2018 IEEE PES/IAS PowerAfrica
Characteristics of Solar Power Generation Plant
Z.P. Ncane
School of Engineering, Howard College Campus
University of KwaZulu-Natal
King George V Ave, Glenwood, 4041, Durban, South
Africa
zehsizwe@gmail.com
Abstract—A solar photovoltaic plant model using MATLAB
Simulink has been developed and presented in this paper. One
solar module is first simulated, followed by a string of series
connected modules and an array of parallel connected strings.
These were initially simulated under standard testing conditions
(STC) i.e. cell temperature of 25 ºC and irradiance level of
1000 W/m2. Furthermore, the plant was subjected to different
ambient conditions, aimed at studying its I-V and P-V
characteristics. Solar irradiance and temperature were varied
and the dependency of the plant to each parameter is observed.
The plant is then simulated using real data for Upington, South
Africa. The paper further highlights future work which will be a
continuation of the work presented in this paper.
Index Terms—Maximum power point, Open circuit voltage, PV
array, PV string, Short circuit current.
I. INTRODUCTION
Renewable power generators such as solar power
generators play a vital role in supporting the power system,
especially under system emergencies. These types of
generators are highly dependent on weather patterns. To
minimize the variation in the plant’s output power brought
about by the varying weather patterns; other applications make
use of storage methods to store power for use under
unfavorable weather conditions. Solar power generators are
environmentally friendly and have an advantage of having
reduced carbon emissions as there is no fuel source required to
burn for power production.
For the purpose of this study, Upington area, located in
Northern Cape, South Africa is chosen to be the location of
the solar power plant under study. The province offers
favorable solar radiation levels and has attracted the majority
of solar PV projects in South Africa [1].
II. PHOTOVOLTAIC SYSTEM
The system under study is a 30 MW PV plant, aimed at
being grid connected in order to supplement conventional
power generators in the event of network constraints. PV
panels used for the study are monocrystalline silicon type
which is fairly more efficient compared to polycrystalline and
978-1-5386-4163-7/18/$31.00 ©2018 IEEE
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE ENGENHARIA DE ITAJUBA. Downloaded on April 27,2021 at 20:46:29 UTC from IEEE Xplore. Restrictions apply.
A.K. Saha
School of Engineering, Howard College Campus
University of KwaZulu-Natal
King George V Ave, Glenwood, 4041, Durban, South
Africa
Saha@ukzn.ac.za
thin-film modules [2]. The data for a monocrystalline silicon
module is therefore used for simulation purposes.
A. Generic PV cell
For a successful modelling and simulation of the plant,
expressions representing PV cell configuration and behavior
need to be obtained. An equivalent circuit of a practical PV
cell is therefore shown in Fig. 1, which have components of
shunt and series resistances; and a bypass diode. The
expressions defining the I-V characteristics of the solar PV
cell are derived from this equivalent circuit and are given by
(1) to (10).
Figure 1: PV cell equivalent circuit with Rs and Rsh
From Fig.1 the output current of a PV cell is:
I = I LG − I D − I R sh
(1)
The bypass diode current characteristic and the current
across the shunt resistance component are then substituted in
(1) to yield (2).
  q
I = I LG − I OS exp (V + RS I ) − 1 − V + Rs I
  AkT   Rsh (2)
Due to the complexity of obtaining the parameters related
to the cell’s temperature and radiation intensity, most
parameters are available in a PV module’s datasheets. The
following equations define other unknown parameters, which
are further defined in Table I. For an ideal PV cell where the
circuit excludes the series and shunt resistances; when the PV
cell is short circuited ISC=ILG. The assumption ISC≈ILG is
generally used in the modelling of PV arrays for a practical
PV cell due to the introduction of series and parallel resistance
as per (2). Equation (3) shows the bypass diode saturation
current which strongly depends on cell temperature. The
250
 2018 IEEE PES/IAS PowerAfrica

bypass diode is used across each cell to avoid hot-spots in TABLE I: SYMBOLS USED IN THE EQUATIONS DEFINING THE CELL’S I-
solar PV modules [3]. Photon current and PV cell output V, P-V CHARACTERISTICS [3]
current are given in (4) and (5), respectively.
3
Symbol Description
T   qE  1 1  V PV Cell output voltage (V)
I OS = I or   exp  GO  −  
T
 r  Bk  Tr T   I PV Cell output current (A)
(3)
S ILG Photon current (A)
I LG = [I SCR + K I (T − 298)] IOS Reverse saturation current of PV cell (A)
1000 (4)
T Operating temperature of PV cell (°K)
   V
  
