Verifying an economic viable load for experimental purposes relating to small scale PV modules

Abstract

1 aswart@cut.ac.za 2 phertzog@cut.ac.za Abstract—Optimizing the output power of any PV module requires a number of factors to be considered, including the tilt angle, orientation angle, environmental conditions and the energy management system. This system often includes a maximum power point tracker that is required to adjust a module's output voltage to a value which enables the maximum energy to be transferred to a given load. A solar controller may also be used in the energy management system to prevent batteries from overcharging, to prevent back flow of current from the batteries to the solar modules and to provide maximum reliability and service life of the whole system. However, when various parameters of PV modules need to be investigated in real life applications, what type of economic viable load is suitable for experimental purposes relating to small scale PV modules? The purpose of this paper is to present empirical evidence contrasting the performance of three identical 10 W polycrystalline modules connected to three unique separate loads. A LabView software program was developed to record and display the voltage and current measurements from the PV modules using a data logging interface circuit and an Arduino board. Results indicate that a solar controller extracts more power from a PV module (on average 3.9% more power), as compared to a regulated LED and a fixed load resistor. However, the regulated LED follows a profile similar to that of the solar controller, drawing on average 2 W less per day than the solar controller. On the other hand, the fixed load resistor draws on average 8 W less per day than the solar controller, following a profile different to that of the solar controller and regulated LED. The regulated LED is therefore verified as an economic viable load for experimental purposes involving small scale PV modules.

Full text

Verifying an economic viable load for experimental purposes
relating to small scale PV modules
Arthur J Swart1 and Pierre E Hertzog2
1,2Department of Electrical, Electronics and Computer Engineering
Central University of Technology, Private BagX20539, Bloemfontein, 9300
1aswart@cut.ac.za
2phertzog@cut.ac.za
Abstract—Optimizing the output power of any PV module
requires a number of factors to be considered, including the tilt
angle, orientation angle, environmental conditions and the
energy management system. This system often includes a
maximum power point tracker that is required to adjust a
module’s output voltage to a value which enables the maximum
energy to be transferred to a given load. A solar controller may
also be used in the energy management system to prevent
batteries from overcharging, to prevent back flow of current
from the batteries to the solar modules and to provide
maximum reliability and service life of the whole system.
However, when various parameters of PV modules need to be
investigated in real life applications, what type of economic
viable load is suitable for experimental purposes relating to
small scale PV modules? The purpose of this paper is to
present empirical evidence contrasting the performance of
three identical 10 W polycrystalline modules connected to three
unique separate loads. A LabView software program was
developed to record and display the voltage and current
measurements from the PV modules using a data logging
interface circuit and an Arduino board. Results indicate that a
solar controller extracts more power from a PV module (on
average 3.9% more power), as compared to a regulated LED
and a fixed load resistor. However, the regulated LED follows a
profile similar to that of the solar controller, drawing on
average 2 W less per day than the solar controller. On the
other hand, the fixed load resistor draws on average 8 W less
per day than the solar controller, following a profile different
to that of the solar controller and regulated LED. The
regulated LED is therefore verified as an economic viable load
for experimental purposes involving small scale PV modules.
Keywords— Arduino, LabView, Metrology, regulated LED,
fixed resistor, solar controller, load
I. INTRODUCTION
“You cannot have a society where you spend more than
you earn. I mean, it's just fundamentally not viable in the
long run.” These words, uttered by Azim Premji, an Indian
businessman, pinpoint a fundamental universal principle in
that one cannot use more than what one produces. This is
not viable or sustainable, and will lead to ruin. This
principle applies equally well to photovoltaic (PV) modules
used in renewable energy systems. It is impossible to use
more energy than what a PV module can provide. However,
it is imperative to try to optimize the output power of a PV
module. Optimum PV module installations (tilt and
orientation angles) are therefore advocated [1, 2] along with
maximum power point trackers (MPPT) that adjust a
module’s output voltage to a value which enables the
maximum energy to be transferred to a given load [3].
However, many MPPT are expensive leading some to
choose a more economically viable option being a solar
controller that regulates current and prevents overcharging
of the storage device [4]. Both these options have been
reported on in the literature as critical components on
renewable energy management system.
