Journal Pre-proof

Evaluation and optimization of OFF-GRID and on-grid photovoltaic power system for
typical household electrification

Qusay Hassan

PII: S0960-1481(20)31416-6
DOI: https://doi.org/10.1016/j.renene.2020.09.008
Reference: RENE 14157

To appear in: Renewable Energy

Received Date: 14 March 2020

Revised Date: 14 August 2020
Accepted Date: 1 September 2020

Please cite this article as: Hassan Q, Evaluation and optimization of OFF-GRID and on-grid
photovoltaic power system for typical household electrification, Renewable Energy (2020), doi: https://
doi.org/10.1016/j.renene.2020.09.008.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2020 Published by Elsevier Ltd.

Qusay Hassan: Conceptualization, Supervision, Investigation, Methodology, Software, Writing - original
draft, Writing - review & editing

of
ro
-p
re
lP
na
ur
Jo
1 EVALUATION AND OPTIMIZATION OF OFF-GRID AND ON-GRID PHOTOVOLTAIC
2 POWER SYSTEM FOR TYPICAL HOUSEHOLD ELECTRIFICATION
3 Qusay Hassan
4 College of Engineering, Department of Mechanical Engineering, University of Diyala, Iraq
5 Correspondence author email: [email protected]
6
7
8 Abstract
9 This study aims to design a renewable energy system that can meet the desired electrical load of
10 households with low energy cost, high renewable energy fraction and low CO2 emissions.
11 Photovoltaic solar power systems used to electrify typical households in Iraq were investigated
12 through simulation and optimisation. One-minute simulations and optimisations were performed to

of
13 determine the performance and net present cost of two photovoltaic power system configurations,
14 namely (i) off-grid and (ii) on-grid solar photovoltaic power systems. Results show that the two

ro
15 systems exhibit excellent performance, but the on-grid photovoltaic power system requires cheaper
16 cost compared with the off-grid PV power system. The total energy generated from the off-grid PV
17 power system meets the desired electrical load of households and recharges the batteries, whereas the
18
19
-p
excess electricity from the on-grid photovoltaic power system feeds the grid. The two designed
systems are environmentally friendly and economically viable. The total net present cost of the off-
re
20 grid solution is $6,244, and its energy cost is $0.196/kWh. By contrast, the total net present cost of the
21 on-grid system is $6,115, with energy cost equal to $0.183/kWh. The obtained results confirm the
22
lP

suitability of photovoltaic power systems for electrifying single households in addition to feeding the
23 national grid.

24 Keywords: Economic analysis, Off-grid photovoltaic system, On-grid photovoltaic system, Solar
na

25 energy, Technical analysis.

26 1. Introduction
ur

27 In the last decades, Middle Eastern countries have not adopted renewable energy resources due to the
28 available oil supply. This policy has changed in the last decade. Governments in that region have devoted
Jo

29 their efforts on considering alternative forms of energy and environmental protection as essential
30 priorities. The electricity sector in Iraq inefficient due to the strike caused by country destruction in 1991
31 and the growing energy demand of its rising population. At present, the electricity generation in Iraq is
32 insufficient to meet the industrial and domestic sectors desired energy. Solar energy is the most promising
33 renewable energy source in Iraq. The exploitation of this source provide clean energy and reduce carbon
34 dioxide emissions, and other harmful gases generated from fossil fuels burning used in power plants. The
35 annual average of incident solar radiation in Baghdad is approximately 5.02 kWh/m2/day [1], indicating
36 that Iraq has great potential to become a global renewable energy supplier by relying on solar energy.
37 Kazem and Chaichan [2] reviewed the renewable energy resources, and the attempts encourage the Iraqi
38 government to exploit these sources because of inadequate use, especially wind, solar and biomass. Their
39 work also examined the lack of electricity and many other energy challenges facing Iraq. Dhrab and
40 Sopian [3] investigated resources of the renewable energy in Iraq to respond to the increasing energy
41 demands associated with reduced resources of fossil fuel and concerns about their adverse environmental
42 impacts. Al-Riah et al. [4] studied the solar energy resources of three cities in Iraq (Baghdad, Nasiriyah
43 and Mosul) from 1971–1985, and they determined the daily average of the global incident solar radiation
44 per year and the duration of annual shining days. Zwaan et al. [5] reviewed the potential renewable
45 energy resources in the Middle East in addition to integrated the energy systems to assess the correlation

1
46 of renewable energy resources. Their study investigated the energy technology requirements to meet the
47 stringent global climate guidance for stabilising the concentrations of greenhouse gases in the atmosphere.
48 This environmental care goal involves the widespread of renewable energy resources deployment in the
49 Middle East, which is expected to be represented by wind and solar energy by around 60% of the total
50 electricity supply by 2050.
51 The CO2 emissions generated from fuel burning in Iraq represents approximately 3.9% of the total
52 emissions since 1960, where the highest value is recorded in 1964 at approximately 16.09%, and the
53 lowest value is reported in 1968 at approximately 1.21% [6]. This consumption caused by the growing
54 population since 1960 reflects an increase in transport demand (the increasing demand for private cars,
55 increase in total fuel consumption by approximately 6% annually, increased oil production and related
56 burnings, increased human industrial activity, and increased carbon dioxide emissions since the 1950s.
57 Power plants have been affected by a shortage of raw materials because of the economic embargo

of
58 imposed on Iraq since 1991. Subsequently, energy production has fluctuated and resulted in the
59 dependence on small generators in many areas, causing an increase in environmental pollution of cities

ro
60 and rural areas.
61 Off-grid and on-grid solar energy systems can be used in households. Hassan et al. [7] presented a design
62
63
-p
and analysed the off-grid photovoltaic (PV) system for village electrification in a rural site in Iraq. Their
study confirmed that the use of PV systems for electrification is suitable for long-term investments with
re
64 the cost of $0.51/kWh. Abdulateef [8] used MATLAB software to design a solar off-grid PV system for
65 supplying the required electricity to residential units in Malaysia. The results showed that the PV array of
66 44 solar modules meets the desired load profile (36 kWh/day). The PV system can supply 12.4 kWh of
lP

67 energy during the day, and the storage units (battery) storage system can store 23.6 kWh (48V, 350 Ah) at
68 night and can meet the desired load. Al-Shamani et al. [9] presented an analysis component required for
na

69 the off-grid PV system design that can power electrical appliances for irrigation near Hilla City, Iraq.
70 Most studies [7–9] have ignored the solar intensity and ambient temperature effects on energy production
71 during simulation. These parameters highly affect the energy degradation from PV modules, especially in
ur

72 hot states, such as Iraq. An off-grid PV system design for the greenhouse was analysed by Elbreki et al.
73 [10] in Sabha City, Libya. Their study analysed the feasibility of PV power system generation for remote
Jo

74 sites. Their results show that the off-grid PV system is competitive compared with diesel-powered
75 installations on the basis of the available incident solar radiation in that site. Alamsyah et al. [11]
76 designed and analysed a PV system for the electrification of single residential households In Malaysia.
77 They evaluated the effects of solar radiation intensity and PV module temperature variation on the power
78 generated by PV panels. The results showed that the generated by PV system electricity for the household
79 is beneficial in terms of the energy prices in that country.
80 Palej et al. [12] used PV modules and wind turbines to feed the electrical demand of households in
81 Poland. Their results reported that the optimal system configuration through optimisation requires low
82 unit cost of energy (COE) compared with other optimisation objectives. Ceran et al. [13] presented an
83 economic analysis of a PV system integrated with wind turbines, fuel cells and an electrolyser with a
84 hydrogen tank working as the energy storage. Loads of three different households were simulated on the
85 basis of experimental data in terms of incident solar radiation, wind speed and ambient temperature. Their
86 results showed the effectiveness of the proposed approach at the optimal configuration and the limits of
87 investment costs necessary to build the system in detail. Styszko et al. [14] and Jaszczur et al. [15]
88 investigated the energy degradation of a PV system caused by the accumulative dust deposition on the
89 module front cover glass during heating (summer) and non-heating (winter) seasons. They presented a
90 new model for predicting the energy degradation caused by accumulative dust deposition on the PV

