Clemson University

TigerPrints
All Theses Theses

12-2015

Power Quality Study of a Microgrid with Nonlinear

Composite Load and PV Integration
Zhanhe Liu
Clemson University, [email protected]

Follow this and additional works at: https://tigerprints.clemson.edu/all_theses

Recommended Citation
Liu, Zhanhe, "Power Quality Study of a Microgrid with Nonlinear Composite Load and PV Integration" (2015). All Theses. 2459.
https://tigerprints.clemson.edu/all_theses/2459

This Thesis is brought to you for free and open access by the Theses at TigerPrints. It has been accepted for inclusion in All Theses by an authorized
administrator of TigerPrints. For more information, please contact [email protected].
POWER QUALITY STUDY OF A MICROGRID WITH NONLINEAR
COMPOSITE LOAD AND PV INTEGRATION

A Thesis
Presented to
the Graduate School of
Clemson University

In Partial Fulfillment
of the Requirements for the Degree
Master of Science
Electrical Engineering

by
Zhanhe Liu
December 2015

Accepted by:
Dr. Elham Makram, Committee Chair
Dr. Keith Corzine
Dr. John Gowdy
 ABSTRACT

Harmonics distortion is a crucial problem in microgrid. Harmonic sources can be

categorized as two main factors: renewable energy integration and nonlinear loads. Both

factors are investigated in this thesis. For renewbale energy, photovoltaic (PV) power is

one of the most effective solutions for energy crisis and it is showing great potential for

serving customers in microgrid. A three phase PV source model is established from both

mathematical equations and power electronic control schemes. A composite load model

by Crossed Frequency Admittance Matrix theory is illustrated and built as well. Due to

the fact that microgrid should be able to run under two different operating modes: grid-

connected and stand-alone, energy storage devices are considered as neccesity. Therefore

the energy storage with droop control is included in this thesis. A practicdal microgrid

loacated at GA, USA is used as a study system. Instead of making the ideal assumption,

the unbalanced feeder structure and historical meteorological data are considered in the

study. The microgrid, PV model, nonlinear load model and energy storage are simulated

in MATLAB/Simulink environment. Multiple PV sources are integrated at different

locations in order to observe the impact of harmonics on the microgrid and power quality

(PQ). The results show the impact of installing PV sources in both grid-connected mode

and stand-alone mode considering linear and composite nonlinear loads. In addition,

three PQ indices are discussed to demonstrate the numerical impacts with various

perspectives. Furthermore, the mitgation of harmonics is developed by adding a active

power filter on energy storage devices in the stand-alone mode.

ii
 ACKNOWLEDGMENTS

I would like to thank my father Suyu and my mother Xinrong for their supporting

and encouraging during this endeavor. Also, I would like to give thanks to my

advisor Prof. Elham Makram for she has been providing precious academic help and

guidance. Thank Clemson University Electric Power Research Association (CUEPRA)

for funding this program. Members of CUEPRA are: Advanced Cable-bus Co.,

ALSTOM, Duke-Energy, South Carolina Electric and Gas (SCE&G), and Santee-

Cooper.

iii
 TABLE OF CONTENTS

Page

TITLE PAGE .................................................................................................................... i

ABSTRACT ..................................................................................................................... ii

ACKNOWLEDGMENTS ..............................................................................................iii

LIST OF TABLES .......................................................................................................... vi

LIST OF FIGURES ....................................................................................................... vii

CHAPTER

I. INTRODUCTION AND LITERATURE REVIEW ..................................... 1

1.1 Introduction ........................................................................................ 1

1.2 Power System Harmonics .................................................................. 3
1.3 Photovoltaic Source ........................................................................... 6
1.4 Nonlinear Load .................................................................................. 8
1.5 Microgrid Characteristic .................................................................. 10

II. MODELING OF MICROGRID .................................................................. 14

2.1 Photovoltaic Source ......................................................................... 14

2.1.1 Photovoltaic Source Modeling ................................................ 14
2.1.2 Inverter Control of Photovoltaic Source ................................ 17
2.1.3 Simulation of Photovoltaic Sources ........................................ 18
2.2 Nonlinear Composite Load .............................................................. 23
2.2.1 Nonlinear Load Modeling Method ......................................... 23
2.2.2 Individual Nonlinear Load Types ........................................... 24
2.2.3 Simulation Model Building..................................................... 30
2.3 Energy Storage ................................................................................. 32
2.4 Active Power Filter .......................................................................... 34
2.4.1 Structure of Active Power Filter ............................................. 34
2.4.2 Simulation of Active Power Filter .......................................... 35
2.5 Summary .......................................................................................... 37

iv
Table of Contents (Continued)

Page
III. HARMONICS IMPACTS ON MICROGRID............................................. 39

3.1 Power Quality Indices ...................................................................... 39

3.2 Test System Information .................................................................. 41
3.3 Harmonics Impacts on Grid Connected Mode................................. 44
3.3.1 Considering PV with Linear Load .......................................... 44
3.3.2 Considering PV with Nonlinear Composite Load .................. 48
3.4 Harmonics Impacts on Stand-Alone Mode ...................................... 52
3.5 Summary .......................................................................................... 57

IV. CONCLUSION AND FUTURE WORK .................................................... 58

APPENDICES ............................................................................................................... 64

A: Crossed Frequency Admittance Matrix of Composite Load ....................... 65

B: Line Data Calculation of the Study System ................................................. 68
C: Line Data Calculation of the Study System ................................................. 71

REFERENCES .............................................................................................................. 61

v
 LIST OF TABLES

Table Page

2.1 Parameters of KC200GT Solar Array at Nominal Operating Condition ..... 15

3.1 Load Data of the Study System ................................................................... 43

3.2 Composite Load Information ....................................................................... 51

3.3 Distortion Power Index ................................................................................ 51

3.4 Waveform Distortion Index ......................................................................... 52

vi
 LIST OF FIGURES

Figure Page

1.1 Simplified PV Cell Circuit ............................................................................. 6

1.2 Current Waveform of DBR ............................................................................ 8

1.3 Typical Microgrid ........................................................................................ 12

2.1 Boost Converter with MPPT........................................................................ 16

2.2 PV System Including Inverter and Its Control Scheme ............................... 18

2.3 I-V Curve of Single PV Cell under Different Solar Irradiance ................... 19

2.4 I-V Curve of Single PV Cell under Different Temperature ......................... 19

2.5 P-V Curve of Single PV Cell under Different Solar Irradiance................... 20

2.6 P-V Curve of Single PV Cell under Different Temperature ........................ 20

2.7 Active Power Output of PV Source System at Nominal Condition ............ 21

2.8 Output Power at Phase A of PV Source System at Nominal Condition ...... 22

2.9 A Closer Look at Output Power of Phase A of PV Source System at Nominal

Condition...................................................................................................... 22

2.10 Composite Load Structure ........................................................................... 25

2.11 Simplified Circuit of Composite Load: (a) CFL; (b) DBR; (c) PAVC; (d)
Structure ....................................................................................................... 26

2.12 Simulation of CFL ....................................................................................... 27

2.13 Admittance Matrix Magnitude Distribution of CFL .................................... 27

2.14 Simulation of DBR ...................................................................................... 28

2.15 Admittance Matrix Magnitude Distribution of DBR ................................... 29

2.16 Simulation of PAVC .................................................................................... 29

vii
List of Figures (Continued) Page

2.17 Admittance Matrix Magnitude Distribution of PAVC ................................. 30

2.18 Composite Load Simulation Scheme Diagram ............................................ 31

2.19 Droop Control Structure .............................................................................. 33

2.20 Active Power Filter Structure ...................................................................... 35

2.21 Current Waveform (at 0.2s the active power filter is connected) ................ 36

2.22 FFT Spectrum of Source Current without Active Power Filter ................... 36

2.23 FFT Spectrum of Source Current with Active Power Filter ........................ 37

3.1 Test System Structure ................................................................................. 42

3.2 Meteorological Data of July Typical Day ................................................... 43

3.3 WSHDL from 7:00AM to 6:00PM (Case I) ............................................... 45

3.4 THD from 7:00AM to 6:00PM (Case I) ..................................................... 45

3.5 WSHDL from 7:00AM to 6:00PM (Case II) .............................................. 46

3.6 THD from 7:00AM to 6:00PM (Case II) .................................................... 47

3.7 Load Current of Bus16A with Nonlinear Load .......................................... 48

3.8 THD Comparison with Different Loads with PV Injection ........................ 49

3.9 Simplified Study System ............................................................................. 49

3.10 Battery Rated Output Active Power ............................................................ 53

3.11 Active Power Curve of Microgrid in Stand-alone Mode ............................. 53

3.12 THD of Battery Current Output from 7:00AM to 6:00AM ......................... 54

3.13 WSHDL of Study System from 7:00AM to 6:00PM in Stand-alone Mode 56

3.14 THD of Study System from 7:00AM to 6:00PM in Stand-alone Mode ...... 56

viii
 CHAPTER ONE

INTRODUCTION AND LIETERATURE REVIEW

1.1 Introduction

Microgrids are considered as viable options for those places where main grid

expansion is either impossible or has no economic justification, such as the electrification

in university campuses, military installations and rural villages[1]. Some researchers

regard the microgrid as a specific distribution system embedded with distributed

generations, which may operate in grid-connected or islanded (stand-alone) mode [2].

Even though there are some common characteristics between distribution system and

microgrid, for example, voltage level and customer types, microgrid has its own unique

structure. Microgrid consists of renewable sources, backup generators, energy storage,

unbalanced load demand on each phase and multiphase lines. Due to this special structure,

both grid-connected mode and stand-alone mode, as one of the most important features of

microgrid, are able to deliver power whether there is a power outage on grid side or not.

Along with this beneficial advantage, there are potential challenges to keep the system

running safely and well. One of main concerns in the microgrid study is the electrical

power quality issue. From large amounts of experience, it is very necessary to assess

electrical power quality for the sake of devices and users. For example, it has been

reported that a 10% increase in voltage stress caused by harmonic current typically results

in 7% increase in the operating temperature of a capacitor bank and can reduce its life

expectancy by 30% [3]. For the purpose of saving lifetime of electric devices and

1
providing more reliable power to customers, harmonics study becomes significant for

researchers.

Scholars who study power system used to ignore the inside functioning process of

components in the large system since what the output brings are far more important.

However, the situation is different. Because of the integration of renewable energy, the

output of electrical sources are not conventional. Aiming to understand what can be the

output of renewable energy sources, looking into the detailed model of each components

can be a solution. Particularly, photovoltaic source is often integrated to local low voltage

level microgrid and should be studied. Given the small scale of loads and generation in

micro-grid, the amount of harmonics produced by PV will cause more significant

influences than in conventional large grid, such as overheating electronic devices, poor

power quality and loss of power. Apparently not just consumers but also utility companies

would like to find a solution to minimize those negative influences. However, before

researchers can actually find a solution for this problem, how to model, how to analyze

and how to quantify harmonics in microgrid should be accomplished first.

Alternating current (AC) power, started since 1886 in North America, is the major

form in electrical grid system. It operates in sinusoidal waveform to transfer electric

power. In the United States, the frequency of it is 60 Hz. According to its own

characteristic, every electronic devices connected to AC power is designed to function

well under clean electric power. Clean power means whether voltage or current only

contains components in 60 Hz. However, recent advances in technology have made the

question of AC power quality even more important [4].

2
 Human beings have been using coal as a source of electric power for over a

hundred years and the pollution it brought to our society has alerted governments around

the world. In order to reduce the pollution, green and renewable energy is considered as

one of solutions. In a single day, enough sun shines in China to meet its energy needs for

more than 10 years, at least theoretically [5]. Huge potential development stimulates both

researchers and utilities working on integration renewable energy to established power

system. Yet challenges are coming along with potential benefits.

Different origins can cause same results. While innovative electric devices, for

example personal computers, have induced concerns on power quality, integration of

renewable energy leads to some negative influences on power quality. Electric Power

Quality (EPQ) is a term that refers to maintaining the near sinusoidal waveform of power

distribution bus voltages and currents at rated magnitude and frequency [3].