However, are these components really necessary for
determining various parameters of a PV module in real life
applications? That would depend to a large degree on what
applications are considered. If the maximum output power
of a PV module is required under varying atmospheric
conditions, then a MPPT would be required. However, if the
relationship between the incident angle of light on a PV
module’s surface and its associated output power is to be
ascertained, then a solar controller would be suitable. An
even more economical viable option would be a regulated
light-emitting diode (LED) that would not require a solar
controller or its associated storage device.
The purpose of this paper is to present empirical evidence
contrasting the performance of three identical 10 W
polycrystalline modules connected to three unique separate
loads in order to establish an economic viable load when
considering similar output power results. These loads
include a 12 V battery connected to a 5 A solar controller, 2
x 4 W 12 V non-regulated LED lamps, 2 x 5 W 12 V
regulated LED lamps and 2 x 39 Ohm 10 W fixed load
resistors connected in parallel. A theoretical comparison
between different PV module energy management systems
is firstly presented. Secondly, the research methodology is
given, followed by a detailed explanation of the
experimental setup. Results and conclusions complete the
paper.
II. PV MODULE ENERGY MANAGEMENT
PV modules receive direct (beam), diffused and reflected
radiation during varying atmospheric conditions [5]. Direct
beam radiation is the component which enjoys direct line-
of-sight between the sun and the PV module. Diffused
radiation is the component scattered by atmospheric
constituents such as molecules and clouds [6]. Reflected
radiation occurs when light energy is reflected off trees or
buildings towards the PV module.
There are various methods to extract the radiation
received by a PV module or array. The simplest method
would be to couple a PV module directly to a given load.
However, this may not be the most efficient way to extract
the maximum amount of energy from a PV module for any
given radiation condition. An alternative, and more
acceptable method, would involve the use of an energy
management system that would regulate the flow of current
between a PV module and a given load. This would involve
the use of a MPPT or a solar controller.
The main role of a solar controller is to protect the storage
device [7]. Solar controllers (also called solar regulators) are
Page 396 Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2016

rated by the maximum amount of current they can regulate
from a PV array or module [8]. Many often include a simple
switching technique (on/off) for both simplicity of design
and operational reliability. Basic solar controllers are
relatively inexpensive in South Africa, with a 15 A pulse
width modulated version costing approximately R180 from
Mantech Electronics in Johannesburg (see Table 1).
TABLE 1: Comparison of load conditions
Load
condition
Principle
used
Local
cost
Main
advantage
Main
disadvantage
MPPT
10 A
Multi
point
power
tracking
R1360
Increase in
charge
efficiency
up to 30%
High relative
cost
Solar
controller
15
A
Pulse
Width
Modula-
tion
R180
Low power
applications
have better
energy
harvest
ing
Less efficient
than
a
MPPT
Regulated
LED
12 V 4 W
Regulated
electronic
circuit
R42
Relative
low
cost
Difficult to
predict the
characteristics
of
the
electronic
circuit
Fixed
resistor
10 Ohm
10 W
Resistive
load
R10
Very low
cost
Not efficient
under varying
input energy
conditions
More advanced solar controllers are referred to as MPPT.
Not only do they regulate the flow of current, but they also
are used to extract the maximum power from one or more
PV modules under various environmental and operating
conditions [9]. However, they are more expensive than solar
controller or regulated LED lamps.
Regulated LED lamps have the advantage of not requiring
a storage device and of being relatively inexpensive. This
makes them ideally suited to evaluate the applications of
identical PV systems in real life scenarios, as the number of
variables is reduced. It is well known that battery-to-battery
variations in e.m.f at a given state of charge may be in the
order of 50 mV for every 2.25 V cell, due to variations in
the manufacturing process, ageing and charge-discharge
cycling [10]. This may result in a variation between two
identical PV systems’ storage device of 13 % (0.05 / 2.25 x
6 cells per battery x 100%). This may impact negatively on
a comparison study in which two or more identical PV
systems are evaluated under specific conditions. Tight
regulated LED current, high efficiency and satisfactory
power factor have all been achieved with only one power
stage within the LED [11].
Fixed load resistors are the cheapest option when it comes
to load conditions for PV modules. However, its main
drawback is that it severely reduces the source voltage if it is
attempting to draw more current than what is available
(resistor value is too small). On the other hand, it will limit
the current drawn from the PV module and not use all the
available current when the resistor value is too high.
