Original Paper

Measurement and Control

1–10
A simple solar simulator with highly Ó The Author(s) 2019
Article reuse guidelines:
stable controlled irradiance for solar sagepub.com/journals-permissions
DOI: 10.1177/0020294019827327

panel characterization journals.sagepub.com/home/mac

Rahmat Awaludin Salam1,2,3 , Muhammad Miftahul Munir1,2 ,

Thariq Warsahemas1, Casmika Saputra1,2, Hamzah Latief2,4 and
Khairurrijal Khairurrijal1,2

Abstract
A solar simulator suitable for universities’ lab was designed and developed using a quartz tungsten halogen lamp as its
light source, an alternating current phase-cut dimmer, a light intensity meter, and an ATMega328p microcontroller with a
computer and a liquid crystal display. Noting that the quartz tungsten halogen lamp suffers bulb overheating and long-
term degradation that leads to the decrease in its light intensity, a control mechanism was applied. The control mechan-
ism employed a proportional–integral–differential action with Some-Overshoot Ziegler–Nichols tuning rule. It was
shown that the control mechanism works well in stabilizing the quartz tungsten halogen lamp irradiance between 273
and 1182 W/m2. The developed solar simulator was then tested to obtain I-V (current–voltage) characteristics of 3 W
peak and 5 W peak commercial solar panels (GH Solar, GH5P-9). Based on the gained I-V characteristics, it was shown
that the obtained characteristics of the commercial solar panels are in the range of the characteristics provided by the
manufacturer’s data sheets. The developed quartz tungsten halogen lamp-based solar cell simulator can therefore be
used to characterize solar cells.

Keywords
Control, light intensity, quartz tungsten halogen lamp, solar simulator, solar cell

Date received: 27 July 2018; accepted: 22 December 2018

Introduction power supply to the monitoring systems. Moreover, other

Solar simulators, which are important tools in solar
energy research, fall into two major categories: non-
concentrating uniformly distributed lights and high-flux
concentrators.1 A group of solar simulators with non-
concentrating uniformly distributed lights are applied
in the examination of solar cells and solar thermal col-
lectors.2–5 Another group of solar simulators with high-
flux concentrators are operated in producing high tem-
peratures for research in materials processing, thermo-
chemical reactions, solar fuels production, and simu-
lated solar thermal power generation.6–10 Solar cells
along with batteries and power managements were used
to supply electrical power for various monitoring sys-
tems, such as natural disaster monitoring systems,11–15
structure health monitoring,16 aquaculture monitoring,17
and underground mining area monitoring,18 which is
placed in remote or inaccessible areas. Tests of solar cells
along with their batteries and power managements using
a non-concentrating uniformly distributed solar simulator
are therefore very important to make sure the continuous
applications of a non-concentrating uniformly distributed
solar simulator can be found in various fields, such as dis-
infection on wastewater treatments,19 phototherapy,20
1
Department of Physics, Faculty of Mathematics and Natural Sciences,
2
Institut Teknologi Bandung, Bandung, Indonesia
Research Center for Disasters Mitigation, Institute for Research and
Community Services, Institut Teknologi Bandung, Bandung, Indonesia
3
Department of Engineering Physics, School of Electrical Engineering,
Telkom University, Bandung, Indonesia
4
Department of Oceanography, Faculty of Earth Science and Technology,
Institut Teknologi Bandung, Bandung, Indonesia
Corresponding authors:
Muhammad Miftahul Munir, Department of Physics, Faculty of
Mathematics and Natural Sciences, Institut Teknologi Bandung, Bandung
40132, Indonesia.
Emails: miftah@fi.itb.ac.id
Khairurrijal Khairurrijal, Department of Physics, Faculty of Mathematics
and Natural Sciences, Institut Teknologi Bandung, Bandung, 40132,
Indonesia.
Email: krijal@fi.itb.ac.id

Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License
(http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without
further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/
open-access-at-sage).
2 Measurement and Control
photo-degradation process,21 photo-catalysis,22 and
Stirling engine investigation.23,24
There are plenty of solar simulators that have been
developed for solar cells testing.25–29 Light sources uti-
lized by the solar simulators are (a) carbon arc, high-
pressure sodium vapor and argon arc lamp, (b) quartz
tungsten halogen (QTH) lamp, (c) mercury xenon
lamp, (d) xenon arc and flash lamp, (e) metal halide
lamp, (f) light emitting diode (LED), (g) super conti-
nuum laser, and (h) multi-light source. In order to get
full spectra of natural sunlight with low power con-
sumption, QTH-based solar simulators were usually
used at present.26,30,31 However, QTH lamps suffer
bulb overheating and long-term degradation of the
bulb and filter.32 These disadvantages then cause QTH
lamp intensity instable and even reduce their intensity
over time.33
To overcome the disadvantages, a control mechan-
ism needs to be applied to the solar simulator system.
Sarwar et al.6 have implemented the control mechanism
to their high flux solar simulator. However, they did
not mention clearly the adjustment of the flux emitted
on their solar simulator. Instead of controlling the light
intensity, the control mechanism was used to control
the temperature and the motion of Lambertian target
that will be tested. A control mechanism was also
implemented to control current that fed to the LED-
based solar simulator developed by Bazzi et al.34 They
control every LED color using a controlled boost con-
verter that provides a sufficient input voltage for each
LEDs and a Hall-effect sensor to measure the current
fed to the LED. A similar control was also applied by
Stuckelberger et al.28 to control their solar simulator.
The difference is that they control their LED-based
solar simulator using any cheap component to drive
the LEDs and a series of analog-to-digital converter
(ADC) and digital-to-analog converter (DAC) to con-
trol the current fed to the LED automatically from the
computer. However, a manual control is still employed
for the QTH-based solar simulator.4,26,31,35,36 Noting
that the light intensity of QTH lamp has a tendency to
decrease over the time and also instable, an automatic
control is therefore important to be applied to QTH-
based solar simulators to keep a certain and stable light
intensity.
This paper reports developments of a solar simula-
tor using a QTH lamp and its automatic control system
to provide long-term illuminance stability. The plant of
the control system was the QTH lamp and the light
intensity was read by the calibrated sensor utilizing a
light-dependent resistor (LDR). The control mechan-
ism used a proportional–integral–differential (PID)
action set by applying the Some-Overshoot (SO)
Ziegler–Nichols (ZN) tuning rule, which is a better tun-
ing rule. After all constants have been gained, the con-
trol mechanism was applied, and the stability of the
control system was achieved for various set points. All
aspects of the developments will be discussed thor-
oughly in this paper.
Simulator design and development
This section describes design and development of the
controlled solar simulator, which consists of two main
parts: system hardware and its software.
System hardware
The QTH lamp–based solar simulator, which consists
of a plant, an AC phase-cut dimmer, a light intensity
meter, and an ATMega328p microcontroller with a
computer and a liquid crystal display (LCD), was
developed as shown by the block diagram in Figure 1(a).
The plant is a QTH lamp with controlled intensity. Since
the QTH lamp needs AC power, its light intensity is con-
trolled by the phase-cut dimmer and measured by the
light meter. All processes of the simulator are controlled
by the microcontroller, which has an Advanced Versatile
RISC (AVR) central processing unit (CPU), an ADC, a
timer interrupt, a 32-kB flash memory, and 2 bytes
RAM. The measurement results are then displayed on
the LCD and computer in real time. All parts of the
simulator except the computer and LCD along with
the solar panel under test were arranged and placed in
the box with a dimension of 40 3 40 3 40 cm3 as shown
in Figure 1(b) that can be used for characterizing a small
solar panel.
Figure 2 displays the AC phase-cut dimmer, which is
a triac-based device that switches the dimmer to con-
duction mode when the triac receives the gating signal,
which comprises two circuits used for zero-crossing
detection and phase cut.37,38 To initiate the dimming
process, the zero-crossing occurrence of AC power line
input voltage is first detected using the zero-crossing
detector, which is built by the MICW10N diode bridge
and the 4N35 optocoupler. The phase cutter, which is
assembled by the MOC3021 phototriac driver and the
BT136 triac, immediately cuts the input voltage phase
by giving a certain delay time. Therefore, the dimmer
controls the output power by controlling the conduc-
tion time; a shorter conduction time, which is a longer
delay time, will cause less power to be delivered to the
output.37–41 The working principle of the phase-cut
dimmer is as follows. The AC power line input voltage
is fed to the diode bridge as a full-wave rectifier. When
the rectified voltage is higher than 1.3 V, the optocou-
pler output is zero. Otherwise, the output is 5 V, which
is as high as the voltage at the optocoupler collector.
Therefore, the optocoupler output is a pulse train in
which the pulse is sent to the microcontroller to make
an interrupt as shown in Figure 1(a). The interrupt then
activates a timer that delays the gating signal given to
the gate of phototriac used to drive the triac. The AC
power line output voltage across the terminals of triac,
which is supplied to the QTH lamp as shown in Figure
1(a), is no longer sinusoid as its phase is cut and the
QTH lamp dims so that its intensity decreases.
As presented in Figure 1, the light intensity is
then sensed by the light intensity meter as shown in
Salam et al.
Figure 1. (a) Block diagram of developed solar simulator and (b) its layout in a box.
Figure 2. Phase-cut dimming circuit.
Figure 3. It consists of a Wheatstone bridge with the
GL5516 LDR as the intensity sensor and a signal con-
ditioning circuit based on a differential amplifier using
the LM358 op-amp. The LDR has resistances of
4.52 Mv at the dark condition and 296.3 v when the
LDR is highly illuminated by the QTH lamp. The
Wheatstone bridge’s potentiometer R1 functions to
make the bridge output voltage (the voltage between
points A and B) null when the light intensity meter
measures a maximum intensity of the QTH lamp. Since
the Wheatstone bridge’s output voltage is low, the
LM358 op-amp-based differential amplifier is used to
fulfill the measurement range of the microcontroller’s
ADC. Noting that the differential amplifier output vol-
tage is given as Vout = (R6/R5)(VA 2 VB), where (R6/
R5) is its voltage gain,42 the signal conditioning circuit
gives a gain of 2.2. The resistance R8 and the capaci-
tance C1 at the differential amplifier output form a
low-pass filter to remove unwanted high-frequency
signals. Hence, the light intensity meter reads the
3
Figure 3. Light intensity meter circuit.
QTH lamp intensity and converts it into an analog
voltage fed to the microcontroller’s ADC as shown in
Figure 1(a).
System software
In order to make the intensity of the solar simulator
continually constant, a closed-loop control mechanism
is then applied to the system. Although an open-loop
control system is easy to build and very stable, a
closed-loop control system is insensitive toward exter-
nal instabilities.43,44 The closed-loop control system
shown in Figure 4 corresponds to the constant-intensity
solar simulator given in Figure 1(a). The plant is a
QTH lamp in the box. Since the plant requires variable
AC power, the AC phase-cut dimmer is employed as
an actuator to supply the variable AC power under a
control action carried out by the controller. The light
intensity meter along with ADC and translator that
employed as the feedback element of the closed-loop
control system convert the analog input to the light
4 Measurement and Control