I = I SC 1 − C1 exp
 − 1  q Charge of an Electron =1.6*10-19 C
   C 2VOC
   (5) K Boltzmann’s constant =1.38*10-23 J/°K
At the maximum power point, when V = Vm, I = Im, (5) is Kl Temperature coefficient
therefore expressed as:
ISC Short circuit current (A/°K)
   V    S Operating solar radiation (W/m2 )
I m = I SC 1 − C1 exp m  − 1  Short circuit current at STC (A)
   C2VOC    (6) EGO Band gap energy of the semiconductor (J)
Under normal temperature (STC), A, B Ideality constant between 1 and 2
Tr Absolute temperature at Standard Test Condition of PV
 V  cell = 301.18°K
exp m  >> 1 IOR Reverse saturation current of PV cell at temperature Tr
 C 2VOC 
(A)
RSh Intrinsic parallel resistance of PV cell (Ω)
Therefore,
RS Intrinsic series resistance of PV cell (Ω)

   V   B. PV cell characteristics
I m = I SC 1 − C1 exp m   (7)
   C 2VOC  
For a 30 MW solar power plant under study, the employed
PV module is a Sunmodule make and its data as extracted
from the datasheet is given in Table II [4]. Some parameters
 I   − Vm  (8) listed in Table II depict the performance of the module and are
C1 = 1 − m  exp 
 I SC   C 2VOC  not necessarily used in the PV model e.g. module efficiency,
operating temperature etc. A single module was first
Substituting (8) to (6) yields (9). simulated, followed by a series connected modules called
strings and parallel connected strings called arrays. Only 1 out
of 19 PV arrays is modelled in this paper, which consists of 40
  I   − Vm   1    (9) series connected cells and 161 parallel strings.
0 = I SC 1 − 1 − m  exp   exp   − 1 
  I SC   C 2VOC   C 2   
TABLE II: 30 MW SOLAR POWER PLANT PARAMETERS [4]
Under normal temperature (STC), Parameter Specification
Plant Rating (assumed for research purposes)
PV panels total output power 30 MW
 1  System voltage 1500 V DC
exp  >> 1 PV Module (datasheet information) [4]
 C2  Cell type Monocrystalline silicon
Open circuit voltage 37.8 V
Short Circuit current 8.28 A
 Vm  Maximum power point voltage 31.1 V
 − 1 (10)
 VOC  Maximum power point current 8.05 A
C2 =
 Im  Module efficiency 15.7%
ln1 − 
 I SC  Maximum power rating 250 W
TC Isc (A) 0.004 %/K
The aforementioned expressions show relationships of TC Voc (B) -0.30 %/K
different parameters that make it possible to obtain the output TC Pmpp -0.45%/K
power from the PV cell. These expressions are therefore used Operating temperature -40 ºC to 85 ºC
to model and simulate the solar PV plant under study.

978-1-5386-4163-7/18/$31.00 ©2018 IEEE 251

Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE ENGENHARIA DE ITAJUBA. Downloaded on April 27,2021 at 20:46:29 UTC from IEEE Xplore. Restrictions apply.
 2018 IEEE PES/IAS PowerAfrica

Fig. 2 and 3 show simulation results for a single PV increases the voltage to 1500 V DC. Fig. 4 was obtained with
module employed in a 30 MW solar power plant. The results 40 modules connected in series.
demonstrate the characteristics of the module, which indicate
short circuit current, open circuit voltage and maximum output 10

power. A short circuit current is obtained when the voltage is 9

0 V i.e. when module terminals are short circuited; and the 8

open circuit voltage is obtained from the module terminals at
no load. The useful current and voltage (i.e. when there is load 7

connected) yield maximum power that can be produced by the

Short Circuit Current (A)

6

module. 5

4
10

9 3

8
2

7
S hort Circ uit Current (A )

1
6
0
5 0 200 400 600 800 1000 1200 1400 1600
Opern Circuit Voltage (V)
4

3 Figure 4: I-V characteristics of a single string rated at 1500 V DC

2
10000
1
9000
0
0 5 10 15 20 25 30 35 40
8000
Open Circuit Voltage (V)