1.2 Power System Harmonics

Harmonics was not firstly used in electric power system, but in acoustics. In

electric power system, this term “harmonics” represents a component of which frequency

is a certain number multiple of the fundamental frequency. It can exist both in voltages

and currents. So for a h order harmonic component, its frequency can be expressed as:

f h  f fundamental  h (1.1)

where h is the order of harmonic component and ffundamental is the fundamental frequency

of the system. Ideally power system in the United States is running at 60 Hz. However,

3
because of a mixture of reasons, harmonic components always exist in voltages and

currents. Measured data shows, in time domain, voltage or current waveform is a

superposition of fundamental frequency and different harmonics components. To apply

Fast Fourier Transformation (FFT), the signal can be analyzed in frequency domain.

Due to the fact that most of time a harmonic signal contains more than one

frequency or one order harmonic signal. In order to quantify the harmonics of electrical

parameters, including voltages and currents, Total Harmonic Distortion (THD) is defined.

X
h2
h
2

THD 
X1 (1.2)

where X represents variable name. It considers the contribution of every individual

harmonic component on the signal [6].

In power system, it is very necessary to monitor THD value. Majority of electrical

components and devices in grid are designed for sinusoidal-wave currents and voltages in

a certain rated frequency. Currents and voltages in different higher order frequency can

harm components and shorten the life time. Another influence of harmonics is that they

can overheat the devices, such as transformers, so that more losses are brought up. Given

these negative impacts of harmonics in power system, researchers have been working on

the analysis and solutions for harmonics.

This thesis focuses on microgrid study, instead of high voltage conventional

power system or transmission line. However, there are some common characteristic of

them in general power system field. Compared to static power flow study, harmonic

research requires different modeling method which can accurately describe the behavior

4
of components in different frequencies. Simulation of the harmonic interaction between

detailed models of the synchronous generator, power transformer and transmission

system is achieved with the development of a unified multi-frequency domain equivalent

[7]. Also resonance phenomena could exist in transmission line because capacitors and

inductors have different reactance in higher frequency condition. Resonance happens

when the reactance of capacitor and inductor cancel out. In FFT harmonic spectrum, the

current near resonant frequency is much more dominant than those in other frequencies.

Because of the potential possibility of resonance, a seemingly small harmonic injection at

one location on the system causes significant problems some distance away such as

telephone interference [8].

There are several effective solutions for compensating harmonics in power system,

and among them active filters (AF) are highly focused by scholars: novel structures of

active filter can have lower voltage rating, smaller size inductor and lower

electromagnetic interference [9] [10]; with conventional concepts it takes one period of

fundamental frequency to eliminated the harmonics after the load current has changed but

newly built control scheme allows much faster harmonics reduction[11].

However, to design more suitable solutions for various conditions, it is important

to analyze the origins of harmonics in power system, which can be classified as two

major types: renewable energy and nonlinear loads. In this paper, only solar energy is

considered for renewable energy.

5
 1.3 Photovoltaic Source

Photovoltaic (PV) is the technology that generates DC electrical power measured

in watts (W) or kilowatts (kW) from semiconductors when they are illuminated by

photons [12]. For the purpose of mathematically analyzing PV module and also

simulation work, a simplified model is shown in Figure 1.1.

Rs Iout
+
Ipv Id
Rp
Vout

Figure 1.1: Simplified PV Cell Circuit

Since the power of solar irradiance on the whole earth can be considered as

infinite for human beings, many scholars hold the opinion that PV is the solution to

energy crisis. Nowadays it is common to see solar panel (a set of PV cells) on residential

rooftops. They work very well on limited housing appliances, for example heating water.

Aiming to let solar energy replace more of conventional energy, utility-scale solar farm is

being studied and built increasingly. Besides PV panels, converters are used to regulate

DC voltage after the output of PV. Along with converters, Maximum Power Point

Tracking (MPPT) is one of the most important applications for PV source. Due to the

characteristic of PV cell researchers have developed various control scheme for MPPT to

provide the most power from PV cell when weather condition is given. If PV source is

6
connected with AC system, inverters are required to install after converters. Normally the

capacity of utility-scale solar farm is over 500 kW. In short duration of seven years, from

2004 to 2010, the total global grid-connected solar PV capacity has increased at an

average annual rate of 55%, to a total capacity of about 40 GW [13]. Even though this

new application has huge potential benefits, there are some technical concerns still

existing.

Previous work pertaining to solar PV applications in power systems can be

categorized into three major categories: modeling, technical impact, and financial

planning [14]. In this thesis, research focuses on the first two aspects. Various models

have been built and analyzed to study PV source’s impacts on power quality in

distribution systems [15-17]. One of the most important perspectives of power quality

study is harmonics analysis. In PV source, inverters are considered as sources of current

harmonics. Therefore, by improving control scheme of inverters, mitigation and

compensation of harmonics can be achieved [18, 19]. Also because of the intermittent

characteristic of solar irradiance, PV sources bring more uncertainty to systems. This

uncertainty includes time-varying penetration level and its impact on voltage profile [20]

and partial shading condition [21].

Large-scale PV source takes large areas. The limitation of PV farm size and

proper location for solar irradiance determine that instead of connecting to transmission

system directly, PV sources will have much more opportunities to serve microgrid. In the

fourth part of this chapter, the characteristic of microgrid is illustrated.

7
 1.4 Nonlinear Load

Loads usually encountered in power systems can be broadly categorized into

industrial, residential, municipal and commercial loads [22]. When voltage and current in

same frequency have a linear relationship, this load can be called linear load. Most of

time, this type of loads is sets of linear resistors or linear inductors. However, different

from classic load theory, in reality there is no load is pure constant power or pure

constant impedance. So practical loads should be viewed as mixtures of linear loads and

nonlinear loads. Because of massive applications of power electronics, many electrical

devices are driven by multiple controllers or rectifiers. Different combinations of diodes

and capacitors can cause significant nonlinearity. For example, there is a battery charger

which is driven by a diode bridge rectifier (DBR) and if supplied by a pure sinusoidal AC

source of 110V the current waveform is shown in Figure 1.2.

Figure 1.2: Current Waveform of DBR

Thus how to model loads becomes a significant topic for power system research

in order to simulate systems accurately. Load models, regardless of the modeling

8
techniques, are divided into static and dynamic load models [23]. Static loads describe

the relationship between the power consumption, voltage and frequency [24], while

dynamic loads provide the additional advantage of representing time-sensitive behavior

of the load [25].

For static load modeling, there are two major theories: ZIP load model and

Crossed Frequency Admittance Matrix. The constant impedance, constant current, and

constant power components of a ZIP load are represented by a second-order polynomial

in bus voltage magnitude Vi [26]:

PDi (Vi )  a1iVi 2  a2iVi  a3i (1.3)

QDi (Vi )  b1iVi 2  b2iVi  b3i (1.4)

where a1i, a2i, a3i, b1i, b2i and b3i are coefficients for power demand. The second order

term of voltage is for constant impedance (Z); the first order term of voltage stands for

constant current (I); and the zero order term of voltage is for constant power (P). This

method can give very accurate description of power consumption, voltage profile and

current flow. Nonetheless, it is not able to provide clear information of frequency

response of loads. Given the fact that harmonics study has crucial significance for power

quality, it is very necessary to consider a frequency dependent model for loads. Therefore

Crossed Frequency Admittance Matrix theory is suitable for harmonic model of loads.

This type of harmonic model takes into account the harmonic influence between

harmonic currents and harmonics voltages of different order. Equation (1.5) shows how

crossed frequency matrix represents this influence:

9
  I1   Y11 Y12 Y13 Y1m   V1 
    
 I 2   Y21 Y22 Y23 Y2 m  V2 
 I3      V3  (1.5)
    
    
 I  Y Ymm  Vm 
 m   m1

where voltages and currents are represented as complex numbers. The subscript m stands

for the harmonic order. For pure fundamental frequency voltage sources, voltage vector

elements are zeroes except V1. But currents flowing into loads still have harmonic

distortion due to the fact that Y21, Y31 and Ym1 are not all zeroes for nonlinear load. On

contrary, for linear load, the crossed frequency admittance matrix becomes a diagonal

matrix and the off diagonal elements are zeroes. The crossed frequency matrix models the

load as a harmonic currents source.

A proper load model is very important for doing case study in power system.

1.5 Microgrid Characteristics

Microgrid is a new concept which is brought up within the recent ten years. There

is no strict definition for microgrid, scholars share some same opinion about the

characteristics of microgrid.

It is running at low voltage level, mostly same voltage level as distribution

system. Microgrid concept assumes a cluster of loads and microsources as a single

controllable system that provides power to its local area [27]. Normally Microgrids are

considered as viable options for those places where main grid expansions is either

10
impossible or has no economic justification, such as the electrification in university

campuses, military installations and rural villages [1].

Conventional power systems have a configuration of one-way structure, which

means that there is only one source supplying the whole system. Therefore in

conventional power system, the power flow is only one-directional. But in microgrid,

these microsources can be installed in every possible locations, and it is also called

distributed generators (DGs). The types of DGs can be bio-mass generation, small-size

gas turbines, solar plant, wind plant and energy storage. The main purpose of microgrid is

to serve local area and finally reach to autonomous operation with less pollution. Based

on this purpose, sources of large capacity are not practical, instead small scale renewable

energy can play a key role.

Some researchers regard the microgrid as a specific distribution system embedded

with DGs, which may operate in grid-connected or islanded (stand-alone) mode [2]. In

grid connected mode, most of the system-level dynamics are dictated by the main grid

due to the relatively small size of microsources; in stand-alone mode, the system

dynamics are dictated by micro sources themselves, their power regulation control and, to

an unusual degree, by the network itself [28]. Also even when microgrid is running under

grid connected mode, if there is a fault happening in the microgrid the connection with

grid should be cut off to protect the grid. So it is not always guaranteed that energy

storage and DGs can fully meet the load demand. In this condition, some loads have to be

cut off power supply and these loads are considered as non-critical loads; while some

11
loads have to maintain power supply for 24 hours and these loads are considered as

critical loads. Figure 1.3 shows a typical microgrid.

DGs

Noncritical
Load
Substation

Grid

Critical Load
Battery
PCC
Breaker
DGs

Figure 1.3: Typical Microgrid

Due to this special structure, along with this beneficial advantage, there are

potential challenges to keep the system running safe and well. One of main concerns in

the microgrid study is the electrical power quality issue.

Researchers have been doing works on how to monitor and simulate the actual

system. The first step of studying electrical power quality issues, particularly harmonics,

should be modeling a microgrid with renewable sources both in grid-connected mode and

stand-alone mode [15, 28, 29]. To address harmonics distortion, filters are designed and

tested by scholars. Compared to active power filters, passive filters consists of only

passive components and cost less. In [30] [31], several novel passive filters were

designed and tested to show the improvement. Yet passive filters can be very limited

when harmonic distortion is very serious and less predictable, since fixed passive filters

can only help decrease harmonics on the default setting orders. Thus, active power filters

12
were studied for harmonics issues in [11] [32]. Unfortunately, most of these researches,

which studied filter designs or inverter controls, were only focused on power electronic

fields regardless of the harmonics response in the microgrid.

According to previous researches, the factors which have influence on power

quality, specifically harmonics distortion, are very comprehensive. Therefore DGs,

inverters, weather conditions, energy storage and load modeling should be included all

together in the harmonics study in order to have better simulation results. Detailed study

is presented in the following chapters.

13
 CHAPTER TWO

MODELLING OF MICROGRID

The study of microgrid can be achieved through different research methods,

including hardware test, computational programming, mathematical analysis and

software based simulation. As it is known that hardware test costs a lot and

computational programming is hard to describe the transient performance of some of

power electronic devices. In order to study the influence of harmonics in microgrid with

PV integration, software based simulation should be the best option. This thesis is based

on the results which come from computer simulations and therefore it is significant to

model each component in the microgrid in the most suitable ways. This chapter

demonstrates the details of modelling works of PV, nonlinear load, energy storage and

active power filter. Along with the process of modelling, the performances of each

components are also presented to prove the practicability.

2.1 Photovoltaic Source

2.1.1 Photovoltaic Source Modeling

In this thesis KC200GT is chosen as the model of PV cell used for integration.