Figure 4. The closed-loop control representation of the

constant-intensity solar simulator.

intensity. To carry out the control action, the controller

receives an input, which is the error e(t) or a difference
between the desired light intensity Iref and the real light
intensity I(t) measured by the feedback element. The
controller and the translator are an algorithm
embedded in the microcontroller, while the desired light
intensity is an input to the program. The PID control
action operated by the controller to make a constant-
intensity solar simulator has the control signal u(t) as
given in Equation (1)44
ðt
1 de(t)
u(t) = Kp e(t) + e(t)dt + Td ð1Þ
Ti dt
0

where e(t) = Iref 2 I(t), Kp is the proportional constant,

and Ti and Td are known as the integral and derivative
action times, respectively. This control signal adjusts
the delay time given to the AC phase-cut dimmer so
that the light intensity can be controlled.
A program for controlling the constant-intensity
Figure 5. Flowchart used to control solar simulator using PID
solar simulator was built from the flowchart presented
control action.
in Figure 5. In the beginning, the program requests the
computer to initialize the hardware including the micro-
controller and the communication protocols. The error
implementing the mathematical manipulation as pre-
e(t), the voltage V(t), and the light intensity I(t) are also
sented by Grandi et al.26 The relationship between the
initialized. The program then asks for the desired light
light intensities measured by the commercial luxmeter
intensity Iref and then initialized the PID parameters
and those done by the developed light intensity meter
Kp, Ti, and Td regarding the desired light intensity.
at any position of the bottom of the box is depicted in
Next, the program reads the actual light intensity I(t)
Figure 6. The higher the light intensity S given in the
obtained by the plant/process output and calculates the
box, the higher the analog output voltage V of the
error e(t) and the control signal u(t). The control signal
developed light intensity meter as represented by the
u(t) successively alters the delay time given to the phase-
fitting curve V = 0.896 3 ln(G + 14.135) + 2.351 V
cut dimmer/actuator to vary the AC voltage V(t) sup-
and shown in Figure 6(a). This finding is due to the fact
plied to the plant. Finally, the plant gives an output
that the resistance R of LDR decays with the intensity
I(t). If the actual light intensity meets the desired light
S as given by the fitting curve R = 26.6433G20.876 O
intensity, then the control action stops. Otherwise, the
and demonstrated in Figure 6(b).
control action is repeated.