Figure 2: I-V characteristic of a single PV module used in a 30 MW plant 7000

under study 6000

Power (W)

250 5000

4000

200
3000

2000
150
P ower (W )

1000

100 0
0 200 400 600 800 1000 1200 1400 1600
Voltage (V)

50 Figure 5: P-V characteristics of a single string rated at 1500 V DC

0
0 5 10 15 20
Voltage (V)
Figure 3: P-V characteristic of a single PV module used in a 30 MW plant
under study
The type of module used in the study has an advantage of
producing higher maximum power compared to that employed
in [3], which means fewer modules can be used to obtain the
same required power. The string and array were then modelled
and simulation results are outlined in C and D respectively.
C. PV string characteristics
A simulation of series connected cells to yield higher voltage
was conducted. The assumption made is that all modules are
identical, which in practice is not the case. Certain parameters
may vary due to the manufacturing process of each individual
module. Furthermore, different operating conditions may exist
in different parts of the entire plant e.g. cleanness, shading
effect on some modules etc. [5]. The string and array models
assume that the modules used are identical. To obtain an
increased desired output voltage of the plant, 40 PV modules
were connected in series. This configuration keeps the system
current the same as that of a single PV module and only
25 30 35 40
Short circuit current can be observed to have remained
unchanged with the series connection of modules, while the
open circuit voltage increased to 1500 V DC. The maximum
output power shown in Fig. 5 is 10 kW as per the expected
string power; with maximum power point voltage of
1244 V DC and maximum power point current of 8.05 A. The
obtained results prove that with the series connection of PV
modules, power will increase because of an increased system
voltage. However; to get an even higher increase in power,
both current and voltage must increase. The extent to which
the system voltage was increased to, took into consideration
the inverter input voltage which was guided by the
requirements of the IEC standards. The IEC standards specify
a maximum DC voltage of 1500 V for low voltage systems
[6].
D. PV array characteristics
For increased system output current, the strings were then
connected in parallel, which keeps the voltage at 1500 V DC
and increases the short circuit current to 1333 A. The model
was further developed to include 40 series connected cells

978-1-5386-4163-7/18/$31.00 ©2018 IEEE 252

Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE ENGENHARIA DE ITAJUBA. Downloaded on April 27,2021 at 20:46:29 UTC from IEEE Xplore. Restrictions apply.
 2018 IEEE PES/IAS PowerAfrica

(string) and 161 parallel connected strings; the obtained results fs

are shown in Fig. 6 & 7. mf =
f1 (11)
1400
Where: mf is the frequency modulation ratio.
1200
The modulation ratio (ma) was chosen to be an odd
number in multiples of 3 to eliminate even harmonics. The
1000
carrier frequency was chosen such that it eliminates
interference with audible range. Equation (12) shows a
Short Circuit Current (A)

800 relationship between the amplitude of 3 sinusoidal voltage

waveforms displaced 120 º apart and that of a triangle [7].
Vcontrol
600

ma =
400 Vtri (12)
200
Where: Vcontrol is sinusoidal waveform amplitude.
0
Vtri is the amplitude of the triangle.
0 200 400 600 800 1000 1200 1400 1600
Open Circuit Voltage (V)
The modulation ratio has an impact on the rms voltage
Figure 6: I-V characteristics of an array rated at1333 A and 1500 V DC output of the inverter and is chosen to be below 1 for an
x 10
5 inverter to operate in a linear region. The relationship of the
16 modulation ratio to the line-to-line rms voltage is given by
(13) [7].
14

3
12
VLLrms = maVd
10
2 2 (13)
Power (W)