III. RESEARCH METHODOLOGY
An experimental research design was used where three
PV modules were set to the same tilt angle of 29º (Latitude
value of the installation site), each with its own 18 Ohm
10 W fixed load resistor. Four weeks of data (October 2015)
were then recorded to observe any significant differences
between the three systems, which could then be calibrated
use specific factors in the software. A coefficient of
variation of 1.4% was calculated indicating that all three
systems were performing equally well.
The three PV modules were then connected to different
load conditions, resulting in only one different variable. All
other variables (environmental conditions, orientation and
tilt angles, etc.) were standard for the three systems. Data
was recorded from mid November 2015 through the middle
of May 2016. On the 17th of February 2016, the 4 W non-
regulated LED was replaced with a 5 W regulated LED.
This was done to compare the results of the two unique LED
configurations, while at the same time increasing the current
drawn from the second PV module. The 4 W non-regulated
LED lamp is a standard design with no built-in regulation
circuit. However, the 5 W regulated LED lamp features new
technology with a built-in regulator circuit to accommodate
larger voltage fluctuations.
IV. EXPERIMENTAL SETUP
The experimental setup consists of three identical PV
systems comprising 10 W polycrystalline PV modules, a
data logging interface circuit, an Arduino board and
LabVIEW software (see Fig. I for the block diagram). Three
different load conditions are used, which include a solar
controller (5 A) connected to a 12 Ah battery, two parallel
LED lamps (4 W non-regulated and then 5 W regulated) and
two parallel 39 Ohm 10 W fixed load resistors. Therefore,
the only variable which is different between the three
identical systems is the load condition.
Figure I: Experimental setup
The solar controller is an entry level controller that can
regulate no more than 5 A. This is suitable for use with the
Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2016 Page 397

10 W PV module that has a short circuit current of 0.78 A
and an open circuit voltage of 20.8 V. The maximum power
point voltage of the PV module is 16.5 V, with a maximum
power point current of 0.61 A. Two 4 W LED’s are
connected in parallel to the output of the solar controller and
serve as the load resistance for the battery. The solar
controller regulates the current flow to these LED’s,
switching then either on or off depending on the state of
charge of the 12 Ah battery. This ensures that optimum
energy is constantly drawn from the PV module during
daylight hours to charge the 12 Ah battery. A 6 Ohm 10 W
series resistor is used between the solar controller and PV
module for current sensing measurements.
The 4 W and 5 W LED lamps were also selected using
the maximum power point voltage and current of the PV
module. Two parallel 4 W 12 V non-regulated LED lamps
were initially connected directly to the PV module, each in
series with its own 10 Ohm 10 W resistor. This series
resistor accommodates the voltage drop resulting from the
difference between the PV modules output voltage and that
required by the LED. A 6 Ohm 10 W series resistor is used
between the two parallel LED’s and the PV module for
current sensing measurements. This means that the
maximum series resistance for one LED branch would be
16 Ohm, resulting in a maximum current flow of 0.281 A
(16.5 V – 12 V divided by 16 Ohm). However, this is for
one LED branch. Two branches exist which means that 92%
of the maximum power point current (0.562 A divide by
0.61 A) should be drawn by the two parallel 4 W LED
lamps during the maximum period of daily solar radiation.
This was eventually changed to a 5 W 11 – 13 V regulated
LED lamp to enable a higher amount of current to be drawn
from the PV module. Using LED lamps, instead of a MPPT
or a solar controller with a given load, has been used before
as an economical viable load in determining the acceptance
zone and switch-on times of specific PV modules [12, 13].
The two parallel 39 Ohm 10 W resistors were also
selected using the maximum power point voltage and
current of the PV module. The parallel branch results in a
series resistance of 19.5 Ohm which is directly connected to
the PV module by means of a 6 Ohm 10 W series resistor
that is used for current sensing measurements. This means
that the total load resistance for the PV module is 25.5 Ohm,
resulting in a maximum current flow of 647 mA during the
maximum period of daily solar radiation. Using a fixed load
resistance, instead of a MPPT or a solar controller with a
given load, is an effective and easy method to start loading
PV modules located outdoors for measurement purposes
[14, 15]. Table 2 summarizes the load conditions.