The KC200GT data is determined by its own manufacture process. The data can be

obtained from the manufacturer data sheet and some parameters are shown in Table 2.1

[33].

14
 TABLE 2.1

Parameters of KC200GT Solar Array at Nominal Operating Conditions

Imp 7.61A
Vmp 26.3V
Pmax,m 200.143W
Isc,n 8.21A
Voc 32.9V
I0,n 9.825 × 10−8 A
Ipv 8.241A
a 1.3
Rp 415.405 Ω
Rs 0.221 Ω
Kv -0.1230V/K
KI 0.0032A/K

The basic mathematical equation of PV array to describe I-V is shown below [34]:

V  Rs I out
( )
I m  I pv  I 0 [e Vt a
 1] (2.1)

G
I pv  [ I pv ,n  K I (T  Tn )] (2.2)
Gn

I sc ,n  K I (T  Tn )
I0  Voc ,n  Kv (T Tn )
(2.3)
( )
e aVt
1

where Ipv is the current directly generated by one PV cell, I0 is the reverse saturation

current of the diode shown in Fig. 1.1, Im is the source current of PV cell, Iout is the

15
output current after resistance of PV cell, V is the output voltage, a is the diode ideality

constant, Vt is the thermal voltage of PV cell, Isc is the short circuit current of PV cell, T

is the temperature and G is the solar irradiance. Kv is the voltage gain of solar irradiance

and KI is the current gain of solar irradiance.

Based on the data in TABLE 2.1 and equation (2.1) to (2.3), mathematical model

was built in MATLAB/SIMULINK platform. After PV cell, a boost converter with

MPPT block was built. By comparing the output power and output voltage of PV cell on

very sample time period, PMW signal can be generated to gate of IGBT. And from there,

maximum power point can be traced. However, the output voltage of PV array itself is

not fit for the next level inverter control and a boost converter can improve the DC

voltage level. The structure is shown in Figure 2.1.

Boost Converter

PV Array V C R

MPPT

Figure 2.1 Boost Converter with MPPT

16
2.1.2 Inverter Control of Photovoltaic Source

From previous researches there are two main methods for inverter control with

PV integration. One is Active Power Control (or called PQ control), and another one is

Voltage Frequency Control (or called VF control). When considering grid-connected

mode of microgrid, PQ control is the better option for lowering the reactive power output

from PV sources; when considering stand-alone mode operating, energy storage is the

slack bus in the system instead of PV so PQ control can help provide the most active

power for critical loads. Therefore, PQ control was built to be connected in PV source

system. Figure 2.2 shows the scheme of conventional configuration of three phase active

power inverter control. To be more specific, this control scheme has another name, feed-

forward decoupling PQ control. The dq transformation block can work as switching three

parameters (ia, ib, ic) of AC to two parameters (id, iq) in DC. The active and reactive

power signals (Pref, Qref) are used to obtain the reference signal (idref, iqref) of inner current

control loop by the matrix solver and equation (2.4) is given [35]:

d  id (t )    R  L  id (t )  ud (t )  ed (t ) 

      (2.4)
dt  iq (t )    L  R   iq (t )   uq (t )   eq (t ) 
L

where L and R represent inductance and resistance of the impedance Za, Zb and Zc between

three phase inverter and voltage feedback in Figure 2.2. Since the purpose of this control

scheme is to force the reactive power output close to zero, the reference on q-axis set to

zero. The ud and uq are the output control signals of the current control block.

17
 Boost Three Phase Za

PV Converter Inverter Utility

Zb
Panel with MPPT
Zc
ia
ib
ic
ea
PWM eb
abc ec
dq

id ed
iq eq
Current
Control idref
Power
iqref Control Pref, Qref

Figure 2.2 PV System Including Inverter and Its Control Scheme

2.1.3 Simulation of Photovoltaic Source

Building PV source as a simulation file is aiming to help solve practical cases and

problems. In that way, it is necessary to prove the function and response of these

simulation blocks. The inputs of this PV cell are solar irradiance and temperature. In spite

of the fact that there are a lot of other parameters which can slightly influence the output,

for example humidity, solar irradiance and temperature are still the dominant factors over

all of others.

18
 10
S=1000 W/m2
S=800 W/m2
S=600 W/m2
8 S=400 W/m2
S=200 W/m2
Current (A)
6

0
0 5 10 15 20 25 30 35
Voltage (V)

Figure 2.3: I-V Curve of Single PV Cell

under Different Solar Irradiances (at 25°C)

10 T=50°C
T=30°C
T=20°C
8 T=10°C
T=0°C
Current (A)

0
0 5 10 15 20 25 30 35
Voltage (V)

Figure 2.4: I-V Curve of Single PV Cell

under Different Temperatures (at 1000 W/m2)

19
 300
S=1000 W/m2
S=800 W/m2
250
S=600 W/m2
S=400 W/m2
Active Power (W)

200
S=200 W/m2

150

100

50

0
0 5 10 15 20 25 30 35
Voltage (V)

Figure 2.5: P-V Curve of Single PV Cell

under Different Solar Irradiances (at 25°C)

300 T=50°C
T=30°C
T=20°C
250 T=10°C
T=0°C

200
Active Power (W)

150

100

50

0
0 5 10 15 20 25 30 35
Voltage (V)

Figure 2.6: P-V Curve of Single PV Cell

under Different Temperature (at 1000 W/m2)

20
 From Figure 2.3 to Figure 2.6, curves are generated to prove the functions of PV

cell simulation block. In Figure 2.3, the I-V curves show that when boosting the voltage

after the maximum power point output currents drop quickly and higher irradiance can

generate higher current (nominal irradiance is 1000 W/m2). In Figure 2.4, the I-V curves

demonstrate that while boosting voltage after the maximum power point currents drop

quickly and lower temperature can slightly cause higher currents.

Figure 2.3 to 2.6 are based on the single PV cell simulation. However, a PV

source system consists of dozens of PV arrays which contains hundreds of PV cells. Here

in this thesis, each single PV source has 3000 PV cells to work together. At the nominal

condition which is at 25°C and 1000 W/m2, the rated output active power is 600 kW, also

shown in Figure 2.7. Furthermore, since the well-tuned inverter control can decrease the

harmonic distortion of output current, the simulation results in Figure 2.8 and 2.9 give

clear look at the output current at nominal condition when the voltage level is 4.16 kV.

5
x 10
7

5
Active Power (W)

0
0 1 2 3 4 5 6 7 8 9 10
Time (s)

Figure 2.7: Active Power Output of PV Source System at Nominal Condition

21
 150

100

50

0
Current (A)

-50

-100

-150

-200

-250

-300
0 1 2 3 4 5 6 7 8 9 10
Time (s)

Figure 2.8: Output Current at Phase A of PV Source System at Nominal Condition

100

50
Current (A)

-50

-100

9.58 9.59 9.6 9.61 9.62 9.63 9.64 9.65 9.66 9.67
Time (s)

Figure 2.9: A Closer Look at

Output Current at Phase A of PV Source System at Nominal Condition

22
 2.2 Nonlinear Composite Load

2.2.1 Nonlinear Load Modeling Method

Many theories have been proposed for nonlinear load modeling but as it is

mentioned in Chapter II Crossed Frequency Matrix can demonstrate the harmonic

response of different types of loads. Therefore, in this study Crossed Frequency Matrix is

applied to model the load. Particularly, instead of only modeling one type of load, here a

composite load is aiming to be modeled. For the purpose of composite load simulation,

with given voltages, output currents can be monitored. When monitoring output currents,

time domain current waveform is analyzed by the Fast Fourier Transformation (FFT) so

that it can be expressed as a complex number matrix. In this complex matrix, each row

stands for a current on a distinct harmonic order. In microgrid, other than resonance

phenomenon, high frequency harmonic currents have quite small magnitude. It is not

practical to consider every order of harmonic currents that exist in the system. In order to

simplify the modeling process, here the complex matrix only includes from fundamental

current up to 13th order of harmonic current. With only fundamental voltage, 60 Hz, first

column elements in crossed frequency admittance matrix can be calculated. Then by

superimposing 3rd, 5th, 7th, 9th, 11th and 13th order harmonic voltage sources, each one at a

time, the other columns can be calculated likewise. Therefore, three 7×7 admittance

matrixes are built for the three types of nonlinear loads. Equation (2.5) demonstrates how

each harmonic order current is determined through the Crossed Frequency Matrix which

is shown in equation (1.5) and m, n both represent harmonic order number. And equation

(2.6) and (2.7) explain how elements in the matrix is calculated.

23
 𝐼1̇ = 𝑌11̇ 𝑉1̇ + 𝑌12̇ 𝑉2̇ + 𝑌13̇ 𝑉3̇ + ⋯ + 𝑌1𝑛̇ 𝑉𝑛̇

𝐼2̇ = 𝑌21̇ 𝑉1̇ + 𝑌22̇ 𝑉2̇ + 𝑌23̇ 𝑉3̇ + ⋯ + 𝑌2𝑛̇ 𝑉𝑛̇

… …

̇ = 𝑌𝑚1
𝐼𝑚 ̇ 𝑉1̇ + 𝑌𝑚2
̇ 𝑉2̇ + 𝑌𝑚3
̇ 𝑉3̇ + ⋯ + 𝑌𝑚𝑛
̇ 𝑉𝑛̇ (2.5)
𝐼𝑘̇
𝑌𝑘1̇ = (2.6)
𝑉𝑘̇
𝐼 −𝑌𝑘1 𝑉1 (2.7)
𝑌𝑘𝑗̇ = 𝑘 𝑉𝑗

2.2.2 Individual Nonlinear Load Types

• Instead of making the assumption of single type load, in order to consider the

practical situation of microgrid load demand, in this study a composite load is modeled,

including three types of nonlinear load and one linear load. The first nonlinear load type

is compact fluorescent light (CFL), which is the basic load in residential systems. Then

the second one is load with diode bridge rectifier (DBR), which is widely applied to a

large amount of home appliances including desktop computers, television sets, battery

chargers, adjustable speed drives for heating pumps and air conditioning etc. The third

type is load with phase angle AC voltage controller (PAVC), which normally used in

light dimmers, heating load, single-phase induction motors. The structure of this

composite load is shown in Figure 2.10.

24
 Figure 2.10: Composite Load Structure

The next step is to use simplified circuit to represent each type of load and

calculate each admittance matrix according to equation (2.6) and (2.7). In Figure 2.11,

each type of nonlinear load simplified circuit is given. First, to connect an AC source

with only fundamental frequency to each circuit and then from simulation results currents

can be recorded and divided into each frequency by FFT. From equation (2.6), elements

in the first column can be found. Secondly, to superimpose one j order harmonic voltage

source to the previous circuit and elements on column j can be calculated by equation

(2.7).

25
 C1

AC R
AC R

C2

(a) (b)

DBR
CFL
PAVC
Linear

AC R

(c) (d)

Figure 2.11: Simplified Circuit of Composite Load: (a) CFL; (b) DBR;

(c) PAVC; (d) Structure

Figure 2.12 shows the simulation of a 15W CFL load, which is under 120V AC

voltage level. According to the results of current FFT, the Crossed Frequency Admittance

Matrix is calculated (the matrix is provided in Appendix A). In Figure 2.13, the value of

Z axis shows the magnitudes of each elements in the admittance matrix while X and Y

stand for the harmonic orders. Obviously, this load is far away from being linear and 5th,

7th, 11th harmonics are the main factors.

26
 Figure 2.12: Simulation of CFL

Figure 2.13: Admittance Matrix Magnitude Distribution of CFL

27
 Similarly, Figure 2.14 shows the simulation of 0.67 kW DBR and Figure 2.15

gives the demonstration of the magnitudes of elements in this matrix. Compared to the

matrix of CFL, this one has different characteristic. Diagonal elements have the largest

magnitude and this can be interpreted that there is relatively small interaction between

different frequencies. Meanwhile, among diagonal elements, ones which represent for

11th and 13th order harmonic frequencies have larger magnitudes, which means that on

high frequency this load has smaller resistance.