8
Where: Vd is the input DC voltage.
The inverter reduces the overall rating of the plant due to
6

4 the losses experienced during switching and the modulation

ratio as depicted by (13). This will be considered during the
2
reliability evaluation study that will be performed.
0
0 200 400 600 800
Voltage (V)
1000 1200 1400 1600 A. Simulation results
The triangular waveform used in the SPWM has amplitude of
Figure 7: P-V characteristics of an array rated at 1333 A and 1500 V DC
1 at 1050 Hz switching frequency. Sinusoidal waveforms have
amplitudes of 0.8 at a fundamental frequency of 50 Hz to
With both current and voltage increased, the maximum power achieve the modulation ratio of 0.8 as per (12) and a frequency
that can be produced by one array is 1.61 MW as depicted in modulation ratio of 21 as per (1). Fig. 8 depicts this
Fig. 7. To obtain this maximum power, the maximum current comparison.
of a single array is 1296 A and maximum voltage is
1244 V DC. For a complete plant model, 19 PV arrays are 2
Vtri
Va-contr

used to obtain 30.6 MW output power. The plant modelled in 1.5

Vb-contr
Vc-contr

this paper will then be used to evaluate its reliability, when 1

subjected to different conditions. Plant reliability evaluation

will first consider the PV arrays, followed by other systems 0.5

used for grid connection. Reliability evaluation will be looked 0

at as part of the future planned work. -0.5

III. DC-AC INVERTER -1

Inverters play a vital role to grid connected solar PV power -1.5

0 0.002 0.004 0.006 0.008 0.01
time (s)
0.012 0.014 0.016 0.018 0.02

plants, as the conversion from DC-AC power is required.

Sinusoidal Pulse Width Modulation (SPWM) is used for the Figure 8: SPWM triangular and sinusoidal waveforms comparison
inverter model simulated using MATLAB Simulink. The
carrier frequency (fs) and fundamental frequency (f1) used in The comparison results of Fig. 8 are pulses that are used as
the SPWM are related as per (11) [7]. inputs to the gates of MOSFETs semiconductor switches
employed in the inverter circuit. With the solar PV plant

978-1-5386-4163-7/18/$31.00 ©2018 IEEE 253

Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE ENGENHARIA DE ITAJUBA. Downloaded on April 27,2021 at 20:46:29 UTC from IEEE Xplore. Restrictions apply.
 2018 IEEE PES/IAS PowerAfrica

output voltage of 1500 V DC, which is the input to the

inverter; the inverter unfiltered line-to-line output voltage is
also 1500 V AC as expected. Fig. 9 depicts the inverter line-
to-line (Vbc) output voltage obtained from the inverter.

1500

1000

500
Vbc (V AC)

-500

-1000

-1500

0 0.002 0.004 0.006 0.008 0.01

time (s)
0.012 0.014 0.016 0.018 0.02 Figure 10: ICT algorithm used in MPPT

Figure 9: Vbc modified sinusoidal wave V. DISCUSSION OF RESULTS

A. Temperature Dependence
IV. MAXIMUM POWER POINT TRACKING In an array model, temperature was varied between 15-
55 ºC in steps of 10 ºC, while keeping irradiance constant at
Tracking the maximum power point (MPP) in the solar power 1000 W/m2. This was aimed at studying the behavior of solar
PV plant under varying temperature conditions. Fig. 11
plant is aimed at operating the plant at its maximum efficiency
depicts the obtained P-V characteristics.
at any environmental condition. This is achieved by using a
MPP tracking control algorithm with the power conditioning x 10
6

2
unit between PV modules and a load. With different
algorithms available for a MPPT, the chosen method for use in 1.8
15 degC

this paper is the incremental conductance technique (ICT) as 1.6

25 degC
per the algorithm in Fig. 10. The ICT offers high accuracy in 1.4

tracking the MPP and its implementation complexity is fairly

35 degC

medium. The technique compares the incremental

1.2 45 degC
Power (W)

conductance ΔI to the instantaneous conductance I of the 1 55 degC

( ) ( )
ΔV V 0.8

plant. The interpretation is that the incremental conductance is

0.6
the same as the instantaneous conductance at the maximum
power point, greater at the left of the MPP and smaller on the 0.4

right side of the MPP [8]. The results will therefore be used to 0.2

control the inverter switching. 0

0 200 400 600 800 1000 1200 1400 1600 1800
Voltage (V)

Figure 11: P-V characteristics at varying temperature

When the ambient temperature is varied, it can be

observed that as the temperature rises, the open circuit voltage
is reduced thereby reducing the output power. This is due to
the temperature coefficients of the modules that impacts on
the modules output power when the temperature is varied.

B. Irradiance Dependence

Varying irradiance while keeping the temperature constant

at 25 ºC was observed and results are depicted in Fig. 12. The
results prove that the solar PV plant is highly dependent on the
irradiance. The plant produces less power at low irradiance
levels and it increases as irradiance levels are raised.