The data logging interface circuit has been reported on by
a number of researchers [16, 17] and provides power
conditioning between the PV system and the Arduino board
which is connected to the personal computer and interfaced
with LabVIEW software. The use of the Arduino board and
the LabVIEW software as a data logger has been reported
on by Hertzog and Swart [1, 18].
The three PV modules were mounted onto an aluminum
frame and set to the same tilt angle equal to the Latitude of
the installation site (29° South). The same load condition
was initially used with all three PV modules, being three
separate 39 Ohm 10 W fixed resistors in order to calibrate
the system. The output power of these modules was then
recorded and analyzed using LabVIEW software in
conjunction with an Arduino board. Results were obtained
over a four week period (October 2015) which indicated a
coefficient of variation of 1.4%. This coefficient of variation
is calculated using the standard deviation and mean of the
collected data. This ensures the reliability and validity of
subsequent electronic measurements when the three PV
modules are connected to different load conditions, as
described earlier.
TABLE 2: Load conditions
PV Module and
load condition
Series
resistors
Calculating current
with Ohm’s law
PV1
– 2
x 39 Ohm
10 W
resistors
in
parallel
6 Ohm 10 W
current sensing
resistor
= 16.5
19.5+ 6
I
max = 0.647 A
PV2
–
2 x 4 W non-
regulated LED
lamps
in parallel
6 Ohm 10 W
current sensing
resistor and a 10
Ohm 10 W
series resistor
per lamp
=!16.5 " 12
10+ 6 # ×2
I
max = 0.537 A
PV2
– 2 x 5 W
regulated LED
lamps in parallel
6 Ohm 10 W
current sensing
resistor and a 10
Ohm 10 W
series resistor
per lamp
=!16.5 " 11.7
10+ 6 # ×2
I
max = 0.6 A
PV3
– Solar
controller
,
12 Ah
battery and 2 x 4 W
non
-
regulated LED
lamps in parallel
6 Ohm 10 W
current sensing
resistor
No calculation
I
max = 0.610 A
Voltage readings are obtained from the Arduino board by
using the analog read function in LabVIEW. The obtained
values are multiplied by a predetermined factor for
calibration and to compensate for any interface losses. This
value is displayed on the front panel of the LabVIEW
software that is visible on the screen. This value is then
filtered by a Butterworth Filter that is used to filter out high
frequency components that come from the Arduino’s analog
read circuit and any other high frequency noise present in
the data logging system.
Voltage readings from the Arduino board represent PV
module output voltages and currents. Current readings are
obtained by measuring the voltage across a low value high
power precision resistor (6 Ohm 10 W 1%). Voltage
readings are obtained by using a standard voltage divider
circuit (147 kΩ resistor in series with a 100 kΩ resistor).
Multiplying the voltage and current readings within
LaBVIEW yields a power reading in Watts that is written to
a matrix for recording purposes. The total amount of power
extracted per day from each PV module was then recorded
in a singular text file for further analysis.
V. RESULTS AND DISCUSSION
Fig. II presents the LaBVIEW software interface which
was developed by Hertzog and Swart to display electronic
measurements obtained by the Arduino board. The
following points have been highlighted:
· A: Date stamp highlighting the date (11 May 2016)
when the data was recorded;
Page 398 Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2016

· B: Number of samples (4320) taken over the
sample time (12 h);
· C: Total Wh recorded from PV module 3 for the
specified day – In this case the total power is 53.92
Wh for the solar controller, which is 3.8 Wh more
than the LED.
· D: Instantaneous power calculated for each sample
for the fixed resistor (black line), the LEDs (red
line) and the solar charger (blue line) – In this case
it is 0 W, as the sample period has ended.
· E: Dip in power is visible for the fixed resistor load
(black line) due to a pigeon which sat on the PV
module.
· F: Start of the sample time is 06:00 with no solar
radiation present – A rise in the current drawn from
the PV modules is evident from 07:06.
· G: Current factors which are multiplied by the
measured values from the Arduino board to obtain
the actual measurements – 3 different factors exist
due to the initial calibration of the system during
October 2015.
· H: Blue line showing the voltage curve of PV
module 3 connected to the solar charger.