2
 91 module. Onat [16] conducted technical and economic analyses of the off-grid PV system of seven regions
92 in Turkey. He developed an algorithm that can be used to calculate the life-cycle cost, COE rate of
93 electrical capacity utilisation, monthly energy generated/excess, initial investment/replacement cost, and
94 operating and maintenance costs for each component in the system. Ghafoor and Munir [17] evaluated an
95 off-grid PV system electrification for households in Faisalabad City, Pakistan. The model specifications
96 were as follows: peak power: 1,928 Wp, area of PV modules: 12.85 m2; battery backup capacity: 9,640.5
97 Wh, size of charge controller: 56.65 A, and inverter power: 1,020 W. The results illustrated that the COE
98 generated by the analysed system is cheaper than the COE from the electric grid supply on the basis of the
99 project lifetime. In Malaysia Nordin and Rahman [18] presented a novel optimisation method for the
100 sizing of off-grid PV for household electrification. The optimisation started by computing the loss power
101 supply probability as the benchmark to determine the possible PV array size and battery capacity. The
102 results demonstrated that the optimisation methodology sizing the PV array firstly and then sizing the
103 battery. In Turkey Olcan [19] conducted a technical and economic analysis with optimisation of an off-

of
104 grid PV system to determine the size of components for a water pumping system used for irrigation of
105 citrus trees. The analysis was conducted on the basis of the reliability and cost indicators in addition to

ro
106 life-cycle costs and power supply. The results showed that the technical and economical sizing method is
107 more suitable than the computational algorithm. Riza et al. [20] designed and optimised an off-grid PV
108
109
-p
power system with battery for lighting system in Indonesia. The results demonstrated that the designed
space approach provides more advantages than using other optimisation methods. Al-Falahi et al. [21]
110
re
reviewed sizing optimisation methods for off-grid PV solar systems. Hassan et al. [22] presented a
111 detailed mathematical model for PV power calculation using an arbitrary PV system. The study analysis
112 also the PV module energy degradation due to the PV cell temperature effect. Zhang et al. [23] optimised
lP

113 an off-grid hybrid solar/wind energy system using three optimisation algorithms (simulated annealing,
114 harmony search and chaotic search) for Island electrification in Iran. The considered system capacity 40
115 kW, were the components optimised using an economic objective to obtain the lowest system life-cycle
na

116 cost $170,900. The authors compared the numerical results of three different algorithms and confirmed
117 that the forecasted renewable energy resources are reliable to feed the desired load. Ma et al. [24]
ur

118 demonstrated the techno-economic feasibility of off-grid hybrid energy systems consisting of
119 wind/solar/battery components to feed a remote island in Hong Kong. The results demonstrated that the
120 use of diesel generator can completely replace the hybrid off-grid system. The energy generated by the
Jo

121 system at the optimum configuration generates 16% from wind turbine units and 84% from the PV array,
122 and the levelised COE is $0.595/kWh. Jaszczur et al. [25] examined the effects of ambient temperature
123 and deposited dust on the energy degradation of a PV system. Their results showed that the observed
124 deposited dust for 1-week exposure periods on a single PV module is approximately 480.0 mg, leading to
125 a loss in module efficiency of approximately 2.1%. Mokheimer et al. [26] introduced a method for size
126 optimisation of an off-grid hybrid power system consists of a WT/PV feeding water desalination system
127 in Saudi Arabia. The model study showed that the PV component feed system by about 73% while the
128 WT components have only 22% of the total energy generated. In Tunisia Bourouni et al. [27] and Cherif
129 and Belhadj [28] proposed a method based on genetic algorithms for optimisation of an off-grid
130 wind/PV/battery power system for remote water desalination system. Their results showed the weather
131 data in the selected site is very helpful for a long period of energy production to feed such systems.
132 Koutroulis and Blaabjerg [29] proposed an interesting methodology for the optimal sizing of an off-grid
133 PV power system. The proposed methodology aimed to select the components of the system from the
134 commercially available in the market that is guaranteed to minimise the total system cost of 20 working
135 years. which the used methodology can improve system configuration. Tina et al. [30] introduced a
136 potential probabilistic approach for integrating the volatile nature of load and resources. The study

3
137 purpose that can eliminate the need for data series for assessing the long-term performance of the
138 renewable power system for off-grid and on-grid connected applications to feed the Island electrification
139 in Italy. Huang et al. [31] modelled and sized an off-grid PV power system to meet 15 kW peak power
140 demand in China. The conducted sizing method consisted of three steps: firstly, the entire mathematical
141 system model was developed on the basis of the battery capacity fade; secondly, the power management
142 strategy was set for the protection of system component; and thirdly, the optimisation model that uses the
143 numerical results was established. The project was simulated for a five-year operating period. The results
144 using the proposed indicator showed that the reliability of the designed system is weakened because the
145 capacity loss is ignored. Pudjianto et al. [32], Keppler and Cometto [33] designed long-term storage
146 dispatchable backup for a PV system to meet the desired load independent of the intermittent nature of
147 solar energy. Ueckerdt et al. [34] used a distribution method of total costs to find the optimum size of the
148 PV system. The analysed costs include grid purchase, production and grid sell costs on the basis of the
149 load curve principle. This method includes arranging the power capacity required for each unit of time in

of
150 descending order. Rai N. and Rai B. [35] used fuzzy logic control for an off-grid PV–battery power
151 system to meet the required electrical load in India. The results of this methodology show that the battery

ro
152 life spans are prolonged because of the reduction in peak current of the battery up to 0.22%. Kaushika et
153 al. [36] presented a sizing model for an off-grid PV power system with interconnected arrays in India.
154
155
-p
Their simulation considered the generated electricity in the suggested PV array storage in the batteries to
serve the desired load. They also used the probability of the loss power supply to include not meeting load
156
re
demand risk. Results are presented with the support of a numerical example with the fluctuated load
157 demand assumption, the number of batteries and PV modules corresponding to the potential for power
158 loss during the year-round operation cycle.
lP

159 Fara and Craciunescu [37] proposed and simulated an off-grid PV system using MATLAB/Simulink
160 software for energy production to households located in mountainous areas in Romania. The off-grid PV
na

161 system was developed to ensure the annual electricity supply for various types of equipment and lighting
162 with low energy consumption. The results calculated using this approach show that the overall
163 performance of these systems depend on the intensity of incident solar radiation and the battery state of
ur

164 charge. Rezk et al. [38] optimised an off-grid PV-battery power system for irrigation in a rural site in
165 Egypt. The designed power system supplies power to water pumping with the desalination system. The
Jo

166 suggested power system was optimised on the basis of economic objective (minimise net present and
167 energy costs). The results showed that the PV power system with a size of 75 kW, battery of 32 units
168 (2.16 kWh/unit) and a converter with a size of 29 kW has the optimum size configuration to meet the
169 desired electrical load. The total net present and energy costs recorded are $10,985 and $0.05/kWh,
170 respectively. Rezk and El-Sayed [39] conducted economic analysis optimisations for many scenarios of
171 hybrid power systems to meet the desired electrical load for a rural telecommunication site in Egypt. The
172 power system scenarios were configurated on the basis of solar PV and battery. The best configuration
173 results have the lowest net present and energy costs of $75,515 and $0.997/kWh, respectively, where the
174 energy cost depends on the performance and the type of renewable energy components. Ghenai et al. [40]
175 compared the off-grid and on-grid hybrid PV power systems for a water desalination station in the United
176 Arab Emirates. The results indicated that the on-grid PV system performs better than the off-grid PV
177 power system. The energy cost of the on-grid system is $0.8/kWh. The on-grid PV power system
178 possibility through fade the grid by excess electricity to reduce the cost of electricity. By contrast, the off-
179 grid PV system is better in relations of reducing greenhouse gas emissions.
180 In the literature review, various systems have been proposed in different locations and with different
181 objectives to optimise the system as shown in Table 1. The system configuration is site-dependent and
182 needs to be analysed using local environmental conditions and local energy consumption profile. No

4
183 analysis has been conducted in Iraq. Such analysis is highly required because of the lack of stable energy
184 supply and environmental protection.
185
186
187 Table 1 On/off-grid power systems investigated in the literature.

Scenario System component location Analysis type Purpose References

Off-grid PV-battery Iraq Modelling Vilage electrification Hassan et al. [7]

Off-grid PV-battery Malaysia Modelling Residential units Abdulateef [8]
Off-grid PV-battery Iraq Modelling Irrigation Al-Shamani et al. [9]
Off-grid PV-battery Libya Modelling Greenhouse Elbreki et al. [10]

of
Off-grid PV-battery Malaysia Experimental Residential households Alamsyah et al. [11]
On-grid PV-WT Poland Experimental Household electrification Palej et al. [12]

ro
On-grid PV-WT Poland Experimental Household electrification Ceran et al. [13]
Off-grid PV-battery Turkey Modelling Household electrification Onat [16]
Off-grid
Off-grid
PV-battery
PV-battery
Pakistan
Malaysia
-p
Modelling
Experimental
Household electrification Ghafoor and Munir [17]
Household electrification Nordin and Rahman [18]
re
Off-grid PV-battery Turkey Modelling Irrigation Olcan [19]
Off-grid PV-battery Indonesia Modelling Lighting Riza et al. [20]
lP

Off-grid PV-WT-hydrogen Iran Modelling Island electrification Zhang et al. [23]

Off-grid PV-WT-battery Hong Kong Modelling Island electrification Ma et al. [24]
na

Off-grid PV-WT Saudi Arabia Modelling Water desalination Mokheimer et al. [26]
Off-grid PV-WT-battery Tunisia Modelling Water desalination Bourouni et al. [27]
Off-grid PV-WT Tunisia Modelling Water desalination Cherif and Belhadj [28]
ur