Figure 2.14: Simulation of DBR

28
 Figure 2.15: Admittance Matrix Magnitude Distribution of DBR

A 0.9 kW PAVC type load is simulated as Figure 2.16. Figure 2.17 shows that a

load with PAVC can be close to linear load, since there is fairly small interaction between

frequencies and magnitudes of diagonal terms are almost the same.

Figure 2.16: Simulation of PAVC

29
 Figure 2.17: Admittance Matrix Magnitude Distribution of PAVC

2.2.3 Simulation Model Building

After obtaining the crossed frequency admittance matrix for each type of

nonlinear load, the next goal is to include these matrices into simulation. Assuming the

total load demand is fixed, and with the same amount of active power and power factor

the numbers of each type of loads which were described above can be estimated. In real

residential system, most of house appliances are connected in parallel so that individual

equipment can work independently. Thus, total current will simply be sum of the

individual currents. And different loads contain nonlinear loads with different ratios.

Figure 2.18 explains how this mathematical model works in the simulation of microgrid.

Firstly, voltage measurement unit is able to obtain voltage information from the feeder in

microgrid. And then through Fourier transformation, signals in time domain can be

30
transformed into frequency domain and be formed into a voltage matrix. This voltage

matrix has only one column complex numbers and each row represents one harmonic

order. According to equation (2.5), with both admittance matrix and voltage matrix, the

current matrix can be calculated. Therefore, using the complex number in current matrix,

a current signal in time domain can be generated to control the controllable current source

to draw currents from the feeder. Along with the process of nonlinear load, linear load is

also connected in parallel.

Feeder in Microgrid

Voltage
Measurement Ilinear
Linear Load

Inonlinear
Va

Magnitude
Fourier
Analysis Phase Angle

Controllable
Current Source

Admittance Matrix of Generate Voltage

Nonlinear Load Matrix

Current
Waveform
Multipli-
cation
Current
Matrix
Magnitude

Phase Angle

Figure 2.18: Composite Load Simulation Scheme Diagram

31
 2.3 Energy Storage

Energy storage is one of the most unique part of microgrid structure. Because

microgrid should be able to operate either under grid-connected mode or stand-alone

mode, energy storage is necessary for stand-alone operating. In stand-alone mode, the

most important requirement is to provide reliable power to customers. However,

renewable energy, such as PV source, has intermittent characteristic and weather

dependent restrictions. Particularly solar irradiance varies during daytime so that the

output of PV source varies greatly with irradiance. Thus, other stable types of distributed

sources are necessary in a microgrid. In most practical applications, batteries and diesel

generators, which work as the slack bus in stand-alone mode, can supply stable voltage

and frequency for the system.

Beside PQ control, v/f control and droop control are often used in inverter control

of distributed sources in microgrid. Since the goal of PQ control is to achieve the

maximum output of active power and minimum reactive power, it is used for PV control

in grid-connected mode and stand-alone mode. However, in stand-alone mode, energy

storage bus should be the slack bus. Slack bus works to provide stable voltage and

frequency for the system, therefore v/f control and droop control are often used for

inverter control of battery. Compared to v/f control, droop control does not maintain the

same stable voltage and frequency but varies in a small range with the output. Higher

active power output can causes small drop on frequency and higher reactive power output

can cause small drop on voltage.

32
 f  f ref  k x ( Pref  P) (2.8)

V  Vref  k y (Qref  Q) (2.9)

However, because of this characteristic, multiple batteries and other type of

microsources can work together to communicate. So in this study, droop control is

appliced to the inverter control of the energy storage, which is a DC battery. The control

scheme and structure is shown in Figure 2.19. Furthermore, the performance of this

simulation model is going to be shown and explained in the next chapter.

Pref Qref fref Vref

Output
3-phase Output Active
Current Signal Power
Minus kx Sum

3-phase Power Output Reactive

Power
Output Calculation
abc/dq
& Minus ky Minus
Voltage Signal
Transformation
Current &
Voltage in dq
axises

PWM signal
PID Controller
to inverter PWM & dq/abc
Vref in
Generator Transformatio
n
dq
axises

Figure 2.19: Droop Control Structure of Storage Battery

33
 2.4 Active Power Filter

The design of active power filter is not the focus point of this research. Yet, when

discussing harmonics distortion problem, the solution is always about filter design and

installation. In a multi-battery microgrid, harmonics can be much higher in stand-alone

mode than in grid-connected mode. Sometimes, too much harmonics can have quite

negative influence on power devices and customers. Thus, a proper designed filter should

be considered as an important part of microgrid system.

2.4.1 Structure of Active Power Filter

Harmonics can cause damages on electrical devices, such as, overheating

transformers. Therefore, it is very important to monitor power quality in the grid,

especially harmonic distortion levels. If harmonic distortion is beyond the regulation, it is

necessary to install filters to mitigate harmonics. There are two types of filters in general:

passive filters and active power filters.

Renewable energy inverters are the main source of harmonics, the magnitudes of

current on each harmonic are changing all the time along with weather conditions, such as

solar irradiance. However, passive filters are set to decrease the current magnitude in some

specific harmonic order. If these magnitudes are changing, it is hard to choose which

frequency to be tuned to improve the power quality significantly because the major

harmonics distortion might not be the one from the initial design. Thus, an active power

filter can help improve power quality by having an active power source with controllers,

shown in Figure 2.20 [9].

34
 Feeder Nonlinear
Grid Load
Zl
Zl
Zl

Active Power
Filter Three-Phase
Inverter

Figure 2.20: Active Power Filter Structure

2.4.2 Simulation of Active Power Filter

In order to show the performance of the active power filter mentioned above, a simple

simulation study was done. The presented parallel three phase inverter can be controlled

by different types of control schemes and the advantages and disadvantages are discussed

in [36]. The proposed model is using a current compensation method, called Instantaneous

Reactive-Power algorithm [37]. When the active power filter is connected to the grid, the

power delivered on different frequencies can be calculated. By a basic 5th order

Butterworth low pass filter, power that is delivered on high frequency can be detected and

be considered for compensation control scheme. Therefore, the inverter can control the

filter to compensate current harmonics through PWM signals, not only in one specific

order but more generally. To test this active power filter simulation model, it is connected

with a nonlinear load on 4.16 kV (phase to phase) voltage source. From current

35
waveforms shown in Figure 2.21, it can be observed that the harmonics distortion is

improved significantly. By FFT, the current magnitude spectrum on frequencies is

presented as follows. Figure 2.22 gives the FFT spectrum of source current without active

power filter while Figure 2.23 shows the spectrum after the filter is connected. From the

comparison between Figure 2.22 and Figure 2.23, Total Harmonic Distortion (THD) drops

from 19.70% to 3.42% and harmonics on 5th order drops from 15% to 0.35%.

Current (A)
100

-100

100
Current (A)

-100

100

-100
0.19 0.2 0.21 0.22 0.23 0.24
Time (s)

Figure 2.21: Current Waveform (at 0.2s the active power filter is connected)

Fundamental (60Hz) = 89.17 , THD= 19.70%

15
Mag (% of Fundamental)

10

0
0 2 4 6 8 10 12 14 16
Harmonic order

Figure 2.22: FFT Spectrum of Source Current without Active Power Filter

36
 Fundamental (60Hz) = 89.22 , THD= 3.42%
0.4

Mag (% of Fundamental) 0.3

0.2

0.1

0
0 2 4 6 8 10 12 14 16
Harmonic order

Figure 2.23: FFT Spectrum of Source Current with Active Power Filter

2.5 Summary

This chapter includes simulation modelling works for each component in this

microgrid study. Specifically the structure of microgrid consists of renewable source,

nonlinear load, energy storage and potential power filter. Traditional simplified model or

equations which describe the relation of input and output cannot meet the requirement of

harmonics study since it mainly supports steady state study. Steady-state study only gives

the values and information of steady state output or parameters in the power system

without considering short-time transient conditions. Nevertheless, the origins of

harmonics are from inverter performance in transient status. For a long time this area

studies have been regarded as the focus area for scholars who major in Power

Electronics. Consequently researches in Power System area have ignored the detailed

model of various types of inverters while dealing with microgrid. Therefore what makes

this chapter important is while studying the harmonics influence of microgrid, the

37
detailed models are not ingored so that the simulation can give more practical

demonstration. Especially loads and sources, which are major origins of current

harmonics in microgrid, are well modelled from the view of harmonics. With the detailed

models built in this chapter, more simulations results can be persuasive and solid.

38
 CHAPTER THREE

HARMONICS IMPACTS ON MICROGRID

The research method of this thesis study is to use computer software to simulate

the influence brought by PV integration in microgrid. Based on the modelling work

mentioned in Chapter 2, multiple simulations can be conducted to analyze harmonic

distortion influence. Specifically, both grid-connected and stand-alone mode are studied.

In order to evaluate harmonics with quantified indices, several power quality indices are

described in the following part.

3.1 Power Quality Indices

Power quality indices are parameters that are able to show some part of electric

characteristic according to current or voltage measurements. In other words, power

quality indices are defined to quantify the distortion for current or voltage. Each index

has different function to describe the distortion. Therefore, multiple indices should help

researchers to have a clear look at the harmonics distortion from various perspectives.

THD is the measurement of the harmonic distortion at each node on each bus in

the system. In order to look at the harmonic distortion in the whole system, whole system

harmonics distortion level (WSHDL) is defined as:


n
THDi
WSHDL  i 1 (3.1)
n

where n is the number of nodes; h represents the harmonic order; m is the considered

39
 maximum harmonic order. I1 is the absolute value of fundamental (60Hz) current. The

meaning of this definition is to have a great picture for the harmonics distortion trend.

Also in low voltage level system, such as microgrid or distribution systems, instead of all

three-phase line structure, there are many two-phase feeders or single-phase feeders.

Therefore, it is important to look into each node on each bus.

In addition to the WSHDL, there are three types of PQ indices are introduced and

discussed along with simulation results. Distortion power DP index is defined in (3.2)

[38]. The total apparent power, fundamental active power and fundamental reactive

power are defined in (3.3) to (3.5).

DP  S 2  P12  Q12  V1  I1  THDV 2  THDI 2  (THDV  THDI )2

(3.2)

m 1 m 1
where the total apparent power S  V ( n) 2 
n 0
 I ( n)
n 0
2
, (3.3)

fundamental active power P1  V1  I1  cos( ), (3.4)

fundamental reactive power Q1  V1  I1  sin( ) (3.5)

By normalizing DP to unity, normalized DPnorm index can be obtained. DP index

has the ability to show contributions of distortion power from individual customers to

PCC. From (3.2), it is easy to estimate the power delivered not on 60 Hz simply by taking

measurements of THD into the equation. This index is crucial for utility to monitor how

much power is lost. Waveform distortion WD index is defined as [39]:

40
  M N 
WD   I m1  I1   I int  
2
 / I1
2 2
I int (3.6)
 eg  h ,i er  h , j
 i 2 j 1 

where I1 is the rated current magnitude and Im1 is the measured fundamental current

magnitude. Iinteg-h,i is the ith integer harmonic component and Iinter-h,j is the jth inter-

harmonic component. WD index expresses how much a component, AC current, is

distorted or deviated from ideal sinusoidal waveform. WD index includes inter-harmonic

components, which can take a large part of harmonics when different types of inverters

involved. Also inter-harmonics cannot be presented by THD which only include integer

order of harmonics. Instead of an average value, WD index gives an instantaneous

distortion ratio and it can be depicted along with time axis to be monitored. Symmetrical

components deviation SCD index is defined as [38]:

 2 2 
SCD   I mp  I1  I mn
2
 I mz  / I1 (3.7)
 

where Imp, Imn, Imz are the measured currents at positive, negative and zero sequences.

SCD index has a significant meaning for microgrid because it can give the level of

unbalance on currents. SCD index can help staff who work in substations to recognize the

unbalance in each node and its impact on the whole system.