· I: Red line showing the voltage curve of PV
module 2 connected to the LED’s.
· J: Black line showing the voltage curve of PV
module 3 connected to the fixed resistors.
· K: Voltage factors which are multiplied by the
measured values from the Arduino board to obtain
the actual measurements – 3 different factors exist
due to the initial calibration of the system during
October 2015.
Fig. III indicates the total power extracted from the PV
modules for November 2015 through February 2016 for
different load conditions. During this period, the 4 W non-
regulated LED lamp (red line) was used which resulted in a
lower amount of power been extracted from the PV module
as compared to the solar charger (green line). The 4 W non-
regulated LED lamp even extracted less power than what the
fixed load resistors (FLR) did (blue line). All power values
below 30 W are considered to be the effect of cloud
movement resulting in a disruption of direct beam radiation
which is required for optimum output power from a PV
module.
Fig. IV highlights the total power extracted from the PV
modules for February 2016 through May 2016. Here the
4 W non-regulated LED was replaced with 2 x 5 W
regulated LED lamps. This resulted in a larger amount of
power been extracted from the PV module, as compared to
the previous three months. In fact, the red line (2 x 5 W
regulated LED lamps) now closely follows the green line
(solar controller).
Figure II: LaBVIEW interface showing results for 11 May 2016
Point A
Point B
Point C
Point D
Point E
Point F
Point G
Point H-------------------
Point I----------
Point J------------
Point K
Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2016 Page 399

Figure III: Total power extracted from the PV modules for November 2015
through February 2016 for different load conditions
Figure IV: Total power extracted from the PV modules for February 2016
through May 2016 for different load conditions
Fig. V portrays the average power extracted from the
three PV modules for November 2015 through May 2016,
which is based on the data shown in Fig. IV and Fig V. The
2 x 4 W non-regulated LED lamps extracted, on average,
45 W per day, being the lowest of the three load conditions.
However, the 2 x 5 W regulated LED lamps extracted 51 W
per day, being only 2 W less than the solar charger.
However, if the optimum output power is to be determined,
then a solar controller is still the best choice as it extracted
56 W between November 2015 and February 2016, and then
53 W between February 2016 and May 2016. This decline in
the power extracted by the solar charger is due to the annual
solar radiation curve which has its peak in December and its
trough in June.
Fig. VI shows the total power count for the three PV
modules from November 2015 to February 2016 for the
different load conditions, while Fig. VII illustrates this
same data for the February 2016 to May 2016 period.
Figure V: Average power extracted from the three PV modules for
November 2015 to February 2016 for different load conditions
Figure VI: Total power counts for three PV modules from November 2015
to February 2016 for different load conditions (94 days in total)
Figure VII: Total power counts for three PV modules from February 2016
to May 2016 for different load conditions (86 days in total)
Reviewing Fig. VI reveals that cloud conditions existed
for approximately 10 days of the 94 days from November
2015 to February 2016. This equates well to recent weather
reports and news broadcasts detailing the ongoing severe
drought in the Free State region [19]. A similar scenario
exists for the February 2016 to May 2016 time period,
where cloud conditions were experienced for approximately
10 days out of the 86 days. This equates to a cloud coverage
period of 11% for the total time period. Noteworthy though
is the decline in the number of days in which 30 – 60 W was
0
10
20
30
40
50
60
70
80
10-Nov
17-Nov
24-Nov
01-Dec
08-Dec
15-Dec
22-Dec
29-Dec
05-Jan
12-Jan
19-Jan
26-Jan
02-Feb
09-Feb
16-Feb
Power (W)
Date
PV1 = FLR = 2 x 39 Ohm 10 W
PV2 = LED = 2 x 4 W Non regulated
PV3 = SC = 12 Ah battery
0
10
20
30
40
50
60
70
80
17-Feb
24-Feb
02-Mar
09-Mar
16-Mar
23-Mar
30-Mar
06-Apr
13-Apr
20-Apr
27-Apr
04-May
11-May
Power (W)
Date
PV1 = FLR = 2 x 39 Ohm 10 W
PV2 = LED = 2 x 5 W Regulated
PV3 = SC = 12 Ah battery
30
35
40
45
50
55
60
Average Nov - Feb
Average Feb - May
Power (W)
Time period
PV1 = FLR = 2 x 39 Ohm 10 W
PV2 = LED = 2 x 4 W OR 2 x 5 W
PV3 = SC = 12 Ah battery
0
20
40
60
80
< 30 W 30 - 60 W > 60 W
Number of days
Power (W)
PV1 = FLR = 2 x 39 Ohm 10 W
PV2 = LED = 2 x 4 W Non regulated
PV3 = SC = 12 Ah battery
0
10
20
30
40
50
60
< 30 W 30 - 60 W > 60 W
Number of days
Power (W)
PV1 = FLR = 2 x 39 Ohm 10 W
PV2 = LED = 2 x 5 W Regulated
PV3 = SC = 12 Ah battery
Page 400 Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2016

extracted from PV module 2 for the 4 W and 5 W LED
lamps. In Fig. VI, 75 days are observed while in Fig. VII
only 48 days are observed. However, the number of days in
which more than 60 W was extracted increased from 5 days
in Fig. VI to 28 days in Fig. VII. This suggests that the
performance of the 5 W regulated LED lamps is very
closely matched to the performance of the solar controller.