Off-grid/
PV-WT Italy Modelling Island electrification Tina et al. [30]
On-grid
Jo

Off-grid PV-hydrogen China Modelling Electric vehicle charging Huang et al. [31]
Off-grid PV-battery India Modelling Multi-application Rai N. and Rai B. [35]
Off-grid PV-battery India Modelling Residential building Kaushika et al. [36]
Off-grid PV-battery Romania Modelling Household electrification Fara and Craciunescu [37]
Off-grid PV-battery Egypt Modelling Water pumping Rezk et al. [38]
Off-grid PV-battery Egypt Modelling Telecommunication Rezk and El-Sayed [39]
Off-grid/ PV-battery-Diesel United Arab
Modelling Water desalination Ghenai et al. [40]
On-grid genertaor Emirates
188
189 This work presents an optimum sizing and performance evaluation of two types of PV power systems,
190 namely, off-grid and on-grid, to feed the electrical load of a typical household in Iraq. The experimental
191 data of electrical load profile, incident solar radiation, ambient temperature and wind speed are used. The
192 two power systems (on/off-grid schemes) are analysed under the configuration of a fixed tilt angle
193 (annual optimal tilt angle), two-axis tracking system to illustrate the generated energy, in addtion the
194 ambient temperature effect on the energy degradation of the systems were invistigated.
195

5
196
197
198
199

200 2. Methodology
201 The experimental measurement of the electrical load demand, environmental data such solar radiation,
202 ambient temperature and wind speed were obtained from the weather station located at the University of
203 Diyala campus (latitude 33.7733° N, longitude 45.1495° E ) for one year from 01.01.2018 to 31.12.2018.
204 Load demand: an experimental data measured of the electrical load demand for a typical household in

of
205 Baqubah City, Iraq by using an energy meter type (AT3010) connected with data transmitter device
206 which recorded power with a temporal resolution at 1-min. The measurements have done for the period of
207 January 1, 2018, to December 31, 2018. The typical considered household is a reception, living room,

ro
208 kitchen and two-bedrooms with 200 m2 including garden and occupied by family consist, two adults and
209 three children. The cooking, hot water and heating system were powered by natural gas. A few devices
210
211
-p
with high nominal electrical power are used: fridge 0.8 kW, washing machine 1.1 kW, 8 Light Emitting
Diode (LED) lamps 10 W, 4 ceiling fan 0.2 kW and 2 air cooler 0.3 kW were air-cooler and ceiling fans
re
212 working only during the summer and washing machine working 2 times per week. The average of daily
213 energy consumption recorded 7.18 kWh/day. Figures 1(a) and 1(b) show the daily electrical load and
lP

214 monthly average energy consumption, respectively. It is observed from Figure 1(b) that during the
215 summertime higher electrical consumption than during winter time due to the used of air conditioning
216 system.
na

217 Solar radiation: the data of the incident solar irradiation measured on the horizontal plane by
218 pyranometer type (MS-40) located at Diyala University in Baqubah city, Iraq (33.7733° N, 45.1495° E).
ur

219 The pyranometers equipped with a data recorder and transmitter. The measurements obtained at 1-min
220 resolution, these measurements data recorded for the year 2018. The selected daily and monthly average
Jo

221 solar irradiance are shown in Figures 2(a) and (b) respectively. The annual average of the incident solar
222 radiation on the horizontal plane recorded approx. 5.14 kWh/m2/day.
223 Wind speed and ambient temperature: the wind speed and ambient temperature measured for the year
224 2018 from January 1, 2018, to December 31, 2018, at 5 min time step resolution. Figures 3 (a) and (b)
225 show the daily and monthly average of the ambient temperature, respectively. Figures 3 (c) and (d) show
226 the frequency and monthly average of the wind speed, respectively.
3.5 300

03.01.2018
(a) (b)
Monthly averaged electrical energy, kWh

3.0 02.07.2018
250

2.5
200

2.0
Load, kW

150
1.5

100
1.0

0.5 50

0.0 0
02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

227 Time Month

6
228 Figure 1 An experimental data for the daily load distribution (a), monthly averaged energy consumption (b).
1.0
8
04.01.2018
(a)
0.9 (b)
25.07.2018 7
0.8
6

Daily radiation,kWh/m /d
0.7
2
Solar radiation, kW/m

2
0.6 5

0.5 4

0.4
3
0.3
2
0.2

1
0.1

0.0 0
02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

229 Time Month

of
230 Figure 2 An experimental data for the incident solar radiation on the horizontal plane, daily (a) and monthly
231 averaged (b).

ro
50
60
28.06.2018
11.01.2018
(a) (b)
40
-p 50

40
Temperature, oC

30
Temperature, C

re
o

30
20
lP

20
10

10
0

0
na

-10 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Ann
02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36
-10
232 Time
ur

15 14
14 (c) (d)
13 12
12
Jo

11
10
10
Wind speed, m/s

9
Frequency,%

8
8
7
6
6
5
4 4

3
2 2
1
0 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

233 Value, m/s Month

234 Figure 3 An experimental data for the daily and monthly average ambient temperature in (a) and (b) respectively,
235 wind speed frequency and monthly average (c) and (d) respectively.

236
237
238
239

7
240
241 3. System component and model description
242 Two PV system scenarios are investigated in this work, as shown in Figure 4. Scenario (A) consists of PV
243 modules and batteries as storage units that can feed the desired load, and Scenario (B) only includes a PV
244 module connected with the grid that feeds the desired load.
245
246
PV
Scenario (A) PV
Scenario (B)
247
248

of
249 Converter
Converter

ro
250
AC home load AC home load

251 Battery
Grid

252
-p
re
253
254 Figure 4 PV power system: Scenario (A) off-grid and Scenario (B) on-grid.
lP

255
256 Photovoltaic module: photovoltaic modules is the primary renewable energy source unit in the designed
na

257 systems. The SAT-6PB photovoltaic module, with technical specifications at STC: nominal power 285
258 Wp, the voltage at nominal power 12V, solar cells per module 72 (6 x 12), NOCT 48 oC, module
259 temperature coefficient of power -0.28 %/oC are used. The CO2 emissions manufacturing process about
ur

260 600 kg/kW. The installation and replacement costs of this size are $ 110 with operation and maintenance
261 cost $ 5/yr with 25 years lifespan, the modules are orientated at slop degree β = 30o, and azimuth degree
Jo

262 γ = 0o for the south with 90% derating factor and 20% ground reflection. The temperature effect on the
263 module energy production is taken into consideration.
264 A mathematical model for PV electrical power generated from an arbitrary orientation PV module has
265 been presented by Hassan et al. [22] is used. The generated electrical power from PV module depends on
266 the total incident solar radiation GT on the module front surface and is affected by module tilted angle β
267 and module azimuth angle γ . The PV module generating power PPV can be calculated as:

 G 
268 PPV ,t = CPV DPV 1 + α P (Tc,t − Tc, STC )   T ,t  (1)
G
 T ,STC 

269 where CPV is the module rated capacity (kW), DPV is the module derating factor (%), αP is the module
270 temperature coefficient of power (%/°C), TC,t is the module temperature s GT,t is the simultaneous
271 incident irradiance (kW/m2) and GT,STC is the incident solar radiation at STC =(1 kW/m2).
272 The module electrical power depends on the cell temperature, where the cell efficiency at standard test
273 conditions (STCs) depend on the cell type (monocrystalline silicon 21%, polycrystalline silicon 17.0%–
274 18%, and thin-film amorphous silicon 8.5%) [41].

8
275
276 Several articles have presented formulas to estimate the temperature and thermal effect on PV module
277 performance. Santiago et al. [42] reviewed various models. Perović et al. [43] prepared a model for
278 estimating module temperature. The model for predicting module temperature described as [44]:
279
280 Tc ,t = GT k + Ta ,t (2)
281
282 where Ta,t is the ambient temperature (oC ), k is the Ross coefficient which represents mounting
283 conditions (cooling conditions) and ranges k = (0.02-0.056) typically (Km2/kW) [45,46], in this
284 calculations the value of k is taken 0.03).
285 The number of PV modules (array size) can be calculated as [46]:

of
Lt
286 PVarray = (3)
GT ,t + ηPV +ηbηinv + α p

ro
287 where Lt is the electrical load demand (kW), ηpv is the PV module efficiency, ηb is the battery efficiency
288

289
and ηinv inverter efficiency. -p
re
290 Batteries: the battery is the electrical energy storage unit used in case of an off-grid system. The
291 LUMINOUS model 12200L type Tubular with size (200 Ah, 12 volts) with 65.5 kg weight for the off-
lP

292 grid PV power system is selected for the simulation in this work [47]. The estimated battery bank lifetime
293 is 5 years, and the unit cost is $230.00/unit, and replacement costs is taken $130.00 /unit while the
294 maintenance cost expected at $ 0.1/hr, the charging percentage has taken 95%, manufacturing process
na