978-1-5386-4163-7/18/$31.00 ©2018 IEEE 254

Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE ENGENHARIA DE ITAJUBA. Downloaded on April 27,2021 at 20:46:29 UTC from IEEE Xplore. Restrictions apply.
 2018 IEEE PES/IAS PowerAfrica

6 TABLE III: OUTPUT POWER OBTAINED FOR SOLAR PV PLANT IN UPINGTON

x 10
2 AREA

1.8
Month Daily Irradiance Power (MW)
1.6 Average (W/m2)
1.4
1000 W/m2
January 627.425 0.996
1.2
February 528.415 0.831
750 W/m2
March 494.535
Power (W)

1
0.775
April 442.373 0.688
0.8
500 W/m2
May 354.587 0.544
0.6
June 304.386 0.463
0.4 250 W/m2 July 323.425 0.494
0.2 August 389.000 0.600
0 September 458.359 0.714
0 200 400 600 800 1000 1200 1400 1600
Voltage (V)
October 530.785 0.835
November 566.789 0.895
Figure 12: P-V characteristics at varying solar irradiance
December 629.266 0.999
C. Solar PV performance in Upington Area
Upington real data as obtained from the South African
Weather Service was used for simulations to compare the VI. CONCLUSION
monthly data of the area, for the purpose of evaluating the
plant’s reliability which is not covered in this paper. Upington Modelling and simulation of a photovoltaic system was
data obtained was limited to solar irradiance; therefore presented and its performance under various irradiance and
temperature was kept at 25 ºC, for the purpose of this study. temperature conditions was observed. The study revealed that
Fig. 13 highlights the performance of the plant under different solar PV plant performance is dependent on both ambient
seasons, showing power obtained from the plant in different temperature and solar irradiance. SPWM inverter was also
months. With daily average irradiance levels per month for the designed and simulated, with the obtained results as expected
area varying from 629.27 W/m2 in December to 304.39 W/m2 i.e. conversion of DC-AC. MPPT method was discussed and
in June, it can be observed that the plant performs better the integrated system model will form part of the future
during summer seasons and decays during winter seasons. studies. Furthermore, evaluation of the plant’s reliability using
different methods i.e. neural network and fuzzy logic will be
conducted.

REFERENCES
[1] DoE, GIZ; “State of Renewable Energy in South Africa”, Dep. Of
Energy, South Africa, September 2015.
[2] Chapter 5 Solar Photovoltaics [online]. Available:
www.uprm.edu/aret/docs/Ch_5_PV_systems.pdf.
[3] S I. Rasel, R.N. Ali et al.,”Design & Simulation of Grid Connected
Photovoltaic System using Simulink”, Dhaka, Bangladesh, 17-19
December 2015.
[4] Sunmodule; “SW 250 mono/ Version 2.0 and 2.5 Frame Technical
specifications”, Solar world.
[5] Tomas Markvart, Solar Electricity, 2nd ed, Wiley, 2000, 1994.
[6] Ruben Inzunza, Ryota Okuyama, Tsuguhiro Tanaka, Masahiro
Figure 13: Solar PV plant monthly performance for Upington area Kinoshita, “Development of a 1500Vdc photovoltaic inverter for
utility-scale PV power plants”, Toshiba Mitsubishi-electric industrial
Plant performance shown in Fig. 13 is also tabulated in systems corporation, Tokyo, Japan, 2015 IEEE.
Table III, showing the actual obtained maximum power per [7] N. Mohan, T.M. Undeland and W.P. Robbins, Power Electronics,
Converters, Applications and Design, John Wiley & Sons, Inc., 3rd ed,
month. chapter 8, 2003.
[8] P.I. Muoka, M.E. Haque, A. Gargoom, M. Negnevitsky, “Modeling,
simulation and hardware implementation of a PV power plant in
distributed energy generation system”, 978-1-4673-4896-6/13, pg.3,
2013 IEEE.

978-1-5386-4163-7/18/$31.00 ©2018 IEEE 255

Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE ENGENHARIA DE ITAJUBA. Downloaded on April 27,2021 at 20:46:29 UTC from IEEE Xplore. Restrictions apply.