2 parameters are set to 2. Due
to the difficulties of illustrating the different places of particles when the inertial weight
is changed, this paragraph illustrates probable locations when w=1, w=0.5, and w=0. Fig.
2 illustrates the new places that the particle could take in this form.

Figure 8-2 The inertial weight's (w) effect on the velocity vector in PSO (Adopted from
[386]).
The vectors toward the personal and global bests can be any length between 0 and twice
their respective distances. This is due to the fact that both c1 and c2 have been multiplied
by a random number between 0 and 1 [386]. As a result, the range of these two parameters
is unlimited inside the range [0,2]. When the value is 0, the component is ignored by the
particle. When the variable's value equals 2, it is considered twice as large. This is why
the vectors in Figure 8-2 have been multiplied by two. Due to the fact that the lower and
upper bounds of c1 and c2 remain constant, the greatest and minimum distances to
personal and global bests are equal, as illustrated in Figure 8-2. However, as illustrated
in this image, the inertial weight is adjusted. As illustrated in Figure 8-2, when the inertial
weight is 0, the particle's next position is somewhere between its current location and its
personal and global bests. Due to the fact that the algorithm searches locally within the
region defined by the current location, personal best, and global best, the PSO model
incorporates exploitation and local search. Indeed, when both c1 and c2 are set to 1, just
the local search is conducted. This principle is illustrated in Figure 8-3.
Figure 8-3 When w=0, c1 =1 and c2 =1, exploitation and local search performance in PSO
(Adopted from [386]).
Exploration is maximised, and global search performance increases correspondingly to

both of these parameters' values. This is depicted in Figure 8-4. As demonstrated, the

particle has a proclivity for travelling outside the region indicated by the global or local
bests. As a result, discovery and global search are enabled. It is worth mentioning that
both personal and global bests are updated continuously. As a result, the shaded area in
Figure 8-4 shows the area of the next position. c1 =1 and c2 =1, exploitation and local
search performance in PSO.
Figure 8-4 Exploration and global search rise proportionately to w, c1 and c2 values
(Adopted from [386]).
The PSO algorithm is based on these simple concepts and is used to find the optimal
global solution to a given optimisation issue. It begins with a population of randomly
picked solutions. It then continues the steps below till the end condition is met:
1. Determination of each particle's objective value
2. w, c1 and c2 are updated. If both c1 and c2 remain constant, just w is altered.
3. Managing personal and global records
4. Determine the velocity vector for each particle using Eq. 8.1.
5. Determine the next position of each particle using Eq. 8.2.
The final optimum for the optimisation problem will be the best solution obtained by the
entire swam at the conclusion of the optimisation process [333], [383]–[385], [386].

8.2 Appendix B: Supplementary Data for Case Study
8.2.1 Average Lagos BRT driving (minutes)

Average Cycle time
S/n Service Lines (min)
1 Ikorodu-Mile12 76
2 Ikorodu-Fadeyi 132
3 Ikorodu-TBS 164
4 Aruna-TBS 160
5 Agric-Mile12 66
6 Agric-Fadeyi 127
7 Agric-TBS 159
8 Agric-Maryland 99
9 Agric-Costain 157
10 Ogolonto-TBS 189
11 Irawo-TBS 177
12 Idera-TBS 176
13 Owode-Ikorodu 72
14 Mile12-Fadeyi 82
15 Mile12-TBS 139
16 Ketu-Ikorodu 102
17 Ketu-Fadeyi 69
18 Ketu-TBS 126
19 Ojota-Ikorodu 90
20 Ojota-Fadeyi 69

21 Ojota-TBS 125
22 Maryland-Ikorodu 115
23 Maryland-Fadeyi 67
24 Maryland-TBS 117
25 Anthony-Ikorodu 124
26 Fadeyi-TBS 74
27 Barracks-Ikorodu 140
28 Barracks-Mile12 85
29 Barracks-TBS 69
30 Costain-Ikorodu 165
31 Costain-Mile12 105
32 Leventis-Ikorodu 207
33 Leventis-Mile12 149
34 CMS-Ikorodu 204
35 CMS-Mile12 139
36 CMS-Fadeyi 62

8.2.2 Lagos BRT weekdays operational Schedule
8.2.3 Lagos BRT Saturday operational Schedule

8.2.4 Lagos BRT Sunday operational Schedule

8.2.5 Lagos BRT Corridor Lanes

Ovel Design of Charging Infrastructure For Electrified Bus Transit Network Systems

Uploaded by

Copyright:

Available Formats

Ovel Design of Charging Infrastructure For Electrified Bus Transit Network Systems

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ovel Design of Charging Infrastructure For Electrified Bus Transit Network Systems

Uploaded by

Copyright:

Available Formats

NOVEL DESIGN OF CHARGING

INFRASTRUCTURE FOR ELECTRIFIED BUS

TRANSIT NETWORK SYSTEMS

Adedayo Ayodeji Asaolu

Institute of Energy and Environment

Department of Electronic and Electrical Engineering

This thesis is submitted for the degree of Doctor of Philosophy

To my wife Mary and my children Enoch, Nathaniel and Joana

Date: August 02, 2022

This novel charging optimisation tool is developed to demonstrate how multiple

First, I would like to express my gratitude to Professor Stuart Galloway, my academic