295 CO2 emissions generate about 55 kgCO2/kWh. The voltage battery model presented in [58-60] to model
296 lead-acid traction can be described as:
ur

 K 
297 Vb,t = VC ,t − DS ,t  Rb +  (4)
 SoC 
Jo

298 where Vb,t is the battery terminal voltage (V), VC,t is the simultaneous charge-dependent open-circuit, Rb
299 is the terminal resistor (ohm), K is the polarization constant, typically 0.1 (ohm), DS,t is the discharge
300 current (amps) and SoC is the state of charge (%). The battery overcharge occurs when the PV array
301 generated power higher than load demand. In such case the battery SoC reaches to the maximum value
302 SoC= 1 then the control stops charging process, meanwhile the SoC decreases to the lowest level the
303 control returns to charging the battery.
304 The SoC is the main effect on the battery power Pb,t depends on the battery state of charge and can be
305 calculated from equations [59]:
306
307 charging,
308 Pb,t = (Pn (t) − Pk (t) / (ηN ) ⋅ηb + Pb (t −1) ⋅ (1− sd ) (5)
309 discharging,
310 Pb,t = Pb (t −1) ⋅ (1− sd ) − (Pn (t) / (ηb ) ⋅ Pk (t)) (6)
311

9
312 where Pb(t-1) is the power at the end of interval t, Pk(t) the load demand at the time t, Pn(t) is the total
313 energy generated by photovoltaic modules at the time t, sd is the self-discharge factor, and ηb, ηN is the
314 battery charger and inverter efficiency, respectively.
315 Inverter (with control unit): an electronic power converter with the battery charge/discharge control
316 unit selected to be 20% higher than electrical load rated power. The RoHS hybrid type with the size 5.0
317 kW is considered in the proposed system which can work in both on-grid and off-grid system [51]. The
318 inverter cost is $450, cost of replacement is $ 250 with operation and maintenance cost $10/yr and the
319 unit lifetime estimate to be 15 years with an efficiency 85 %, (including cable losses). The inverter
320 manufacturing process generates about 60 kg/kW of CO2 emission.
321
322 4. Economic model
323 The PV power system optimal size can achieve the best compromise between two major approaches:

of
324 economical and technical objectives.

ro
325
326 Net Present Cost (CNPC) include all the cost accrue on the system during whole project lifespan, can be
327 expressed as [52]:

328 C NPC =
-p C ann (7)
CRF ( i , R p )
re
lP

329 where, Cann is the total annualised cost ($/year), CRF() is capital recovery factor, Rp is the project lifespan
330 (year) and i is the value of interest rate (%).
331 Cost of Energy (COE): in PV power systems cost of energy depends on several parameters initial
na

332 capital cost, operation and maintenance costs, energy takend and fade gird in case on-grid power
333 system, can be expressed as [53]:
ur

334 (a) Scenario (A):

Jo

Cann
335 COEScenario (A) = (8)
EnergyAC ,DC

336 (b) Scenario (B):

Cann
337 COEScenario (B) = (9)
EnergyAC ,DC + Energy fade grid

338 where Energy AC,DC is the energy cost serve the electrical load form both buses AC and DC ($/kWh),
339 primary load served, Energy fade grid is the total energy grid sales ($/kWh) to the grid and Cann is the total
340 annualised cost of the system that includes the annualised costs of each system component, with the
341 others, annualised costs.
342 Life cycle cost (LCC): is the economic feasibility of hybrid power systems including the total cost
343 incurred over the productive life of the system such as initial capital cost, installation cost, operation and
344 maintenance cost, replacement cost, fuel cost, salvage value etc. Those expenses are calculated for the

10
345 whole system lifetime, and the better system usually has lower LCC. This parameter can be calculated as
346 follows [54]:

347 LCC = Initial cost- Operation and Maintenance cost - Replacment cost-Slvage value (10)

348 5. Elecarical load and emission equations:

349 The equations of simultaneous electrical load that ensure feed the desired load from the systems
350 components and the amount on emissions for the selected scenario can be expressed as the following:
351
352 Scenario (A): the off-grid PV simultaneous load balance equation :

353 PL,t scenario (A) = PPV ,t + Pb,t (11)

of
354
355 the power flows between the system components can be expressed as [54]:
356

ro
Pb,t for PPV ,t = 0 ; Pb charge,t = 0

357 PL,t scenario (A) =  Pb,t + PPV ,t
P
 PV ,t
-p for PPV ,t < PL,t ; Pb charge,t = 0
for PPV ,t ≥ PL,t ; Pb charge,t ≥ 0
(12)
re
358
359 The amount of CO2 emission in the off-grid PV power system generated due to the manufacturing process of
lP

360 the system components, which can be calculated as follows:

361
362 CO2 scenario (A) = PVCO2 + BatteryCO2 + InverterCO2 (13)
na

363
364 Scenario (B): the on-grid PV simultaneous load balance equation:
ur

365 PL,t scenario (B) = PPV ,t + PG,t (14)

366
Jo

367 the power flows between the system components can be expressed as [54]:
368
 PG,t for PPV ,t = 0 ; PTo grid ,t = 0

369 PL,t scenario (B) =  PG,t + PPV ,t for PPV ,t < PL,t ; PTo grid ,t = 0 (15)
P for PPV ,t ≥ PL,t ; PTo grid ,t ≥ 0
 PV ,t
370
371 For on-grid PV power system amount of CO2 emission generated due to the manufacturing process of the
372 system components and the energy taken from the grid. In Iraq the natural gas in its electrical power stations
373 which generates 0.35 kgCO2/kWh [7] can be calculated as follows:
374
375 CO2 scenario (B) = PVCO2 + InverterCO2 +GridCO2 (16)
376 where PL,t, Pb,t, PPV,t , PG,t , PTo grid,t is the load, battery power, PV power, power from grid and power sent
377 to grid respectively in kW. PVCO2 , InventerCO2 , BatteryCO2 and GridCO2 is the CO2 emissions generated due
378 to the manufacturing process of PV, Inverter, Battery and energy taken from grid respectively that described
379 in system component and model description section.
380

11
381 Energy self-consumption(SC):
382 In the case of scenario (B), the self-consumption is the electrical energy generated by the PV component
383 and consumed locally. Self-consumption (SC) can be expressed as [54]:
384
385
386 S C = E co n su m ed / E g en e ra ted (17)
387 where Econsumed and Egenerated represent the energy consumed, and the excess Egenerated energy generated
388 energy in (kWh).
389
390 6. Optimisation methodology
391 Figure 5 shows the optimisation process flow chart. The decision variables included in the process of
392 optimisation are the number of PV modules (Npv) and number of batteries (Nbat) in case of scenario (A).

of
393 The whole year analysis performed using one minute's time step and included electrical load, incident
394 solar radiation, wind speed and the ambient temperature. The initial assumption of PV system

ro
395 configuration subject to the following constraint [58-60]:
396 Scenario (A):
397
-p
M in ( N p v ) ≥ 0 (18)
re
398 Scenario (B):
lP

399 M in ( N pv , N bat ) ≥ 0 (19)

400 Metalorgical data (year data)

na

401
402 Initial guess of (Npv, Nbat)
ur

Model of (PV, solar radiation,

403 Temperature)
Restriction inequalities
Jo

evaluation
404
PV system size (on-grid) or (off-
405 grid)

406
Load profile Battery model size (off-grid)
407
408 Fitness function evaluation

409 Yes
Optimum component
Criterion reached? size
410 No

411 Size selection operation

412
Crossover and mutation operation
413
414 New size configuration

12
415
416 Figure 5 Optimal sizing model flow chart.

417
418 7. Results and discussion
419 The analysis was performed on the basis of the experimental measurement of electrical load, solar
420 radiation, wind speed and ambient temperature. Scenarios (A) and (B) were modelled as electrical loads,
421 as presented in Figure 1. The analysis results were presented using 1-min time step simulations to avoid
422 the errors generated during the simulation process, as recommended by Jaszczur et al. [54]. The project
423 lifespan is 25 years, annual inflation is 2% , the annual discount rate is 2%, and interesting nominal rate is
424 4%. The prices of grid energy are based on the average electrical energy prices for 2018 in Iraq. The
425 energy purchase cost is 0.24 $/kWh, and the energy sold cost is 0.09 $/kWh. The PV modules were

of
426 installed in the yearly optimum orientation at the selected site (γ = 0°, β =30°), and the effects of wind
427 speed and ambient temperature on the PV module energy production were considered.

ro
428 This analysis mainly aims to verify the feasibility of PV solar systems to meet the electrical load needs of
429 a typical household in Iraq. The systems were analysed using optimisation functions with a nondominated
430 sorting genetic algorithm.
-p
re
9000 4600 260
Cycles to Failure
8000 (a) Lifetime Thrpt. 240
(b)
4400
lP