Testing results and discussion The developed AC phase-cut dimmer

The developed light intensity meter
Since the developed light intensity meter output is
analog voltage, the commercial luxmeter (Lutron, LX-
101A) with light intensity output was used as a bench-
mark. Noting that the output of the luxmeter is in lux,
the value is then converted to the W/m2 by discretizing
the spectral radiant distribution curve of the lamp and
As described in the previous section, the AC phase-cut
dimmer is used to control the light intensity. In order
to test the AC phase-cut dimmer, the AC power source
with the root-mean square (RMS) voltage of 18 V and
50 Hz was used. The AC and the zero-crossing detector
waveforms are demonstrated in Figure 7. It was found
that the output waveform of the zero-crossing detector
is a pulse train with the height of 5 V and the frequency
Salam et al. 5

Figure 6. Relationships between light intensities and (a) the output voltages given by the developed light intensity meter and (b)
the resistances of the LDR.

of 100 Hz (twice the AC power source frequency).

However, the pulse shape was not a square wave due to
the 4N35 optocoupler response time.
Figure 8 shows various delay times of the gating sig-
nal given to the phase cutter in Figure 2 and their out-
put waveforms. As given in Figure 8(a), there is no
phase cut occurs for the delay time Toff of 0 ms (the gat-
ing signal is continuously high at 5 V or the duty cycle
D of the gating signal of 100%), and the AC power is
fully fed to the QTH lamp to obtain a maximum inten-
sity. By increasing the delay time, which implies that
the duty cycle of the gating signal reduces, the phase
cut increases and the AC power that is supplied to the
QTH lamp becomes lesser and its intensity therefore
lowers as depicted in Figure 8(b) and (c). At the delay
time Toff of 10 ms, which is the same as the period of
the pulse train (the gating signal is continuously low at
0 V or the duty cycle D of the gating signal of 0%),
there is no AC power provided as shown in Figure
Figure 7. AC and zero-crossing detector waveforms.
8(d); the QTH lamp therefore delivers no intensity.
These findings agree well with the theory.45 Note that
the filament of QTH lamp is a partially inductive load Appendix. However, the black square symbol repre-
to the AC power. It therefore stores the electrical sents the light irradiance that was experimentally mea-
energy derived from the change in current.46 The higher sured by using the luxmeter (Lutron, LX-101A). It is
the current flows, the greater the electrical energy is clearly seen that the theoretically calculated light irradi-
stored in it. As a result, it has been observed that the ance does not fit the experimentally measured one. This
delay time up to 1.75 ms (or the duty cycle higher than is because the theoretical formulation of light intensity
82.5%) will not cut the phase of the AC power so that in Equation (A.11) of the Supplemental Appendix does
the maximum AC power is fed to the QTH lamp. In not take into account the QTH lamp as a partially
addition, the delay time down to 8.25 ms (or the duty inductive filament and the commutating dV/dt charac-
cycle less than 17.5%) will cut the phase of the AC teristic of the assembly of the MOC3021 phototriac
power so that no AC power is provided to the QTH driver and the BT136 triac. The effect of the delay time
lamp. on the light intensity of the partially inductive QTH
The light irradiance as a function of the delay time is lamp is given when discussing Figure 8. The delay time
depicted in Figure 9. The theoretically calculated light of 2 ms that will not cut the phase of the AC power
irradiance, which is given by the red line, was obtained must be used as a correction when fitting the theoreti-
by applying Equation (A.11) of the Supplemental cally calculated light intensity to the experimentally
6 Measurement and Control