Additionally, I would like to express my profound gratitude to the Nigerian Petroleum

It is difficult to find the appropriate words to articulate my heartfelt appreciation for my

2.9 ELECTRIC VEHICLES CHARGING DEMONSTRATION WITH MICROGRIDS AND OFF-

TABLE 2-2 CHARACTERISTICS OF ELECTRIC TRANSIT BUSES .......................................... 33

TABLE 2-3 COMPARATIVE EVALUATION OF HEB, FCEB AND DB ( [12]) ........................ 34

TABLE 4-2 AVERAGE BEBS ENERGY CONSUMPTION BASED ON TRAFFIC FLOW

TABLE 4-5 SIMULATION RESULT FOR THE SCENARIO I ................................................... 133

TABLE 5-1 BEB PARAMETER (ADOPTED FROM [101]) ................................................... 151

TABLE 5-2 RESULT OF ALLOCATION OF INDUCTIVE SIMULATION ................................... 152

TABLE 5-3 SENSITIVITY ANALYSIS OF THE MULTI-TERMINAL CONFIGURATION OF

INDUCTIVE CHARGING TRANSIT DESIGN ................................................................. 157

FIGURE 2-1 STRATEGIES FOR SUSTAINABLE BRTS [137] ................................................. 24

FIGURE 2-2 EL METROPOLITANO'S TRUNK CORRIDOR [148] ........................................... 26

FIGURE 2-3 BUS TERMINAL IN OSHODI, LAGOS [151] ...................................................... 27

FIGURE 2-4 LAGOS BRT BUS STOP [154] ......................................................................... 28

FIGURE 2-5 LAGOS BRT BUS DEPOT [156] ...................................................................... 28

FIGURE 2-6 A TYPICAL BOARDING PLATFORM FOR EASY ENTRY/EXIT [157]..................... 29

FIGURE 2-7 TRANSMILENIO PREPAID BOARDING AND PASSENGER INFORMATION IN BOGOTA

FIGURE 2-9 BATTERY ELECTRIC BUS CONFIGURATION (ADOPTED FROM [12])............... 33

FIGURE 2-10 BEB CHARGING TECHNOLOGIES ................................................................. 44

FIGURE 2-11 TRANSIT INDUCTIVE SYSTEM (ADOPTED FROM [101])................................. 46

FIGURE 2-12 PANTOGRAPH SYSTEM (SOURCE: [227]) ..................................................... 48

FIGURE 3-2 TRANSIT SCHEDULE FOR ASSIGNMENTS D ...................................................... 70

FIGURE 3-7 MULTI-TERMINAL ROUTE BASED CHARGING MODEL ..................................... 87

FIGURE 3-8 MODEL OF TWO-TRANSIT TERMINAL ............................................................. 92

FIGURE 3-9 ALLOCATION OF INDUCTIVE CABLE LENGTH ................................................. 94

FIGURE 3-10 BRT ROUTE WITH BUSES ............................................................................. 96

FIGURE 3-12 PROPOSED BRT INDUCTIVE CHARGING SYSTEM .......................................... 98

FIGURE 3-13 MATHEMATICAL MODEL TRANSIT INDUCTIVE SYSTEM ................................ 99

FIGURE 3-14 ALLOCATION OF INDUCTIVE CABLE ........................................................... 101

FIGURE 4-1 MODELLING ELEMENTS AND SIMULATION OVERVIEW ................................. 105

FIGURE 4-3 ADVISOR VEHICLE INPUT WINDOW ........................................................... 112

FIGURE 4-4 ADVISOR SIMULATION SETUP WINDOW .................................................... 112

FIGURE 4-5 ADVISOR RESULTS WINDOW ..................................................................... 113

FIGURE 4-6: FREE TRAFFIC (AVERAGE SPEED: 33KM/H) ................................................. 116

FIGURE 4-7 MOVE TRAFFIC (AVERAGE SPEED: 21KM/H) ................................................ 117

FIGURE 4-8: APPROPRIATE TRAFFIC (AVERAGE SPEED: 18KM/H).................................... 117

FIGURE 4-9: INAPPROPRIATE TRAFFIC (AVERAGE SPEED: 12KM/H)................................. 118

FIGURE 4-10 CRITICAL TRAFFIC (AVERAGE SPEED: 7KM/H) ........................................... 118

FIGURE 4-11 THE ROUTE ELEVATION PROFILE [332] ...................................................... 119

FIGURE 4-12 RESULT (FREE TRAFFIC) ............................................................................ 120

FIGURE 4-13 PSEUDOCODE OF THE PSO ALGORITHM ..................................................... 124

FIGURE 4-14 GRAPHICAL REPRESENTATION EXPLAINS THE CONSTRAINT-HANDLING

MECHANISM INCORPORATED INTO OUR PSO ALGORITHM (ADAPTED FROM [339]).126

FIGURE 4-16 SCENARIO I, TRADE-OFF ANALYSIS ........................................................... 134

PRIORITY (50KW) SCENARIOS ................................................................................ 137

PRIORITY (200KW) SCENARIOS .............................................................................. 139

PRIORITY (400KW) SCENARIOS .............................................................................. 140