7000 220
Lifetime throughput, kWh

4200
Cycles of failure

6000 200
Capacity, Ah

4000
5000 180
na

3800
160
4000

3600 140
3000

120
ur

2000 3400

100
1000 3200
20 30 40 50 60 70 80 0 20 40 60 80 100 120
Jo

431 Depth of discharge, % Discharge current, A

432 Figure 6 Throughput and lifetime cycles to the failure of a battery (a), capacity curve (b).

433
434 Results for scenario (A)
435 The simulation results at one-minute time step showed the meteorological data and load profile
436 consumption. The summary of system results with the best (most optimised) configuration, including the
437 economic data are presented in Table 2. The capacity factor of the PV array is 4.275 kW (15 modules)
438 with four batteries. The battery wear cost is $0.033/kWh approximately 30% of the COE. The best system
439 configuration has levelised COE of $0.2 kWh and monthly cost of $43.2. These values are extremely low
440 compared with the monthly cost from the grid-based system on the Iraqi Ministry of Electricity prices.
441 The daily power flows of the optimal system for two days sunny (02.04.2018), and part cloudy
442 (05.06.2018) are shown in Figures 7 (a) and (b), and the energy distributions for these days are shown in
443 Figures 7 (d) and (e), respectively. In this system, the battery bank is recharged when the renewable
444 energy fraction higher than desired electrical load and discharges when the desired load exceeds the PV
445 power output and during the night. The daily energy distributions for two different days during the winter

13
446 and summer periods are shown in Table 3. In the sunny day, the PV has hight energy production observed
447 due to higher incident solar radiation.
448
449
450 Tabel 2 Scenario (A) performance operating components - annual analysis

PV Parameters Data Units Battery parameters Data Units

Capacity 4.275 kW Number of batteries 4 -
Output avarage 0.17 kW Strings in parallel 4 -
Avarage output/day 17.0 kWh Nominal capacity 6.72 kWh
Capacity factor 16.6 % Autonomy 22.5 hr

of
Porduction 6,212 kWh/yr Lifetime throughput 16,051 kWh
Max. output 3.55 kW Input energy 1,714 kWh/yr

ro
Operation hours 4,317 h/yr Output energy 1,637 kWh/yr
Cost 0.0435 $/kWh Wear cost 0.033 $/kWh
451
4.0 100
-p 4.0 100
re
Load (a) 02.04.2018 Load
(b)
PV PV 05.06.2018
3.5 3.5 Excess electricity
Excess electricity
Battery input 90 Battery input 90
Production/Consumption, kW

Production/Consumption, kW

3.0 Battery output 3.0 Battery output

lP

Battery SoC Battery SoC

Battery SoC, %
2.5 2.5
Battery SoC

80 80

2.0 2.0
na

70 70
1.5 1.5

1.0 1.0
60 60
0.5 0.5
ur

0.0 50 0.0 50
02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36 02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36

452 Time Time

Jo

4.0 4.0
(c) 02.04.2018 (d) 05.06.2018
PV PV
3.5 Battery output 3.5 Battery output
PV_SC PV_SC
Production/Consumption, kW

Production/Consumption, kW

3.0 3.0

2.5 2.5

2.0 2.0

1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0
02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36 02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36

453 Time Time

454 Figure 7 Scenario (A), the daily power flow (a) and (b), the energy distribution (c) and (d) for sunny day
455 (02.04.2018) and partcloudy day (05.06.2018).

456
457 Table 3 Scenario (A) daily energy distribution for sunny and part cloudy days in kWh.
458

14
 Day Load PV Energy Battery out Battery in
Sunny (02.04.2018) 7.78 23.97 4.48 5.19
Part cloudy (05.06.2018) 8.98 20.29 5.42 4.06
459 The battery SoC presented in Figures 8 (a) and (b). The values of SoC between 60% and 100%
460 approximately 72% of the time, and the SoC values are higher than 80% for more than 85% of the time,
461 as shown in Figure 8 (b). The result shows the SoC monthly battery bank’s profile during the simulation.
462 The SoC during the winter months is lower than during the summer months. Therefore, the system needs
463 sufficient battery capacity to supplement the supplies in winter.
22 100
20 (a) (b)
90
18
80
16
70

of
14
Frequency, %

12 60

SoC, %
10 50

ro
8 40
6 30
4
20
2
0
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
-p 10

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
re
SoC, %
464
465 Figure 8 The frequency battery state of charging (a) and the monthly battery bank SoC (b).
lP

466 The daily power productions with and without the temperature effect for partially cloudy and sunny days
467 at the PV array size (4.28 kW) are presented in Figures 9 (a) and (b), respectively. For the presented days,
468 the PV power decreases because of the temperature effect. The decrease is higher for sunny days than for
na

469 partially cloudy days. The calculated values for partially cloudy days (25.07.2018) with the temperature
470 effect and without the temperature effect are 23.5 and 24.8 kWh, respectively, and the error value caused
ur

471 by the temperature effect is 5.5%. The outcomes for sunny day (20.08.2018) with the temperature effect
472 and without the temperature effect are 23.7 and 25.6 kWh, respectively, and the error value caused by the
473 temperature effect is 6.5%.
Jo

474 The monthly averages of electric energy production at the optimum PV size (4.28 kW) are presented in
475 Figure 9 (c). For the presented systems, two different positioning systems use a fix angle system with an
476 optimum orientation at the selected site (γ = 0 and β = 30) and a two-axis tracking system. The
477 temperature effect on the PV energy production is shown in Figure 9 (c). The maximum energy
478 generations for the modules with the tracking system without the temperature effect and with the
479 temperature effect are 7595.38 and 7,312.63 kWh, respectively, with consideration of approximately 4%
480 energy drop, and its degrees by decreasing the module tilted angle. The ambient temperature effect
481 becomes positive during winter time (cold months, efficiency increases), whereas the negative effect is
482 observed most of the time and decreases the energy production, as shown in Figure 9 (d). The highest
483 energy drop is found during July (–7.39 %) for the tracking system, and the highest energy drop (–7.02%)
484 is found for the fixed module tilted angles (β=30°).

15
 4.5 4.5
With temperature effect 25.07.2018 With temperature effect 20.08.2018
4.0 Without temperature effect 4.0 Without temperature effect

(a) (b)
3.5 3.5

3.0 3.0
Power,kW

Power,kW
2.5 2.5

2.0 2.0

1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0
02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36 02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36

485 Time Time

1200
Fix angle with temperature effect
(c) 2 (d)
Fix angle without temperature effect

of
Track with temperature effect
1000 Trackwithout temperature effect 1

0
800 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

ro
-1
Energy, kWh

Error, %

-2
600

-3

400

200
-p -4

-5
Fix tilted angle
Tracking system
re
-6

0 -7
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

486 -8
lP

Month

487 Figure 9 The temperature effect on the PV power for cloudy and sunny days in (a) and (b) respectively and monthly
488 energy (c), the monthly error percentage due to temperature effect on PV array energy (d).
na

489 Energy distribution

490 The PV power system annual energy distribution illustrated in Figure 10. The PV array is the main energy
ur

491 source for the system which produces approximately 6,212 kWh/yr. However, about 2,485 kWh/yr from
492 the annual energy generated was consumed to charge batteries. The excess energy by about 1,468 kWh/yr
Jo

493 shifted to the dump as split energy [38]. These excess renewable energies are unused because of the lack
494 of demand or storage capacity. Among the total generated energy, approximately 2.52% is used to
495 compensate losses in batteries, and about 3.02% for converter losses.
496

Load consumbtion
(2,485 kWh/yr)
40%

Battery bank charging

(1,914 kWh/yr)
30.81%
PV Energy
output
(6,212 kWh/yr) Dump load (1,468kWh/yr)
(100%) (23.63%)

Battery bank losses

(157 kWh/yr)
(2.52%)

Converter losses
(188kWh/yr)
(3.02%) 16
497
498 Figure 10 Energy flow distribution after one-year of simulation.

499
500
501
502
503
504
505

of
506

ro
507
508
509
-p
re
510
511
lP

512
513
na

514
515
ur

516
Jo

517
518 Results for scenario (B)
519 The on-grid PV system was analysed using a similar procedure as of the off-grid PV system. The PV
520 system components include only PV modules at fixed angles (γ = 0 and β = 30) and a two-axis tracking
521 system, with the temperature effect on the PV energy production connected with the grid. The daily
522 power generation and energy distribution for two different days during the winter (03.01.2018) and
523 summer periods (22.05.2018) are shown in Figures 11 (a and b) and (c and d), respectively. The daily
524 energy distributions for two different days during the winter and summer periods are presented in Table 4.