3.2 Test System Information

Study system shown in Figure 3.1 is a 14-bus 4.16 kV microgrid (located in GA,

USA). The detailed feeder configuration and its impedance matrices are presented in

41
Appendix B. Bus 10 is the slack bus and the balance point in grid-connected mode. The

slack bus voltage is set to 1.05 pu to keep all node voltages at least 0.95 pu. In Figure 3.1,

three lines represent three phase feeder. Two lines represent two phase feeder and single

line means one phase feeder. Load information is given in Table 3.1. The PV sources as

the only type of renewable energy are integrated to the system as shown in Figure 3.1.

The inverter control inputs ea, eb, ec, ia, ib, ic are obtained from monitoring the integration

point in microgrid. If these values are away from ideal balanced condition and changing

along with time, then PV sources will provide more harmonics than running under ideal

balanced condition because of the performance of inverters. Figure 3.2 shows the solar

irradiance and temperature during 24 hours (average value of typical day of July within

10 years). Since in the simulation model, PV’s output is only influenced by temperature

and solar irradiance, only these two factors are considered in the typical day of July. In

this work, the simulation only runs during the sunniest period (7am to 6pm).

PV
Grid

PCC
Bus 40
Bus 10

Bus 41

Bus 50
Bus 20

Bus 30
Bus 51
Bus 60

Bus 31 Bus 62
Bus 34

Bus 32 Bus 33 Bus 61

Figure 3.1 Test System Structure

42
 Table 3.1 Load Data of the Study System

Load Data for the System

Bus Reactive
Active
Note Power
Power (kW)
(kVar)
10 — — —
20 Three phase 630 212
30 Three phase 412 112
31 — — —
32 One phase 37 12
33 One phase 24 8
34 Three phase 343 122
40 Three phase 175 100
41 Three phase 133 51
50 Three phase 298 151
51 Two phase 68 12
60 Three phase 200 70
61 Three phase 74 28
62 One Phase 32 11

A Typical Day of July

800
solar irradiance (w/m2 )

700
600
500
400
300
200
100
0
0 5:00 10:00 15:00 20:00
Time (hour)

35
33
temperature (°C)

31
29
27
25
23
21
0:00 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 24:00
Time (hour)

Figure 3.2 Meteorological Data on Typical July Day

43
 3.3 Harmonics Impacts on Grid Connected Mode

The test system is simulated in this section in MATLAB/Simulink and analyzed

considering the PV model and the composite load model explained in 2 to study the

harmonic distortion and PQ indices described in section 3.1. The PV data are set as in

Table 3.1. After running the power flow in the basic microgrid (without PV integration

and without nonlinear loads), voltages on each node are found in the acceptable

fluctuation range (0.95, 1.05 pu).

3.3.1 Considering PV with Linear Load

Two cases are studied based on PV locations. In case I, PV locations shown in

Figure 3.1 are considered. Applying equation (3.1) gives the harmonics distortion data for

each bus from 7:00 am to 6:00 pm. WSHDL values demonstrates the harmonics

distortion in the microgrid system as shown in Figure 3.3. In Figure 3.3, the harmonic

distortion is significant at noon time. In other words, as the percentage of power supplied

by PV is going higher, the WSHDL is going higher. This phenomenon can be explained

by the fact that inverter produces harmonics and more power coming from PV means

more distorted power is generated. THD provides analysis in each bus as shown in

Figure 3.4. According to the standards in [40], when voltage level is under 69 kV and

Isc/IL is smaller than 20, THD of current is regulated under 5% by standards. So the THD

values above 5% represent nodes that largely influenced by PV integration and some

potential damage might be caused by harmonics to shorten the lifetime of electrical

components.

44
 Whole System Harmonics Distortion Level
2.4
Injection at Bus 50 60 & 61
2.2

2.0

1.8

1.6
percent %

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00
time (hour)

Figure 3.3 WSHDL from 7:00AM-6:00PM (Case I)

THD Plots of Case I

15
20A
30A
34A
Harmonics Distortion (percentage%)

40A
41A
10 50A
5% harmonics is the
regulated limit 60A
61A
32
33
51A
5
51B
62

0
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00
Time (hours)

Figure 3.4 THD from 7:00AM-6:00PM (Case I)

45
 In case II, the PV sources are integrated at buses 20, 40 and 60. Figure 3.5 shows

the WSHDL of case II. In Figure 3.6, the red line (phase A of bus 50) is the most

affected. However, THD value on all these nodes are under 5%, which is acceptable for

microgrid system. Compared to case I, the integration locations affect the THD. In case I

WSHDL is large when the penetration level is high; however in case II WSHDL is not

affected much by the penetration level change. This results give a clear demonstration

about how the integrations of PV affect the microgrid distortion. However, when THD is

over 5% the power quality is required to be improved.

Whole System Harmonics Distortion Level

0.35

Injection at Bus 20 40 & 60

0.30

0.25

0.20
percent %

0.15

0.10

0.05

0
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00
time (hour)

Figure 3.5 WSHDL from 7:00AM-6:00PM (Case II)

46
 THD Plots of Case II
4.0
20A
30A
3.5
34A
Harmonics Distortion (percentage%)

40A
3.0 41A
50A
2.5 60A
61A
32
2.0
33
51A
1.5 51B
62
1.0

0.5

0
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00
Time (hours)

Figure 3.6 THD from 7:00AM-6:00PM (Case II)

Compared to the case I, the integration locations in case II are more dispersed on

this radial system. According to the theory of PQ control inverter which is introduced in

Chapter 2, since each PV source are separated and slack bus can provide nearly ideal

balanced power from the grid side in grid-connected mode, the inverter PQ control

should be improved by less distortion reference currents and voltages. Figure 3.3 and

Figure 3.5 give a clear presentation of the difference between two location choices for

three PV sources.

47
3.3.2 Considering PV with Nonlinear Composite Load

This part investigates the effect of nonlinear loads. According to experiences from

Grainger Industrial Supply [41], a normal business building can withstand up to 15% of

nonlinear load without apparent negative influences. But when it comes larger than 15%,

necessary devices should be installed to improve power quality. Nonlinear loads are

added at different percentage w.r.to the linear loads. Blue solid line in Figure 3.7 shows

the current waveform with 10% of nonlinear load at Bus 61, phase A. At this time, the

THD is 4.0% as shown in Figure 3.8. Red solid line in Figure 3.7 represents the 15% of

nonlinear load at the same node and the corresponding THD is 5.6%. Figure 3.8 shows

how the current harmonic distortion increase with increasing the percentage of nonlinear

load. The nonlinear load is the same as the one mentioned in Chapter 2 and each type of

the admittance matrix is presented in Appendix A.

Comparison

with 10% nonlinear

15 with 15% nonlinear

10

5
Current (A)

-5

-10

-15

0.26 0.27 0.28 0.29 0.3 0.31

Time (s)

Figure 3.7 Load Current of Bus 61A with Nonlinear Load

48
 8%

7%
Regulated Limit
6%

5%
THD

4%

3%

2%

1%

0
0 5% 10% 15% 20%
Nonlinear Proportion

Figure 3.8 THD Comparison with Different Loads with PV Injection

Substation

AC
DC
PV PCC

Bus 51 Bus 60 Bus 61 Bus 62

Small Factory Business Business Residential
Building Building Building
Two Phase Single
3 Phase 3 Phase Phase

Figure 3.9 Simplified Study System

49
 For the purpose of PQ indices investigation, the microgrid system in Figure 3.1 is

simplified as shown in Figure 3.9. An equivalent PV source is considered at the PCC

point. The simplified system consists of four main loads. The composite load information

is shown in Table 3.2. Bus 51 (considered as a small scale factory and because of large

amount of motors and heating pumps) is a PAVC type of nonlinear loads. Buses 60 and

61 (two business buildings which contain many devices like computers) are classified as

DBR. Load of bus 62 is considered residential. PQ indices in section 3.1 are applied on

the simplified system. Harmonics distortion ranking (HDR) can be found through

descending sorting of DP and WD indices. The simulation results are shown in Tables 3.3

and 3.4. All these results are based on currents measurement because voltage distortion is

not dominant compared with currents. Table 3.3 lists the DP index which implies that the

large amount of load contributing more to DP and the ranking gives the same

information. However in Table 3.4, the WD index reflects that without considering power

but the waveform of currents. Clearly, due to large use of motors, heating pumps and

other heavy duty nonlinear loads, even load on Bus 51 does not consume the most power

but it still has the worst impact on current waveform. In Table 3.4, the HDR is different

with that in Table 3.3; this can give substation engineers multiple views of power quality.

The SCD index can indicate how much unbalance the system have from data obtained at

PCC. But before applying the SCD equation, one assumption has to be made that in this

case study mutual coupling between lines can be neglected due to the short length of each

feeder. According to simulation results, the magnitude of current sequences running

through PCC is [25.87 12.64 10.23] Amps ([positive negative zero]). By applying

50
equation (3.7), the SCD index is 0.54. SCD index could weight from 0 to 1 and larger

number shows more unbalanced. The contribution of harmonics from PV inverter,

different types of nonlinear loads on each phase and system configuration have

determined that this microgrid is very unbalanced. Engineers who work in substations

can also use SCD as power quality index for regulating the system.

Table 3.2 Composite Load Information

Bus Number Phase DBR CFL PAVC Linear Load

51 A&B 15% 15% 30% 40%
60 ABC 20% 20% 10% 50%
61 ABC 25% 25% 10% 40%
62 C 15% 10% 15% 60%

Table 3.3 Distortion Power Index

Bus
Phase DPnorm HDR
Number
51 A 0.0791 7
51 B 0.0787 8
60 A 0.1417 3
60 B 0.1798 1
60 C 0.1645 2
61 A 0.0793 6
61 B 0.1072 4
61 C 0.0977 5
62 C 0.0720 9

51
 Table 3.4 Waveform Distortion Index

Bus
Phase WD HDR
Number
51 A 0.482 2
51 B 0.511 1
60 A 0.301 9
60 B 0.336 7
60 C 0.324 8
61 A 0.382 6
61 B 0.453 3
61 C 0.419 5
62 C 0.437 4

3.4 Harmonics Impacts on Stand-Alone Mode

In stand-alone mode, load demands are supported by distributed sources which are

in different locations of the system. Instead of connecting Bus 10 to grid, here in stand-

alone mode Bus10 is connected to energy storage as slack bus. Microsources include, in

this case study, three PV sources, two batteries and one diesel generator.

For the models proposed in Chapter 2, the rate output power of battery on 4.16 kV

voltage level is presented in Figure 3.10. It is proven that the simulation models are well

tuned to be able to give steady output. Batteries are charged by PV sources when the

microgrid is running under grid-connected mode, while PV’s output varies along with

time, batteries give maximum output and the diesel generator covers the rest of the load

demand. In this case, total load demand is 2,426 kW and it is also assumed that only 80%

of loads is critical load in stand-alone mode. The power curve of this system is given in

Figure 3.11.

52
 5
x 10
3

2.5
Active Power (W)

1.5

0.5

0
0 0.4 0.8
Time (s)

Figure 3.10 Battery Rated Output Active Power

1400

Battery
1200 PV Source
Diesel Generator

1000
Active Power (kW)

800

600

400

200

0
7 8 9 10 11 12 13 14 15 16 17 18
Time (hour)

Figure 3.11 Active Power Curve of Microgrid in Stand-alone Mode

53
 Both nonlinear loads and inverters generate harmonics and the harmonic

distortion is able to influence the output current of energy storage. Figure 3.12

demonstrates how THD of battery output current changes during the day from 7:00 AM

to 6:00 PM. In Figure 3.12, the red line stands for THD values without active power filter

while the green line stands for THD values with active power filter. Without filter,

harmonics value can be higher than 5%, which can potentially harm the transformer

located between the battery and the microgrid. Even when the nonlinearity ratio of load

increases, this THD value could be much higher. After being connected with the active

power filter, the THD values are largely decreased. In this case, the highest value of THD

is not over 1%. Though active power filter can work effectively, it is not practical to

install many active power filters in one microgrid considering the cost. Installing one

active filter on the slack bus to protect energy storage is practical since the initial cost of

battery and larger size transformer are more expensive.