This is substantiated by Fig. V (2 W difference between
the two load conditions) and by Fig. II (blue and red power
curve is very similar). The fundamental difference arises in
the voltage curve of the solar controller and the 5 W
regulated LED lamps, with the solar controller maintaining a
higher PV module voltage than does the LED lamps. The
fixed load resistor, shown in Fig. II, has the worst
performance, with a lower voltage value before 09:00 and
after 15:00. Resistors can be connected in various networks
to acts as a voltage dropper, voltage divider, or current
limiter [20]. However, it can severely reduce the source
voltage if it is attempting to draw more current than what is
available or it may draw less current than what is available.
VI. CONCLUSIONS
The purpose of this paper was to present empirical
evidence contrasting the performance of three identical
10 W PV modules connected to three unique separate loads
in order to establish an economic viable load when
considering similar output power results. The theoretical
analysis highlighted that the local cost for a 15A solar
controller is R180, while 2 x 5 W regulated LED lamps may
cost less than R90. Two fixed load resistors (39 Ohm 10 W)
are the cheapest option at around R10 each.
The experimental setup revealed that only one variable is
different between three identical PV systems, being the load
conditions. Data from these three PV systems was recorded
from November 2016 through May 2016. Results indicate
that a solar controller extracts more power from a PV
module (on average 3.9% more power), as compared to a
regulated LED and a fixed load resistor. The load resistor
extracts approximately 16.7% less power than a solar
controller. However, the 2 x 5 W regulated LED lamps
follows a profile similar to that of the solar controller,
drawing on average 2 W less per day.
It is recommended that regulated LED lamps be closely
matched to the maximum power point voltage and current of
a specific PV module. This may be done by determining the
total series resistance that should be used in conjunction
with the LED lamp to draw a current closely matched to the
maximum power point current of the PV module. This will
enable its use as a viable load, being closely matched to the
performance of an appropriately selected solar controller.
Future research may consider using this practical setup
with larger sized PV modules (20 W and 50 W), matching
the number of regulated LED lamps to their output power.
Obtaining additional results from this practical setup for
winter and spring months may further cement the usefulness
of regulated LED lamps as an economic viable load for
experimental purposes involving small scale PV modules.
Proven advantages include lower costs (less than 50% of the
price for a solar controller) and its close emulation of a solar
controller’s performance. This will adhere to the
fundamental universal principle that one cannot use more
power than what one produces, but can produce an amount
of power which is very close to the maximum extractable
power from a PV module.
VII. REFERENCES
[1] P. E. Hertzog and A. J. Swart, "Determining the optimum tilt angles
for PV modules in a semi-arid region of South Africa for the summer
season," presented at the SATNAC 2015, Arabella, Hermanus,
Western Cape, 2015.
[2] O. Asowata, et al., "Optimizing the output power of a stationary PV
panel," presented at the SATNAC 2013, Stellenbosch, South Africa,
2013.
[3] A. Torres, "Aproveitamento Fotovoltaico Controlado por Redes
Neurais Artificiais Interligado ao Sistema Elétrico," MSc Thesis,
GPEC–DEE–UFC, 1998.