17
 4.0 4.0
Load
(a) 28.08.2018 Load (b)
3.5 PV 3.5 PV 27.05.2018
From grid From grid
Production/Consumption, kW

Production/Consumption, kW
Fed grid Fed grid
3.0 3.0

2.5 2.5

2.0 2.0

1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0
02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36 02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36

525 Time Time

4.0 4.0
(c) PV
PV 28.08.2018 27.05.2018
From grid (d)

of
3.5 From grid 3.5 Fed grid
Fed grid PV_SC
PV_SC
Production/Consumption, kW

Production/Consumption, kW
3.0 3.0

ro
2.5 2.5

2.0 2.0

1.5

1.0
-p 1.5

1.0
re
0.5 0.5

0.0 0.0
lP

02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36 02:00 04:48 07:36 10:24 13:12 16:00 18:48 21:36

526 Time Time

527 Figure 11 Scenario (B), the daily power flow in (a) and (b) and the energy distribution (c) and (d) for sunny day
na

528 (28.08.2018) part cloudy day (27.05.2018).

529
530 Table 4 Scenario (B) daily energy distribution for sunny and part cloudy days in kWh.
ur

531
Fed to the PV Self-
Day Load PV energy From gird
grid Consumption
Jo

Sunny (28.08.2018) 6.54 25.67 3.51 20.07 3.02

Part cloudy (27.05.2018) 7.55 19.71 4.56 14.74 2.99
532
533 The monthly electrical energy distributions (energy from the grid and fed to the grid) are presented in
534 Figure 12. Two different position systems, namely, (i) with fixed angles (optimum orientation in the
535 selected site γ = 0 and β = 30) and with (ii) a two-axis tracking system, were analysed. The temperature
536 effect on PV energy production was considered. The system summary and selected economic parameters
537 of the on-grid PV power system are presented in Figure (12). The capacity factor of the PV array is 4.275
538 kW (15 modules) and generates 6,212 kWh/yr of electrical energy. The annual energy taken from the grid
539 is 1,630 kWh/yr, and the energy fed to grid is 4,600 kWh/yr.

18
 650
Energy from grid
600 587
Energy fed to grid
542 551
550 536

500
461 468
450
409
400
Energy, kWh

350
303
300 277

250

200 181 186

159 156
169 162 170
150 126 135 133
109 111 116
97
100 88

of
50

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

ro
540 Month

541 Figure 12 The monthly energy from grid and fed to the grid.

542
-p
re
543 8. Systems comparison
544 Cost break-down
lP

545 The cash flow breakdown in terms of cost type and components is shown in Figure 13. The currency used
546 in this study is US dollar ($). For the off-grid PV system, the initial capital, total net present and operation
na

547 costs of the analysed system are $2,720, $6,244 and $274/yr, respectively. The corresponding levelised
548 COE is $0.196/kWh on the basis of the selected household consumption (7.2 kWh/d). The monthly
549 energy cost is $42.34, and this value is two times lower than the tariff provided by the National Grid of
ur

550 the Iraqi Ministry of Electricity [56].

551 For the on-grid PV system, the initial capital, total net present and operation costs are $1,800, $6,132 and
Jo

552 $388/yr, respectively. The corresponding levelised COE is $0.183/kWh on the basis of the selected
553 household consumption (7.2 kWh/d). The monthly energy cost is approximately $39.53, and this value is
554 more than two times lower than the current electrical tariff provided by National Grid of the Iraqi
555 Ministry of Electricity [56].

19
 7000
Salvage
6000 O&M
Replacement
5000 Capital

4000
NPC,$

3000

2000

of
1000

ro
0

-1000
-p
re
PV Battery Converter System PV Grid Converter System
Scenario (A) Scenario (B)
lP

556
557 Figure 13 Off-grid, on-gridPV system cash flow break-down by cost type for the best configuration.
na

558
559 The effect of the PV capacity
ur

560 The effects of PV capacity on the total NPC and COE for the off-grid and on-grid PV systems are shown
561 in Figures 14 (a)–(b). In this case, the converter size is fixed at 5 kW. The results show that the optimal
Jo

562 system size is 4.2 kW PV for both systems. The increase in PV capacity can increase the total NPC of the
563 off-grid system and decrease the NPC of the on-grid system because of the renewable energy sold to the
564 grid. However, the COE increases with the increase in PV capacity in any case.
7500
0.24
Scenario (A)
(a)
Scenario (A) (b)
Scenario (B)
7000 Scenario (B)
0.22

6500 0.20
NPC, $

COE, $

6000 0.18

5500 0.16

5000 0.14

4500 0.12
4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4

565 PV capacity, kW PV capacity, kW

566 Figure 14 The effect of PV capacity on total NPC (a) and COE (b).

20
567
568 The effect of the load consumption
569 The results analysed in this work were based on daily load consumption of approximately 7.2 kWh/d.
570 Load consumptions varying from 7.2 kWh/d to 10 kWh/d were investigated to analyse the impact of large
571 energy consumption on the optimisation results. The configuration for the optimal off-grid PV system
572 under different load consumption levels is presented in Table 5. The PV capacity, converter size and
573 battery number gradually increase with the increase in load consumption.
574 Table 5 Scenario (A) optimal configuration at different load coniguration.

daily load Excess

System PV size Battery Converter
average electricity
No. (kW) number (kW)
(kW h) (%)

of
1 7.18 4.275 4 5 54
2 7.5 4.275 4 5 52.4

ro
3 8 4.845 4 5 54.9
4 8.5 5.130 4 5 55.1
5
6
9
9.5
5.130
5.415
-p 5
5
5
6
52.3
52.4
re
7 10 5.980 5 6 54.5
575
lP

576 Table 5 presents the results based on the electrical load measured in 2018, where the increase in the
577 desired electrical load has high effects on system size, configuration, and cost, as shown in Figure 14.
na

578 Figures 15(a)–(b) demonstrate the average daily load impact on the NPC and COE (b). The increase in
579 daily load increases the NPC and COE of the off-grid PV system, which is larger than of the on-grid PV
580 system. For the daily load average of 8.5 kWh/d, the NPC recorded is approximately $7,253, and the
ur

581 COE is $0.248/kWh. The NPC, COE of off-grid and on-grid PV systems are $6,795 and $0.171/kWh and
582 $0.248 and $0.171/kWh, respectively in terms of excess electricity contribution cost.
Jo

10000 0.30

9500 Scenario (A) (a) Scenario (A) (b)

0.28 Scenario (B)
Scenario (B)
9000
0.26
8500
0.24
8000
NPC, $

0.22
7500
COE, $

0.20
7000
0.18
6500
0.16
6000
0.14
5500

5000 0.12

4500 0.10
7.0 7.5 8.0 8.5 9.0 9.5 10.0 7.0 7.5 8.0 8.5 9.0 9.5 10.0
Daily load average, kWh
583 Daily load average, kWh

584 Figure 15 The effect of average daily load on total NPC (a) and COE (b).

585 Figure 16 (a) shows the estimated annual energy generated by the off-grid and on-grid PV power systems
586 with the increase in daily load. The off-grid system generates energy more than the on-grid system,
587 although the two PV systems have the same energy production. As shown in Figure 15, the off-grid one
588 generates more energy than the on-grid system because of the energy generated from storage units

21
589 (batteries). Figure 16 (b) shows the difference between the off-grid and on-grid PV systems in terms of
590 NPC and COE. The COE difference increases when the PV system size increases and this difference is
591 because of the storage units (batteries) used in the off-grid PV system. The differences in NPC increases
592 when the PV system size increases and this difference is because of the energy fed to the grid from the
593 on-grid PV system.
594
9000 36
Scenario (A) (a) NPC (b)
Scenario (B) 32
8000
PV system energy production, kWh/yr

COE
28
7000
24

6000 20

%
16

of
5000

12
4000

ro
8

3000
4

595
2000
7.0 7.5 8.0 8.5 9.0
Daily load average, kWh
9.5 10.0
-p 0
7.0 7.5 8.0 8.5 9.0

Daily load average, kWh

9.5 10.0
re
596 Figure 16 The systems annual expected energy generated (a), systems difference percentage between NPC and
597 COE in (b).
lP

598
599 CONCLUSIONS
na

600 This study confirmed that PV energy systems can replace/compensate the desired load using renewable
601 energy power. The solar energy systems without or with battery can supply power to typical households
ur

602 in Iraq. A PV power system that meets the desired load of a typical household in Diyala, Iraq was
603 developed through modelling, simulation and optimisation. The PV power systems include (i) off-grid
604 (PV–battery–inverter) and (ii) on-grid (PV–inverter–grid) systems. The input data of electrical loads,
Jo

605 solar radiation, ambient temperature and wind speed in Baqubah City, which is the capital of Diyala
606 Government, were used to achieve economic optimisation using a genetic algorithm. Simulations and
607 optimisations were conducted to determine the cost and performance of the proposed PV power systems.
608 The results showed that the two PV power systems meet the annual desired electrical load of the
609 household. The proposed PV on-grid power system provides excess electricity to the grid requires
610 cheaper energy cost than the off-grid power system and is suitable to supply energy to the grid.
611 - For the power system consist (PV= 4.275kW PV, battery= 2.4 kW) at off-grid (scenario A), the
612 expected total NPC is $6,244, and the COE is $0.196/kWh. The increase in system size leads to an
613 increase in NPC and COE. For the on-grid (scenario B), the expected total NPC is $6,115, and the
614 COE is $0.183/kWh. The increase in system size decreases the NPC for investment payback because
615 of the energy sent to the grid and increases the COE (Figures 14 (a) and (b)). For the long investment
616 period, the off-grid PV system has higher NPC than the on-grid PV system.
617
618 - Despite the long hours of the day in Iraq during summer that leads to more solar energy production
619 than winter days, the high temperature in summer has the high reduction in energy production because
620 of the high temperature of solar cells, which reaches more than 70 °C, as shown in Figure 9.
621