6 1
wtih filter
5.5 without filter

5 0.8

4.5
THD in Percentage

THD in Percentage

4 0.6

3.5

3 0.4

2.5

2 0.2

1.5

1 0
7 8 9 10 11 12 13 14 15 16 17 18
Time (h)

Figure 3.12 THD of Battery Current Output from 7:00AM to 6:00PM

54
 In Figure 3.13, WSHDL is plotted from 7:00 AM to 6:00 PM. While the solar

irradiance is increasing along with time, the increase of WSHDL means that the harmonic

distortion gets worse. For a grid-connected system, if the value of WSHDL is above 1%,

that means some of nodes are suffering from significant harmonic distortion. Because

when system is running under grid-connected mode THD near PCC is very small if

assuming the grid does not deliver much harmonics, there must be some buses which

suffer from high THD. Nevertheless, in Figure 3.13, it is easy to notice that WSHDL is

keeping above 2%. This can be viewed as one of the major difference between grid-

connected mode and stand-alone mode. Since the slack bus in stand-alone mode is the

one which is connected with the energy storage and generators, there are always a certain

amount of high frequency currents from the slack bus. If the nonlinearity of load becomes

more significant, it is even harder to keep THD lower than 5% on each node. When the

average THD of each node is 5%, WSHDL is 2.236%. Figure 3.14 presents THD on each

node along the same time period.

According to the comparison between Figure 3.13 and Figure 3.14, WSHDL has

the advantage of showing the trend of the whole system with one simple parameter based

on THD values. Without looking into every line in Figure 3.14, how the harmonics

distortion varies in the whole microgrid can be found in Figure 3.13. However, THD

plots provides a more accurate close look at each current, which can be analyzed

individually.

55
 3.5

2.5
Percent %

1.5

0.5

0
7 8 9 10 11 12 13 14 15 16 17 18
Time (hour)

Figure 3.13 WSHDL of Study System from 7:00AM to 6:00PM in Stand-alone Mode

12
Bus 20
Bus 30
Harmonics Distortion (percentage%)

Bus 34
10
Bus 40
Bus 41
Bus 50
8
Bus 60
Bus 61
Bus 32
6 Bus 33
Bus 51a
Bus 51b
4 Bus 62

0
7 8 9 10 11 12 13 14 15 16 17 18
Time (hours)

Figure 3.14 THD of Study System from 7:00AM to 6:00PM in Stand-alone Mode

56
 3.5 Summary

This chapter discussed the impact of two harmonic sources: PV sources and

nonlinear load, in a microgrid in both grid-connected mode and stand-alone mode. Firstly

the PV source integration is studied using linear loads; then PV sources are combined

with nonlinear loads. While considering only linear loads, this study provides a

comparison between two types of PV integrating locations. Then the whole system

harmonics distortion level as well as the THD are considered as the main measures to

evaluate the power quality and impact of PV integrations. Instead of single PV source,

this study investigates mutual influences with multiple PV sources. An actual microgrid

with historical-based weather conditions is used as a study system. Simulation results can

indicate a discernible look at these negative impacts on the microgrid power quality. In

order to make numerical comparison and ranking power quality indices are used to

estimate the current distortion from several different perspectives. Three PQ indices are

applied and compared to investigate the power distortion, waveform distortion and

system unbalance. And using the same models that were applied in grid-connected mode

study, stand-alone mode was investigated mainly by THD and WSHDL. Because the

slack bus in stand-alone mode is the energy storage battery and diesel generators, it

cannot provide power of the same quality as from the grid side. From multiple figures,

the microgrid suffers much more from harmonics distortion in stand-alone mode.

Simulation models and used matlab codes are shown in Appendix C.

57
 CHAPTER FOUR

CONCLUSION AND FUTURE WORK

The initial motivation of this thesis study is from the special structure of

microgrid which might be the future solution of power delivery. As it has been presented

above, microgrid is a low-voltage power system with small size capacity and complicated

feeder configuration. While analyzing the whole microgrid system, different with

simplified mathematical model, detailed models are able to present better view on

monitoring and assessment of harmonics. An accurate harmonics study can help reduce

power loss, financial loss and operating stability of microgrid.

Given the fact that microgrid is aiming to serve local area, it becomes one of the

most promising application platform of renewable energy. Specifically in this thesis, PV

integration is analyzed. The major issue of the PV integration is harmonics distortion

which is the focus point in this research.

In Chapter One, some related basic background knowledge were presented as the

foundation of the research. In addition, previous works by other scholars were mentioned

as literature review to give a clear picture of what has been accomplished and what need

to be improved. These foundation explains the significance of electrical power system

harmonic study, what is microgrid and why it is different from conventional distribution

system. In Chapter Two, the main content is that how the components in microgrid are

modelled for simulation. PV cell and nonlinear loads were modelled from mathematical

58
equations while inverter control schemes in both PV system and energy storage were

achieved through control blocks in Matlab/Simulink platform. Mathematical equations

express the relationship between input and output, mostly in steady state condition.

However, the Crossed Frequency Admittance Matrix is able to describe the load response

in different harmonic orders. At the same time, control blocks in Simulink can simulate

the transient response of inverters. In Chapter Four, the research demonstrated the

harmonic distortion monitoring and assessment in an actual microgrid. In order to get

valuable data for practical uses, the study system was modeled based on actual load

demand, line configuration and meteorological information. An index named as WSHDL

is proposed to evaluate the impact of harmonics distortion in both grid-connected and

stand-alone mode. Besides WSHDL, there are other three power quality indices, which

are Distortion Power Index, Waveform Distortion Index and Symmetrical Components

Deviation Index. Each of them can present how electrical power is distorted by

harmonics from three different perspectives.

There are three directions for future study: 1) optimization of PV locations in

unbalanced microgrid; 2) improvement on renewable energy focused active power filter;

3) communication and control scheme between multiple distributed sources. Given the

fact that PV sources can be installed for individual customers or for different locations in

microgrid level, then the choices of locations becomes important. As it is studied in this

thesis, the mutual influence between two solar inverters cannot be ignored when the

capacity is large enough. Therefore the optimization might be able to help improve in

both power quality and load sharing aspects. Another direction is to do research in

59
developing novel active power filter specifically for microgrid. As it was shown in this

thesis, harmonics of PV sources can vary a lot and there is no fixed dominant harmonic

order because of various inverters’ performance. Future work can be extended on

adaptive control scheme of filters. The third direction is also focused on the special

structure of microgrid. The biggest difference between microgrid and distribution system

is the existence of multiple distributed sources. Because each of the distributed sources

has different capacity and weather-dependent characteristic, it is not practical to assume

that they will cover the load with stable output in the most optimal way. For renewable

energies, for example PV and wind turbines, the output can drop down a large amount

within a short time period and meanwhile some other sources, for example batteries or

diesel generators, should be able to cover the drop to maintain the power delivery. For

this reason, a flexible and optimal communication scheme between distributed sources

might show the benefits in the future as well.

60
 REFERENCE

[1] G. Venkataramanan and C. Marnay, "A larger role for microgrids," Power and
Energy Magazine, IEEE, vol. 6, pp. 78-82, 2008.
[2] X. Xufeng, J. Mitra, W. Tingting, and M. Longhua, "Evaluation of Operational
Reliability of a Microgrid Using a Short-Term Outage Model," Power Systems,
IEEE Transactions on, vol. 29, pp. 2238-2247, 2014.
[3] S. Chattopadhyay, M. Mitra, and S. Sengupta, Electric power quality. West
Bengal, India: Springer, 2011.
[4] J. C. Whitaker, AC power systems handbook, 3rd ed. Boca Ranton, FL: CRC
Press, 2006.
[5] The Power of Renewables: Opportunities and Challenges for China and the
United States. Washington, DC: The National Academies Press, 2010.
[6] F. De la Rosa, Harmonics and Power Systems. Boca Ranton, FL: CRC Press,
2006.
[7] H. Gerber, "Nonlinear transmission line analysis for determining harmonics,"
Energy Conversion, IEEE Transactions on, vol. 14, pp. 1402-1406, 1999.
[8] J. Harries and J. Randall, "Harmonic resonance on parallel high voltage
transmission lines," Power Delivery, IEEE Transactions on, vol. 12, 1997.
[9] H.-L. Jou, J.-C. Wu, Y.-J. Chang, and Y.-T. Feng, "A novel active power filter for
harmonic suppression," Power Delivery, IEEE Transactions on, vol. 20, pp.
1507-1513, 2005.
[10] G. Casaravilla, A. Salvia, C. Briozzo, and E. H. Watanabe, "Selective active filter
with optimum remote harmonic distortion control," Power Delivery, IEEE
Transactions on, vol. 19, pp. 1990-1997, 2004.
[11] J. Allmeling, "A control structure for fast harmonics compensation in active
filters," Power Electronics, IEEE Transactions on, vol. 19, pp. 508-514, 2004.
[12] A. Luque and S. Hegedus, Handbook of photovoltaic science and engineering:
John Wiley & Sons, 2011.
[13] Renewable Energy Policy Network for the 21st Century, Renewables 2011:
Global status report Sep. 2011 [Online]. Available: http://www.ren21.net
[14] B. Tamimi, C. Canizares, and K. Bhattacharya, "System stability impact of large-
scale and distributed solar photovoltaic generation: The case of Ontario, Canada,"
Sustainable Energy, IEEE Transactions on, vol. 4, pp. 680-688, 2013.
[15] P. P. Dash and M. Kazerani, "Dynamic modeling and performance analysis of a
grid-connected current-source inverter-based photovoltaic system," Sustainable
Energy, IEEE Transactions on, vol. 2, pp. 443-450, 2011.
[16] I. T. Papaioannou, M. C. Alexiadis, C. S. Demoulias, D. P. Labridis, and P. S.
Dokopoulos, "Modeling and field measurements of photovoltaic units connected
to LV grid. Study of penetration scenarios," Power Delivery, IEEE Transactions
on, vol. 26, pp. 979-987, 2011.
[17] A. Yazdani, A. R. Di Fazio, H. Ghoddami, M. Russo, M. Kazerani, J. Jatskevich,
et al., "Modeling guidelines and a benchmark for power system simulation studies

61
 of three-phase single-stage photovoltaic systems," Power Delivery, IEEE
Transactions on, vol. 26, pp. 1247-1264, 2011.
[18] A. Kulkarni and V. John, "Mitigation of lower order harmonics in a grid-
connected single-phase PV inverter," Power Electronics, IEEE Transactions on,
pp. 5024-5037, 2013.
[19] S. Munir and Y. W. Li, "Residential distribution system harmonic compensation
using PV interfacing inverter," Smart Grid, IEEE Transactions on, vol. 4, pp.
816-827, 2013.
[20] A. Canova, L. Giaccone, F. Spertino, and M. Tartaglia, "Electrical Impact of
Photovoltaic Plant in Distributed Network," Industry Applications, IEEE
Transactions on, vol. 45, pp. 341-347, 2009.
[21] K. Sundareswaran, P. Sankar, P. S. R. Nayak, S. P. Simon, and S. Palani,
"Enhanced Energy Output From a PV System Under Partial Shaded Conditions
Through Artificial Bee Colony," Sustainable Energy, IEEE Transactions on, vol.
6, pp. 198-209, 2015.
[22] L. G. Dias and M. E. El-Hawary, "Nonlinear parameter estimation experiments
for static load modelling in electric power systems," Generation, Transmission
and Distribution, IEE Proceedings C, vol. 136, pp. 68-77, 1989.
[23] P. Jazayeri, W. Rosehart, and D. T. Westwick, "A Multistage Algorithm for
Identification of Nonlinear Aggregate Power System Loads," Power Systems,
IEEE Transactions on, vol. 22, pp. 1072-1079, 2007.
[24] A. Miranian and K. Rouzbehi, "Nonlinear Power System Load Identification
Using Local Model Networks," Power Systems, IEEE Transactions on, vol. 28,
pp. 2872-2881, 2013.
[25] "Bibliography on load models for power flow and dynamic performance
simulation," Power Systems, IEEE Transactions on, vol. 10, pp. 523-538, 1995.
[26] C. W. Taylor, Power system voltage stability: McGraw-Hill, 1994.
[27] R. H. Lasseter, "MicroGrids," in Power Engineering Society Winter Meeting,
2002. IEEE, 2002, pp. 305-308 vol.1.
[28] N. Pogaku, M. Prodanovic, and T. C. Green, "Modeling, analysis and testing of
autonomous operation of an inverter-based microgrid," Power Electronics, IEEE
Transactions on, vol. 22, pp. 613-625, 2007.
[29] C. Yea-Kuang and G. Jyh-Cherng, "Modeling and control of stand-alone
photovoltaic generation system," in Power System Technology (POWERCON),
2010 International Conference on, 2010, pp. 1-7.
[30] K. H. Ahmed, S. J. Finney, and B. W. Williams, "Passive Filter Design for Three-
Phase Inverter Interfacing in Distributed Generation," in Compatibility in Power
Electronics, 2007. CPE '07, 2007, pp. 1-9.
[31] W. Weimin, H. Yuanbin, T. Tianhao, and F. Blaabjerg, "A New Design Method
for the Passive Damped LCL and LLCL Filter-Based Single-Phase Grid-Tied
Inverter," Industrial Electronics, IEEE Transactions on, vol. 60, pp. 4339-4350,
2013.