Figure 8. Gating signals of the phase cutter for various delay times: (a) Toff of 0 ms (D of 100%), (b) Toff of 2 ms (D of 80%),
(c) Toff of 6 ms (D of 40%), and (d) Toff of 10 ms (D of 0%) and its output waveforms.

measured one. Therefore, the corrected theoretical

delay time Toff, corr is given by Toff, corr = A 3 Toff 2 B,
where Toff is the uncorrected delay time given in
Equation (A.11) of the Supplemental Appendix, B is
the constant of 2 ms due to the delay time that will not
cut the phase of the AC power, and A is the constant to
change the slope of the curve in Figure 9 due to the
commutating dV/dt characteristic which is shown by
the blue line.

The developed control system

The tuning rules of PID controller usually used are the
classic ZN tuning rule and its modifications. Generally,
these tuning rules get the control parameters (Kp, Ti,
and Td) based on experimental results. The common
method in performing the ZN tuning is the marginal Figure 9. Comparison of the light irradiances, which were
stability of Kp response that requires the knowledge of theoretically calculated and experimentally measured, as a
ultimate/critical gain, Ku, and also the ultimate period, function of the delay time.
Salam et al. 7

Table 1. Constants for the PID control.

Set point (W/m2) Kp Ti (ms) Td (ms)

273 4.95 520 260

363 4.25 420 210
454 4.00 420 210
545 3.50 400 200
636 3.62 400 200
727 3.00 400 200
818 3.33 400 200
909 3.67 400 200
1000 4.00 400 200
1091 4.00 400 200
1182 4.50 400 200

Figure 11. Transient responses of the system with and without

control mechanism.

parameters were obtained first. The PID control para-

Figure 10. Transient responses of the system after employing
the Ziegler–Nichols (ZN) tuning rule and its modifications for
the set point irradiances (Gsps) of 273 and 636 W/m2.
Tu.44,47 Since the step response may have an unaccepta-
ble huge maximum overshoot,44 modifications of the
ZN tuning rule have widely been proposed including
the SO- and No-Overshoot (NO)-ZN rules. The SO-ZN
is a simple ZN tuning rule modification that reduces the
overshoot,47 and it is the better tuning type due to its
quick response and minimum overshoot.48
Figure 10 illustrates transient responses under the
applications of the ZN tuning rule and its modifica-
tions. It is shown that the SO-ZN tuning rule gave bet-
ter responses for all tested set points. The differences
are clearly shown for the set point irradiance (Gsp) of
636 W/m2. The SO-ZN rule gave the lower overshoot
compared to that of the ZN, while the set point was
achieved faster than those of the other tuning rules.
Moreover, although the Gsp of 273 W/m2 was reached
as fast as the ZN tuning rule, the transient response
obtained by the SO-ZN tuning rule was smoother than
that of the ZN as like as that resulted from the NO-ZN
rule. Therefore, the PID control mechanism used in the
developed system was tuned by using the SO-ZN tuning
rule. To implement the control mechanism, the control
meters under the SO-ZN tuning rule were Kp = 0.33 Ku,
Ti = 0.5 Tu, and Td = 0.33 Tu.44,47 The obtained control
parameters are then presented in Table 1. By perform-
ing the control system on the solar simulator, the tem-
poral stability of the solar simulator can absolutely be
determined as an A class anytime is needed.
Figure 11 shows the time stability of lamp irradiance
of 637 W/m2 with and without implementing control
mechanism which are represented by blue and red lines,
respectively. It is seen that without implementing the
control mechanism, the lamp irradiance was unstable.
The lamp irradiance tended to decrease down to 6.22%
after running for 1 h. It is also shown that the decre-
ment of the intensity occurred when the system was
running for the first 15 min. This condition is detrimen-
tal to the testing process that requires a solar simulator
with constant light irradiance. On the other side, the
employment of control mechanism made the lamp irra-
diance constant during the simulation process. This sta-
bility also applied for the lamp intensity of 1000 W/m2.
Testing the developed solar simulator
After having been developed, the solar simulator was
then tested to obtain I-V characteristics of a 3 W peak
(GH Solar, GH5P-9), and a 5 W peak (WP) commer-
cial solar panel (GH Solar, GH5P-9). The photovoltaic
characteristics extracted from the obtained I-V charac-
teristics were then compared to those provided by the
manufacturer’s data sheet. The measured I-V charac-
teristics of the commercial solar panel at standard test
condition are shown in Figure 12. By inspecting Figure
12, the parameters of the solar panel can be extracted
such as the short-circuit current (Isc), open-circuit vol-
tage (Voc), the maximum power (Pm), power maximum
current (Imp), and power maximum voltage (Vmp). The
extraction of the measured I-V characteristics and the
8 Measurement and Control

Table 2. Obtained I-V characteristics of 3 W peak solar panel (GH Solar, GH5P-9).