22
622 - Iraq uses natural gas in its electrical power stations, natural gas is the second polluter after coal, and
623 the natural gas available in Iraq generates 0.35 kgCO2/kWh [7]. The use of renewable energy from PV
624 power systems with seven sizes shown in Table 5 can avoid CO2 emissions, as shown in Figure 17.
625 For the off-grid system, the emissions caused by the replacement of batteries are four times during the
626 project lifetime, and the on-grid system avoids emissions by feeding the energy to the grid.
2200
Scenario (A)
2000 Scenario (B)
1800
Annual avoidance of CO , Kg

1600
2

1400

1200

of
1000

800

ro
600

400

200
-p
re
0
1 2 3 4 5 6 7

PV system number
lP

627
628 Figure 17 Avoidance of CO2 emissions for seven sizes of PV system.
na

629
630 The obtained results encourage Iraqi people to use PV power systems for electrifying their houses and
ur

631 feeding the national grid.

632 Acknowledgements
Jo

633 The work was supported by University of Diyala grant no. 12.12.43.2019
634
635 References
636 1. Al-Riahi M, Al-Kayssi A. Some comments on time variation in solar radiation over Baghdad, Iraq.
637 Renewable energy 1998;41:479-484. https://doi.org/10.1016/S0960-1481(98)00107-4
638
639 2. Kazem A, Miqdam T. Status and future prospects of renewable energy in Iraq. Renewable and
640 Sustainable Energy Reviews 2012: 16: 6007-6012. https://doi.org/10.1016/j.rser.2012.03.058
641
642 3. Dihrab S, Kamaruzzaman Sopian. Electricity generation of hybrid PV/wind systems in Iraq.
643 Renewable Energy 2010; 35: 1303-1307. https://doi.org/10.1016/j.renene.2009.12.010
644
645 4. Al-Riahi M, Al-Hamdani N, Al-Saffar H. Some aspects of solar radiation climatology of Iraq.
646 Renewable energy 1992: 2 : 167-173. https://doi.org/10.1016/0960-1481(92)90102-9
647

23
648 5. Van der Zwaan B, Lachlan C, Tom. Potential for renewable energy jobs in the Middle East. Energy
649 Policy 2013:60 : 296-304. https://doi.org/10.1016/j.enpol.2013.05.014
650
651 6. Almusaed A, Asaad A. Building materials in eco-energy houses from Iraq and Iran. Case Studies
652 in Construction Materials 2015:2: 42-54. https://doi.org/10.1016/j.cscm.2015.02.001
653
654 7. Hassan Q, Jaszczur M, Mohamed M, Styszko K, Szramowiat K, Gołaś J. Off-grid photovoltaic
655 systems as a solution for the ambient pollution avoidance and Iraq’s rural areas electrification. In
656 E3S Web of Conferences 2016:10: 00093. https://doi.org/10.1051/e3sconf/20161000093
657
658 8. Abdulateef J. Simulation of solar off-grid photovoltaic system for residential unit. International
659 Journal of Sustainable and Green Energy 2014; 4: 29-33. https://doi:
660 10.11648/j.ijrse.s.2015040301.15

of
661
662 9. Al-Shamani N, Othman H, Mat S, Ruslan H, Abed M, Sopian K. Design & sizing of stand-alone

ro
663 solar power systems a house Iraq. Recent Advances in Renewable Energy Sources 2015:145-150.
664
665
666
10. -p
Elbreki A, Sopian K, Ruslan H, Al-Shamani N, Elhub B. A stand-alone Photovoltaic system
design and sizing: A greenhouse application in Sabha city as a case study in Libya. Proceeding of
667
re
The 3rd Engineering Science And Technology 2017:3.
668
669 11. Alamsyah I, Kamaruzzaman S, Shahrir A. Technoeconomics analysis of a photovoltaic system to
lP

670 provide electricity for a household in Malaysia. Proceedings of the International Symposium on
671 Renewable Energy: Environment Protection and Energy Solution 2006: 762.
672
na

673 12. Palej P, Hassan Q, Kleszcz S, Hanus R, Jaszczur M. Analysis and optimization of hybrid renewable
674 energy systems. Polityka Energetyczna 2019: https://doi: 22. 10.33223/epj/109911
ur

675
676 13. Ceran B, Hassan Q, Jaszczur M, Sroka K. An analysis of hybrid power generation systems for a
677 residential load. In E3S Web of Conferences 2017;14: 01020.
Jo

678 https://doi.org/10.1051/e3sconf/20171401020
679
680 14. Styszko K, Jaszczur M, Teneta J, Hassan Q, Burzyńska P, Marcinek E, Samek L. An analysis of
681 the dust deposition on solar photovoltaic modules. Environmental Science and Pollution Research
682 2019; 26: 8393-8401. https://doi.org/10.1007/s11356-018-1847-z
683
684 15. Jaszczur M, Teneta J, Styszko K, Hassan Q, Burzyńska P, Marcinek E, Łopian N. The field
685 experiments and model of the natural dust deposition effects on photovoltaic module efficiency.
686 Environmental Science and Pollution Research 2019; 26: 8402-8417.
687 https://doi.org/10.1007/s11356-018-1970-x
688
689 16. Onat N. Comparison of Turkey’s Geographical Regions in terms of Stand-Alone PV System
690 Design and Cost Parameters. International Journal of Photoenergy 2017
691 https://doi.org/10.1155/2017/6392635
692

24
693 17. Ghafoor A, Munir A. Design and economics analysis of an off-grid PV system for household
694 electrification. Renewable and Sustainable Energy Reviews 2015; 42: 496-502.
695 https://doi.org/10.1016/j.rser.2014.10.012
696
697 18. Nordin D, Rahman A. A novel optimization method for designing stand alone photovoltaic system.
698 Renewable Energy 2016; 89: 706-715.
699
700 19. Olcan C. Multi-objective analytical model for optimal sizing of stand-alone photovoltaic water
701 pumping systems. Energy conversion and management 2015; 100: 358-369.
702 https://doi.org/10.1016/j.enconman.2015.05.018
703
704 20. Riza L, Gilani H, Aris S. Standalone photovoltaic systems sizing optimization using design space
705 approach: case study for residential lighting load. Journal of Engineering Science and Technology,

of
706 2015;10: 943-957.
707

ro
708 21. Al-Falahi D, Jayasinghe G, Enshaei H. A review on recent size optimization methodologies for
709 standalone solar and wind hybrid renewable energy system. Energy Conversion and Management
710
711
-p
2017; 143: 252-274. https://doi.org/10.1016/j.enconman.2017.04.019

712
re
22. Hassan Q, Jaszczur M, Przenzak E. Mathematical model for the power generation from arbitrarily
713 oriented photovoltaic panel. In E3S Web of Conferences 2017; 14: 01028.
714 https://doi.org/10.1051/e3sconf/20171401028
lP

715
716 23. Zhang W, Maleki A, Rosen A, Liu J. Sizing a stand-alone solar-wind-hydrogen energy system
717 using weather forecasting and a hybrid search optimization algorithm. Energy conversion and
na

718 management 2019; 180: 609-621. https://doi.org/10.1016/j.enconman.2018.08.102

719
ur

720 24. Ma T, Yang H, Lu L. A feasibility study of a stand-alone hybrid solar–wind–battery system for a
721 remote island. Applied Energy 2014 ;121: 149-158. https://doi.org/10.1016/j.apenergy.2014.01.090
722
Jo

723 25. Jaszczur M, Hassan Q, Styszko K, Teneta J. Impact of dust and temperature on energy conversion
724 process in photovoltaic module. Thermal Science 2019; 23:1199-1210.
725 https://doi.org/10.2298/TSCI19S4199J
726
727 26. Mokheimer M, Sahin Z, Al-Sharafi A, Ali I. Modeling and optimization of hybrid wind–solar-
728 powered reverse osmosis water desalination system in Saudi Arabia. Energy Conversion and
729 Management 2013; 75: 86-97. https://doi.org/10.1016/j.enconman.2013.06.002
730
731 27. Bourouni K, M’Barek B, Al Taee A. Design and optimization of desalination reverse osmosis
732 plants driven by renewable energies using genetic algorithms. Renewable energy 2011; 36: 936-
733 950. https://doi.org/10.1016/j.renene.2010.08.039
734
735 28. Cherif H, Belhadj J. Large-scale time evaluation for energy estimation of stand-alone hybrid
736 photovoltaic–wind system feeding a reverse osmosis desalination unit. Energy 2011; 36: 6058-
737 6067. https://doi.org/10.1016/j.energy.2011.08.010
738