62
[32] P. Acuna, L. Moran, M. Rivera, J. Dixon, and J. Rodriguez, "Improved Active
Power Filter Performance for Renewable Power Generation Systems," Power
Electronics, IEEE Transactions on, vol. 29, pp. 687-694, 2014.
[33] M. G. Villalva, J. R. Gazoli, and E. R. Filho, "Modeling and circuit-based
simulation of photovoltaic arrays," in Power Electronics Conference, 2009.
COBEP '09. Brazilian, 2009, pp. 1244-1254.
[34] H. S. Rauschenbach, "Solar cell array design handbook-The principles and
technology of photovoltaic energy conversion," 1980.
[35] C. Hui, Z. Hui, J. Wentao, and W. Santong, "Research on PQ Control Strategy for
PV Inverter in the Unbalanced Grid," in Power and Energy Engineering
Conference (APPEEC), 2012 Asia-Pacific, 2012, pp. 1-3.
[36] P. Fang Zheng, "Application issues of active power filters," Industry Applications
Magazine, IEEE, vol. 4, pp. 21-30, 1998.
[37] H. L. Jou, "Performance comparison of the three-phase active-power-filter
algorithms," Generation, Transmission and Distribution, IEE Proceedings-, vol.
142, pp. 646-652, 1995.
[38] J. Si-Hun, S. SeoEun, and P. Jung-Wook, "On Improving Distortion Power
Quality Index in Distributed Power Grids," Smart Grid, IEEE Transactions on,
vol. 4, pp. 586-595, 2013.
[39] L. Tao and A. Domijan, "On power quality indices and real time measurement,"
Power Delivery, IEEE Transactions on, vol. 20, pp. 2552-2562, 2005.
[40] S. M. Halpin, "Comparison of IEEE and IEC harmonic standards," in Power
Engineering Society General Meeting, 2005. IEEE, 2005, pp. 2214-2216 Vol. 3.
[41] Available: http://www.grainger.com/content/safety-electrical-power-system-
harmonics

63
APPENDICES

64
 Appendix A

Crossed Frequency Admittance Matrix of Composite Load

Table A-1 DBR Matrix

(DBR: 0.67kW with 120V AC power)

0.08∠22.00° 0.15∠-145.39° 0.12∠4.36° 0.04∠-128.50° 0.07∠-96.07° 0.04∠-97.07° 0

0.06∠-137.00° 0.38∠49.34° 0.29∠-163.58° 0.17∠13.94° 0.07∠-63.94° 0.10∠99.93° 0.18∠65.00°

0.03∠76.00° 0.27∠-125.28° 0.38∠40.04° 0.32∠-168.50° 0.06∠-23.73° 0.03∠-45.48° 0.05∠145.87°

0.01∠-25.00° 0.06∠59.86° 0.24∠-132.11° 0.41∠35.77° 0.31∠177.78° 0.14∠-33.37° 0.05∠-42.37°

0.01∠-90.00° 0.08∠84.92° 0.09∠61.91° 0.25∠-139.35° 0.48∠33.26° 0.31∠174.67° 0.08∠-7.74°

0.01∠142.00° 0.06∠-85.55° 0.03∠97.14° 0.09∠56.82° 0.31∠-134.84° 0.46∠37.03° 0.29∠-178.74°

0 0.02∠-70.87° 0.05∠-76.69° 0.03∠72.41° 0.09∠73.95° 0.29∠-133.32° 0.52∠34.79°

65
 Table A-2 CFL Matrix

(CFL: 15W with 120V AC power)

0.0016∠38.00° 0.0023∠-133.36° 0 0.0015∠-97.12° 0.0018∠-88.29° 0.002∠-57.10° 0.0023∠-79.83°

0.0012∠-81.00° 0.0058∠83.31° 0.0037∠-148.97° 0.0036∠63.03° 0.0045∠-113.27° 0.0034∠144.30° 0.0049∠139.50°

0.0008∠172.00° 0.0059∠--60.59° 0.0076∠78.08° 0.0079∠-123.18° 0.0076∠-22.81° 0.0019∠-50.85° 0.0039∠7.03°

0 0.0034∠107.18° 0.0145∠-31.62° 0.0207∠71.80° 0.0069∠-152.46° 0.0101∠33.42° 0.0094∠-57.11°

0 0.0011∠108.69° 0.0073∠-171.54° 0.0132∠-64.63° 0.0134∠80.29° 0.0149∠-143.40° 0.0098∠-27.19°

0.0026∠-86.56° 0.0259∠-44.56° 0.0469∠-4.72° 0.0457∠69.96° 0.0286∠-8.84° 0.0665∠80.06° 0.0423∠-52.52°

0 0.0020∠--61.40° 0.0025∠-23.92° 0.0028∠98.85° 0.0075∠-166.76° 0.0136∠--46.33° 0.0230∠80.89°

66
 Table A-3 PAVC Matrix

(DBR: 0.9kW with 120V AC power)

0.0966∠ -5.00° 0.0178∠-174.81° 0.0101∠164.85° 0 0 0 0

0.0080∠156.00° 0.0771∠3.7301° 0.0137∠96.76° 0.0030∠173.73° 0 0.0012∠86.00° 0

0.0046∠7.00° 0.0118∠-6.60° 0.0839∠5.01° 0.0034∠68.28° 0.0016∠98.91° 0.0008∠-153.50° 0.0003∠-113.10°

0.0010∠-86.00° 0.0011∠-59.29° 0.0028∠-61.91° 0.0908∠5.54° 0 0 0

0 0.0010∠68.94° 0.0007∠-73.71° 0.0009∠-72.93° 0.0929∠4.3399° 0 0

0 0.0010∠32.54° 0 0 0.0286∠-8.84° 0.0940∠80.06° 0

0 0.0006∠10.60° 0 0 0 0 0.0945∠2.80°

67
 Appendix B

Line Data Calculation of the Study System

Line Construction Types:

B a
b
A 18”
a 10” c
b

30” 30” 18”

42” c

36”
G
Ground wire position is at 45
feet for all pole types G

c
C 6”
c’ D a b
24” 24”

36” Same for A and

B phases 30”

G
G

68
 Table B-1 Line Data of the Study System

Line Line Length Conductor

From Bus To Bus Phasing
Number (Miles) Type and Code

1 10 20 0.48 ABC A 336

2 20 30 0.4 ABC A 336
3 30 31 0.2 AB D 4/0
4 30 34 0.45 ABC B 4/0
4 30 40 0.5 ABC B 336
5 31 32 0.2 A C 4/0
6 31 33 0.2 B C 4/0
7 40 41 0.5 ABC B 4/0
8 40 50 0.6 ABC B 336
9 50 51 0.4 AB D 4/0
10 50 60 0.55 ABC B 336
11 60 61 0.4 ABC B 4/0
12 60 62 0.6 C C 4/0

In the initial design all phase conductors are either 336,400 - 26/7 strands ACSR or 4/0 -
6/1 strands ACSR. See line data for each line section’s configuration and size. Neglect
line capacitance. Ground conductors are all 2/0 – 6/1 strands ACSR

Type A336 Impedance:

0.4503 + 1.0721𝑖 0.1722 + 0.5045𝑖 0.1723 + 0.4265𝑖

𝑍𝑎𝑏𝑐 = [0.1722 + 0.5045𝑖 0.4501 + 1.0728𝑖 0.1722 + 0.5045𝑖 ] (Ω/mi)
0.1723 + 0.4265𝑖 0.1722 + 0.5045𝑖 0.4503 + 1.0721𝑖

69
Type B4/0 Impedance:

0.5450 + 1.4311𝑖 0.1021 + 0.4886𝑖 0.1053 + 0.3924𝑖

𝑍𝑎𝑏𝑐 = [0.1021 + 0.4886𝑖 0.5497 + 1.4134𝑖 0.1082 + 0.4665𝑖 ] (Ω/mi)
0.1053 + 0.3924𝑖 0.1082 + 0.4665𝑖 0.5574 + 1.3863𝑖

Type B336 Impedance:

0.3780 + 1.2981i 0.1021 + 0.4886i 0.1053 + 0.3924i

𝑍𝑎𝑏𝑐 = [ 0.1021 + 0.4886i 0.3827 + 1.2804i 0.1082 + 0.4665i] (Ω/mi)
0.1053 + 0.3924i 0.1082 + 0.4665i 0.3904 + 1.2533i

Type C4/0 Impedance:

𝑍𝑐 = [0.3333 + 0.9913𝑖] (Ω/mi)

Type D4/0 Impedance:

0.5560 + 1.3908𝑖 0.1110 + 0.3378𝑖

𝑍𝑎𝑏 = [ ] (Ω/mi)
0.1110 + 0.3378𝑖 0.5560 + 1.3908𝑖

70
 Appendix C

Simulation Models and Matlab Codes

1. Test System Simulation Model

-T- + v [V32]
+ i -
-

Discre te ,
Ts = 1e -005 s.

powe rgui
Conn1
-T-
In1 i
+ -

C
A
B
Conn2
+ v [V33]
-
In2

Aa

Bb

Cc
B40
Conn3

Subsystem2

A
aA A

B
bB B

C
cC C

Bus10(slack_bus) B41
Aa

Bb

Cc

B30
Aa

Bb

Cc

B20

C
A
B
aA

Conn1 bB
+
In1 v [V51_A]

Aa

Bb

Cc
cC B50 -
A
B
C

Conn2 B34
A
B
C

A
B
C [I51_A]
In2
i
Conn3 +
-
Subsystem1
[I51_B]
+ i
-

+
- v
[V51_B]

[I62]

i
+
-

+
- v
[V62]
Aa

Bb

Cc

B60
aA A
bB B
cC C
B61
A
B
C

Ia

-T- Vabc Ib

Ic
From
Nonlinear_load_model

Conn1
29+273.15 In1

Temperature Conn2
[K]
In2
789 Conn3

Subsystem

Figure C-1: Microgrid Test System

71
2. Composite Nonlinear Load Simulation Model
1 Ia

+
C ontrolled C urrent Source

s
-
In1 Out1

2 Ib
nonlinear_load_a
1

+
Vabc
C ontrolled C urrent Source1

s
-
In1 Out1

nonlinear_load_b
3 Ic

+
C ontrolled C urrent Source2
In1 Out1

-
nonlinear_load_c

Figure C-2: Three Phase Nonlinear Load Connection

DBR 0.67kW per

Constant1

Load in W
DBR percentage

10e3 0.1 1/670

Matrix
In1 Out1
Multiply

current_waveform

1
Out1

1 Voltage v ol_matrix
In1

matrix_cal
Load in W
CFL percentage

10e3 0.1 1/15

CFL15W Add

Y1

Constant3

Matrix
In1 Out1
Multiply

current_waveform1

Load in W
PAVC percentage

10e3 0.1 1/900

PAVC 0.9 kW

Y2

Constant6

Matrix
In1 Out1
Multiply

current_waveform2

Figure C-3: Nonlinear Load by Cross Frequency Admittance Matrix

72
 1 Voltage
m agnitude |u|
signal
angle 1/360*2*pi u

Fourier
Gain1

1/sqrt(2)
m agnitude |u|
signal
angle 1/360*2*pi u 1/sqrt(2)