[4] R. Gouws and E. van Niekerk, "Prototype super-capacitor
photovoltaic streetlight with xLogic superRelay functionality,"
International Journal of Emerging Trends in Electrical and
Electronics (IJETEE–ISSN: 2320-9569), vol. 7, pp. 34-39, 2013.
[5] A. A. El-Sebaii, et al., "Global, direct and diffuse solar radiation on
horizontal and tilted surfaces in Jeddah, Saudi Arabia," Applied
Energy, vol. 87, pp. 568-576, 2010.
[6] T. V. Ramachandra and B. V. Shruthi, "Spatial mapping of renewable
energy potential," Renewable and Sustainable Energy Reviews, vol.
11, pp. 1460-1480, 2007.
[7] P. Simona, et al., "ANALYSIS OF RESIDENTIAL
PHOTOVOLTAIC ENERGY SYSTEMS," Universitatii Maritime
Constanta. Analele, vol. 13, p. 143, 2012.
[8] M. Moustafa and E. E. El-bokl, "SOLAR ENERGY FOR RIVER
NILE CRUISERS," Brodogradnja, vol. 65, pp. 61-72, 2014.
[9] A. R. Thomas and S. Mathew, "Solar powered single phase induction
motor using single source five level inverter," in Emerging Research
Areas: Magnetics, Machines and Drives (AICERA/iCMMD), 2014
Annual International Conference on, 2014, pp. 1-6.
[10] M. L. Gopikanth and S. Sathyanarayana, "Impedance parameters and
the state-of-charge. II. Lead-acid battery," Journal of Applied
Electrochemistry, vol. 9, pp. 369-379, 1979/05/01 1979.
[11] P. Fang and Y. F. Liu, "Single stage primary side controlled offline
flyback LED driver with ripple cancellation," in Applied Power
Electronics Conference and Exposition (APEC), 2014 Twenty-Ninth
Annual IEEE, 2014, pp. 3323-3328.
[12] A. J. Swart and P. E. Hertzog, "Validating the acceptance zone for PV
modules using a simplified measuring approach," presented at the
ECTI-CON, 13th International Conference on Electrical Engineering,
Electronics, Computer and Information Technology, Chiang Mai,
Thailand, 2016.
[13] A. J. Swart and P. E. Hertzog, "Correlating the switch-on time of two
identical PV modules using a simplified measuring approach,"
presented at the ECTI-CON, 13th International Conference on
Electrical Engineering, Electronics, Computer and Information
Technology, Chiang Mai, Thailand, 2016.
[14] P. Debora, et al., "Degradation and others parameters analysis of
different technologie photovoltaic modules," in Photovoltaic
Specialists Conference (PVSC), 2011 37th IEEE, 2011, pp. 002297-
002300.
[15] C. R. Osterwald, et al., "Resistive loading of photovoltaic modules
and arrays for long-term exposure testing," Progress in Photovoltaics:
Research and Applications, vol. 14, pp. 567-575, 2006.
[16] A. Ozemoya, et al., "Experimental Assessment of PV Module
Cooling Strategies," presented at the SATNAC 2014, Boardwalk
Conference Centre, Nelson Mandela Bay, Eastern Cape, South Africa,
2014.
[17] A. J. Swart, et al., "Cost-effective energy monitoring of domestic off-
grid PV systems," JEPE, Journal of Energy and Power Engineering,
vol. 5, pp. 182-188, 2013.
[18] P. E. Hertzog and A. J. Swart, "A customizable energy monitoring
system for renewable energy systems," presented at the SAUPEC
2015, University of Johannesburg, 2015.
[19] T. Letshaba, "Taps to run dry in three Free State towns," in SABC, ed,
2016.
[20] I. Kostić, et al., "Thermal Characterization of the Overload Carbon
Resistors," International Journal of Photoenergy, vol. 2013, 2013.
James Swart received his DTech: Electrical: Engineering degree
in 2011 from the Vaal University of Technology. His research
interests include engineering education and alternative energy. He
is currently an Associate Professor at CUT.
Pierre Hertzog received his DTech: Electrical: Engineering degree
in 2004 from the Central University of Technology. His research
interests include alternative energy and automation and
manufacturing systems.
Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2016 Page 401