25
739 29. Koutroulis E, Blaabjerg F. A new technique for tracking the global maximum power point of PV
740 arrays operating under partial-shading conditions. IEEE Journal of Photovoltaics 2012; 2: 184-190.
741 https://doi.org/10.1109/JPHOTOV.2012.2183578
742
743 30. Tina G, Gagliano S, Raiti S. Hybrid solar/wind power system probabilistic modelling for long-term
744 performance assessment. Solar energy 2006; 80: 578-588.
745 https://doi.org/10.1016/j.solener.2005.03.013
746
747 31. Huang Z, Xie Z, Zhang C, Chan H, Milewski J, Xie Y, Hu X. Modeling and multi-objective
748 optimization of a stand-alone PV-hydrogen-retired EV battery hybrid energy system. Energy
749 conversion and management 2019; 181: 80-92. https://doi.org/10.1016/j.enconman.2018.11.079
750
751 32. Pudjianto D, Djapic P, Dragovic J, Strbac G. Grid integration cost of photovoltaic power

of
752 generation. Energy Futures Lab, Imperial College: London, UK 2013.
753

ro
754 33. Keppler H, Cometto M. Nuclear energy and renewables: system effects in low-carbon electricity
755 systems. Nuclear Energy Agency 2012, OECD: Paris, France.
756
757 34.
-p
Ueckerdt F, Brecha R, Luderer G. Analyzing major challenges of wind and solar variability in
758
re
power systems. Renewable energy 2015; 81: 1-10. https://doi.org/10.1016/j.renene.2015.03.002
759
760 35. Rai N, Rai B. Control of fuzzy logic based PV-battery hybrid system for stand-alone DC
lP

761 applications. Journal of Electrical Systems and Information Technology 2018; 5: 135-143.
762 https://doi.org/10.1016/j.jesit.2018.02.007
763
na

764 36. Kaushika D, Gautam K, Kaushik K. Simulation model for sizing of stand-alone solar PV system
765 with interconnected array. Solar Energy Materials and Solar Cells 2005; 85: 499-519.4.
ur

766 https://doi.org/10.1016/j.solmat.2004.05.024
767
768 37. Fara L, Craciunescu D. Output analysis of stand-alone PV systems: modeling, simulation and
Jo

769 control. Energy Procedia 2017; 112: 595-605.

770
771 38. Rezk H, Abdelkareem A, Ghenai C. Performance evaluation and optimal design of stand-alone
772 solar PV-battery system for irrigation in isolated regions: A case study in Al-Minya (Egypt).
773 Sustainable Energy Technologies and Assessments 2019; 36: 100556.
774 https://doi.org/10.1016/j.seta.2019.100556
775
776 39. Rezk H, El-Sayed M. Sizing of a stand alone concentrated photovoltaic system in Egyptian site.
777 International Journal of Electrical Power & Energy Systems 2013; 45: 325-330.
778 https://doi.org/10.1016/j.ijepes.2012.09.001
779
780 40. Ghenai C, Merabet A, Salameh T, Pigem C. Grid-tied and stand-alone hybrid solar power system
781 for desalination plant. Desalination 2018; 435: 172-180.
782 https://doi.org/10.1016/j.desal.2017.10.044
783
784 41. Luque A, Hegedus S. Handbook of photovoltaic science and engineering. John Wiley & Sons
785 2011.

26
786
787 42. Santiago I, Trillo-Montero D, Moreno-Garcia M, Pallarés-López V, Luna-Rodríguez J. Modeling
788 of photovoltaic cell temperature losses: A review and a practice case in South Spain. Renewable
789 and Sustainable Energy Reviews 2018; 90: 70-89. https://doi.org/10.1016/j.rser.2018.03.054
790
791 43. Perović B, Klimenta D, Jevtić M, Milovanović M. An analytical model for estimating the
792 temperature of a photovoltaic module based on the principle of energy balance. Zbornik
793 Međunarodne konferencije o obnovljivim izvorima električne energije–MKOIEE 2017; 5: 89-95.
794
795 44. Jaszczur M, Teneta J, Hassan Q, Majewska E, Hanus R. An Experimental and Numerical
796 Investigation of Photovoltaic Module Temperature under Varying Environmental Conditions. Heat
797 Transfer Engineering, 2019, 1-14. https://doi.org/10.1080/01457632.2019.1699306
798

of
799 45. Nordmann T, Clavadetscher L. Understanding temperature effects on PV system performance.
800 In 3rd World Conference on photovoltaic Energy Conversion. Proceedings 2003; 3: 2243-2246.

ro
801
802 46. Mahmoud M, Ibrik H. Techno-economic feasibility of energy supply of remote villages in Palestine
803
804
-p
by PV-systems, diesel generators and electric grid. Renewable and sustainable energy reviews
2006; 10: 128-138. https://doi.org/10.1016/j.rser.2004.09.001
805
re
806 47. LUMINOUS battered available on 10 March 2020 at the website https://www.luminousindia.com,
807 2020 (accessed 15 January 2020)
lP

808
809 48. Jaszczur M, Hassan Q, Teneta J. (2018). Temporal load resolution impact on PV/grid system
810 energy flows. In MATEC Web of Conferences 2018; 240: 04003.
na

811 https://doi.org/10.1051/matecconf/201824004003
812
ur

813 49. Chan L. A new battery model for use with battery energy storage systems and electric vehicles
814 power systems. IEEE Power Engineering Society Winter Meeting. Conference Proceedings 2000;1:
815 IEEE. https://doi.org/10.1109/PESW.2000.850009
Jo

816
817 50. Jaszczur M, Hassan Q, Al-Anbagi N, Palej P. A numerical analysis of a HYBRID PV+ WT power
818 system. In E3S Web of Conferences 2019; 128: 05001.
819 https://doi.org/10.1051/e3sconf/201912805001
820
821 51. ROHS available on 10 March 2020 at the website https://www.rohsguide.com/rohs-faq.htm., 2020
822 (accessed 15 January 2020)
823
824 52. Jaszczur M, Hassan Q, Palej P. An optimisation of the hybrid renewable energy systems. In E3S
825 Web of Conferences 2019; 113: 03022. https://doi.org/10.1051/e3sconf/201911303022
826
827 53. Hassan Q, Jaszczur M, Abdulateef J. Optimization of PV/wind/diesel hybrid power system in
828 homer for rural electrification. In Journal of Physics: Conference Series 2016; 745: 032006.
829 https://doi.org/10.1088/1742-6596/745/3/032006
830

27
831 54. Jaszczur M, Hassan Q, Teneta, J. Temporal load resolution impact on PV/grid system energy flows.
832 In MATEC Web of Conferences 2018; 240: 04003.
833 https://doi.org/10.1051/matecconf/201824004003
834
835 55. Jaszczur M, Hassan Q, Palej P, Abdulateef J. Multi-objective optimisation of micro grid hybrid
836 power system for Household application. Energy 2020; 117738.
837 https://doi.org/10.1016/j.energy.2020.117738
838
839 56. Iraq Energy Institute report 2018 avalible in the website https://iraqenergy.org/wp-
840 content/uploads/2019/02/Iraq-Energy-Institute-2018-Annual-Report.pdf, 2020 (accessed 15
841 January 2020)
842
843 57. Elliston B, Diesendorf M, MacGill I. Simulations of scenarios with 100% renewable electricity in

of
844 the Australian National Electricity Market. Energy Policy 2012; 45: 606-613.
845 https://doi.org/10.1016/j.enpol.2012.03.011

ro
846
847 58. Bhende N, Panda S, Mishra S, Narayanan A, Kaipia T, Partanen J. Optimal Power Flow
848
849
-p
Management and Control of Grid Connected Photovoltaic-Battery System. International Journal of
Emerging Electric Power Systems 2019; 20. https://doi.org/10.1515/ijeeps-2019-0056
850
re
851 59. Riffonneau Y, Bacha S, Barruel F, Baghzouz Y, Zamaï E. Optimal reactive supervision of grid
852 connected PV systems with batteries in real conditions. International Review of Electrical
lP

853 Engineering 2012; 7; 4607-4615

854
855 60. Riffonneau Y, Bacha S, Barruel F, Ploix S. Optimal power flow management for grid connected
na

856 PV systems with batteries. IEEE Transactions on sustainable energy 2011; 2; 309-320.
857 https://doi.org/ 10.1109/TSTE.2011.2114901
ur
Jo

28
Highlights

• Photovoltaic power system for household application.

• Simulation-based economic optimisation approach.
• Local weather resources highly influence on the optimal design.
• Established optimal Off/On grid photovoltaic systems configurations.
• Promising solution to decrease CO2 emissions.

of
ro
-p
re
lP
na
ur
Jo
Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.

☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests:

of
ro
-p
re
lP
na
ur
Jo

Qusay Hassan
10.03.2020