Fourier1
Gain2
1/sqrt(2)

Reshape 1/120 1
m agnitude |u|
vol_matrix
signal
Reshape
angle 1/360*2*pi u 1/sqrt(2)
(7) to [7x1]
Fourier2 column-major
Gain3

1/sqrt(2)

m agnitude |u|
signal
angle 1/360*2*pi u 1/sqrt(2)
Fourier3 Gain4

1/sqrt(2)
|u|
m agnitude
signal
1/360*2*pi u
angle

Fourier4 Gain5

|u|
m agnitude
signal
1/360*2*pi u
angle

Fourier5 Gain6

|u|
m agnitude
signal
1/360*2*pi u
angle

Fourier6 Gain7

Figure C-4: Voltage Matrix Formation

3. Active Power Filter Inverter Control

Current Compensation Control

Vs Vabc
ICabc

Iload Iabc

ICabc1 [Icomp]
Vdc Ploss Ploss

PI controller
+
v - I_ref
g
I_meas

g
+
Iabc
A A
a
B B
A A
- b B B
C C c
C C

a 1 b 2 c 3

Figure C-5: Active Power Filter Inverter Control

73
4. DC Battery Droop Control Blocks

5
1 abc
Ic dq0 ICd 6
sin_cos ICq

abc_to_dq0 Terminator
Transformation

2 abc 3
VId dq0 VIdd 4
sin_cos
VIdq
abc_to_dq0 Terminator1
Transformation1
4
sin In Me an
Product 1

Discrete P
Add
Mean value

Product1

Product2 In Me an 2
Q
Add1
Discrete
Product3 Mean value1

3 abc
Iinv dq0
sin_cos

abc_to_dq0 Terminator2
Transformation3

Figure C-6: Active Power Calculation

3
w
5
f
f
1
s
mod sin
Integrator
3 2*pi 4
Trigonometric
fn sin
Gain1 2*pi Math Function
1 5e-6
Function cos
Pn Constant
Gain Trigonometric
2
Function1
P
sin

Trigonometric
Function2

1
VIdd*

Product abc
dq0 2
sin_cos VIdq*

abc_to_dq0
Transformation
Terminator
4 1e-5
Q
Gain2 Product1

5 (2/3)*pi sin
E0
Add Trigonometric
Constant3
Function3
Product2

sin
(2/3)*pi
Add1 Trigonometric
Constant4 Function4

Figure C-7: Voltage Reference in dq Axis

74
 Scope1

1 PI k Memory
VIdd* dq0
2 Discrete abc Signal(s) Pulse s 1
VIdd PI C ontroller sin_cos PWM
-K- 0 PWM Generator
5 dq0_to_abc
0 Product1 Iod Gain2 Constant1 Transformation
8
Constant
7 Product sin
w Product2 5

k1 Memory1
3 PI
6 -K-
VIdq*
Discrete Ioq
Gain3
4 PI C ontroller1
VIdq

Scope

Figure C-8: PWM Signal Generation

5. System Impedance Calculation Example (from Bus 10 to Bus 30)

%%Calculate impedance from Bus 10 TO BUS 30

%%Phase ABC
%%From Bus 10 To Bus 20 length 0.48mi
%%A336

%% Impedance ohms per mile

Zaa=0.3733+i*1.413;
Zbb=0.3733+i*1.413;
Zcc=0.3733+i*1.413;

Zab=0.09528+i*0.8451;
Zac=0.09528+i*0.7674;
Zbc=0.09528+i*0.8451;

Zww=0.8013+i*1.603;
Zwa=0.09528+i*0.7857;
Zwb=0.09528+i*0.7848;
Zwc=0.09528+i*0.7857;

Zabc=[Zaa Zab Zac Zwa;

Zab Zbb Zbc Zwb;
Zac Zbc Zcc Zwc;
Zwa Zwb Zwc Zww;]

%******************************
%Kron Reduction

75
 K=Zabc(1:3,1:3);
L=Zabc(1:3,4);
M=Zabc(4,4);
Lt=Zabc(4,1:3);
Zabc_sim=K-L*inv(M)*Lt

%Length 0.48 mi
Zabc_10_20=Zabc_sim*0.48;

%%From Bus 20 To 30 length 0.4mi

Zabc_20_30=Zabc_sim*0.4;

Zabc_com=[Zabc_10_20 Zabc_20_30;
Zabc_20_30.' Zabc_10_20+Zabc_20_30;];

%%Impedance from Bus 10 To Bus 30

Zabc10_30=Zabc_com

6. Crossed Frequency Admittance Matrix Calculation Example (PAVC)

%%%%%%%%%%%%%PAVC Matrix 0.9kW %%%%%%%%%%%%%%

%% in this case only 1st, 3rd, 5th, 7th, 9th, 11th, 13th, 15th
Y2=zeros(7,7);

%%120 V rms
V1=120;

I1=zeros(1,7);
I1(1,1)=16.4*cos(-5/180*pi)+16.4*sin(-5/180*pi)*j;
I1(1,2)=1.36*cos(156/180*pi)+1.36*sin(156/180*pi)*j;
I1(1,3)=0.78*cos(7/180*pi)+0.78*sin(7/180*pi)*j;
I1(1,4)=0.172*cos(-86/180*pi)+0.172*sin(-86/180*pi)*j;
I1(1,5)=0.057*cos(171/180*pi)+0.057*sin(171/180*pi)*j;
I1(1,6)=0.044*cos(72.8/180*pi)+0.044*sin(72.8/180*pi)*j;
I1(1,7)=0.034*cos(-1/180*pi)+0.034*sin(-1/180*pi)*j;

for k=1:1:7;
Y2(k,1)=I1(1,k)/sqrt(2)/V1;

end

%%120*10% V rms 60degree 3rd harmonics

76
V3=120*0.1*cos(60/180*pi)+120*0.1*sin(60/180*pi)*j;

I2=zeros(1,7);
I2(1,1)=16.3*cos(-6/180*pi)+16.3*sin(-6/180*pi)*j;
I2(1,2)=1.85*cos(111/180*pi)+1.85*sin(111/180*pi)*j;
I2(1,3)=0.93*cos(16/180*pi)+0.93*sin(16/180*pi)*j;
I2(1,4)=0.174*cos(-80/180*pi)+0.174*sin(-80/180*pi)*j;
I2(1,5)=0.07*cos(162/180*pi)+0.07*sin(162/180*pi)*j;
I2(1,6)=0.06*cos(78.2/180*pi)+0.06*sin(78.2/180*pi)*j;
I2(1,7)=0.038*cos(12.5/180*pi)+0.038*sin(12.5/180*pi)*j;

for k=1:1:7;
Y2(k,2)=(I2(1,k)/sqrt(2)-Y2(k,1)*V1)/V3;

end

%%120*5% V rms 100degree 5th harmonics

V5=120*0.05*cos(100/180*pi)+120*0.05*sin(100/180*pi)*j;

I3=zeros(1,7);
I3(1,1)=16.4*cos(-5.3/180*pi)+16.4*sin(-5.3/180*pi)*j;
I3(1,2)=1.45*cos(159/180*pi)+1.45*sin(159/180*pi)*j;
I3(1,3)=0.98*cos(53/180*pi)+0.98*sin(53/180*pi)*j;
I3(1,4)=0.16*cos(-79/180*pi)+0.16*sin(-79/180*pi)*j;
I3(1,5)=0.052*cos(167/180*pi)+0.052*sin(167/180*pi)*j;
I3(1,6)=0.045*cos(69/180*pi)+0.045*sin(69/180*pi)*j;
I3(1,7)=0.035*cos(1/180*pi)+0.035*sin(1/180*pi)*j;

for k=1:1:7;
Y2(k,3)=(I3(1,k)/sqrt(2)-Y2(k,1)*V1)/V5;

end

%%120*5% V rms 50degree 7th harmonics

V7=120*0.05*cos(50/180*pi)+120*0.05*sin(50/180*pi)*j;

I4=zeros(1,7);
I4(1,1)=16.4*cos(-5/180*pi)+16.4*sin(-5/180*pi)*j;
I4(1,2)=1.37*cos(157/180*pi)+1.37*sin(157/180*pi)*j;
I4(1,3)=0.77*cos(9/180*pi)+0.77*sin(9/180*pi)*j;
I4(1,4)=0.645*cos(46/180*pi)+0.645*sin(46/180*pi)*j;
I4(1,5)=0.05*cos(173/180*pi)+0.05*sin(173/180*pi)*j;
I4(1,6)=0.04*cos(69/180*pi)+0.04*sin(69/180*pi)*j;
I4(1,7)=0.034*cos(-5/180*pi)+0.034*sin(-5/180*pi)*j;

77
for k=1:1:7;
Y2(k,4)=(I4(1,k)/sqrt(2)-Y2(k,1)*V1)/V7;

end

%%120*5% V rms 45degree 9th harmonics

V9=120*0.05*cos(45/180*pi)+120*0.05*sin(45/180*pi)*j;

I5=zeros(1,7);
I5(1,1)=16.4*cos(-5/180*pi)+16.4*sin(-5/180*pi)*j;
I5(1,2)=1.36*cos(156/180*pi)+1.36*sin(156/180*pi)*j;
I5(1,3)=0.77*cos(7.7/180*pi)+0.77*sin(7.7/180*pi)*j;
I5(1,4)=0.165*cos(-85/180*pi)+0.165*sin(-85/180*pi)*j;
I5(1,5)=0.76*cos(53/180*pi)+0.76*sin(53/180*pi)*j;
I5(1,6)=0.042*cos(68/180*pi)+0.042*sin(68/180*pi)*j;
I5(1,7)=0.035*cos(-4/180*pi)+0.035*sin(-4/180*pi)*j;

for k=1:1:7;
Y2(k,5)=(I5(1,k)/sqrt(2)-Y2(k,1)*V1)/V9;

end

%%12*0.1 rms 70 degree 11th harmonics

V11=120*0.1*cos(70/180*pi)+120*0.1*sin(70/180*pi)*j;

I6=zeros(1,7);
I6(1,1)=16.4*cos(-5/180*pi)+16.4*sin(-5/180*pi)*j;
I6(1,2)=1.38*cos(156/180*pi)+1.38*sin(156/180*pi)*j;
I6(1,3)=0.78*cos(6/180*pi)+0.78*sin(6/180*pi)*j;
I6(1,4)=0.17*cos(-88/180*pi)+0.17*sin(-88/180*pi)*j;
I6(1,5)=0.063*cos(165/180*pi)+0.063*sin(165/180*pi)*j;
I6(1,6)=1.64*cos(72/180*pi)+1.64*sin(72/180*pi)*j;
I6(1,7)=0.038*cos(1/180*pi)+0.038*sin(1/180*pi)*j;

for k=1:1:7;
Y2(k,6)=(I6(1,k)/sqrt(2)-Y2(k,1)*V1)/V11;

end

%%120*0.05 rms 30 degree 13th harmonics

V13=120*0.05*cos(30/180*pi)+120*0.05*sin(30/180*pi)*j;

I7=zeros(1,7);

I7(1,1)=16.4*cos(-5/180*pi)+16.4*sin(-5/180*pi)*j;
I7(1,2)=1.36*cos(156/180*pi)+1.36*sin(156/180*pi)*j;
I7(1,3)=0.78*cos(6.8/180*pi)+0.78*sin(6.8/180*pi)*j;
I7(1,4)=0.172*cos(-86/180*pi)+0.172*sin(-86/180*pi)*j;
I7(1,5)=0.06*cos(170/180*pi)+0.06*sin(170/180*pi)*j;

78
I7(1,6)=0.045*cos(73/180*pi)+0.045*sin(73/180*pi)*j;
I7(1,7)=0.83*cos(31.5/180*pi)+0.83*sin(31.5/180*pi)*j;

for k=1:1:7;
Y2(k,7)=(I7(1,k)/sqrt(2)-Y2(k,1)*V1)/V13;

end

Y2

for x=1:1:7
for y=1:1:7

fprintf('%4.4f %4.4f\n',abs(Y2(x,y)),angle(Y2(x,y))/pi*180);

end
end

79