Photovoltaic characteristics Typical (from data sheet) Testing (from experiment)

Open-circuit voltage (Voc) (10.80 6 0.32) V 11.04 V

Short-circuit current (Isc) (0.367 6 0.011) A 0.365 A
Maximum power (Pmax) (3.00 6 0.09) W 3.08 W
Maximum power voltage (Vmp) (9.0 6 0.27) V 9.29 V
Maximum power current (Imp) (0.334 6 0.01) A 0.331 A

Table 3. Obtained I-V characteristics of 5 W peak solar panel (GH Solar, GH5P-9).

Photovoltaic characteristics Typical (from data sheet) Testing (from experiment)

Open-circuit voltage (Voc) (9.67 6 0.29) V 11.148 V

Short-circuit current (Isc) (0.65 6 0.0195) A 0.65 A
Maximum power (Pmax) (5.00 6 0.15) W 5.06 W
Maximum power voltage (Vmp) (8.93 6 0.27) V 9.02 V
Maximum power current (Imp) (0.56 6 0.0168) A 0.56 A

typical factory characteristics of the solar panel are

then summarized in Tables 2 and 3. As can be seen
from the extracted data, the obtained characteristics
using the developed solar simulator are similar to the
factory characteristics. Although there are some differ-
ences between them, the differences are insignificant. It
can be concluded that the light irradiance emitted by
the solar simulator is comparable to the standard one.

Conclusion
A QTH lamp–based solar simulator has been success-
fully developed. A control mechanism was implemented
in the solar simulator to overcome the decrease in the
QTH lamp light intensity due to the bulb overheating
and long-term degradation. The control mechanism
functioned well in stabilizing the lamp irradiance Figure 12. The measured I-V characteristics of the commercial
between 273 and 1182 W/m2 under a PID action with solar panel.
SO ZN tuning rule. The developed solar simulator was
then employed to obtain I-V (current–voltage) charac-
teristics of two (3-WP and 5-WP) commercial solar Funding
panels. It was obtained that the photovoltaic character-
This work was financially supported by ‘‘Riset Unggulan
istics of the commercial solar panels extracted from the
ITB—PP’’ grant in the fiscal year 2018, the University’s
I-V characteristics were closely similar to those provided Excellence Research (PUPT) grant in the fiscal year 2016, and
by the manufacturer’s data sheets. the Dissertation Research Grant by Ministry of Research,
Technology and Higher Education of the Republic of
Acknowledgements Indonesia in the fiscal year 2017.

R.A.S. gratefully acknowledged the Ministry of Research,

Technology and Higher Education of the Republic of Supplemental material
Indonesia for the Dissertation Research Grant in the fiscal Supplemental material for this article is available online.
year 2017.

ORCID iDs
Declaration of conflicting interests
Rahmat Awaludin Salam https://orcid.org/0000-0002-
The author(s) declared no potential conflicts of interest with 1799-8255
respect to the research, authorship, and/or publication of this Muhammad Miftahul Munir https://orcid.org/0000-0003-
article. 0894-9109
Salam et al.
References
1. Ekman BM, Brooks G and Akbar Rhamdhani M.
Development of high flux solar simulators for solar
thermal research. Sol Energy Mater Sol Cells 2015; 141:
436–446.
2. Ito S, Matsui H, Okada KI, et al. Calibration of solar
simulator for evaluation of dye-sensitized solar cells. Sol
Energy Mater Sol Cells 2004; 82: 421–429.
3. Sabahi H, Tofigh AA, Kakhki IM, et al. Design, con-
struction and performance test of an efficient large-scale
solar simulator for investigation of solar thermal collec-
tors. Sus Energ Technol Assess 2016; 15: 35–41.
4. Namin A, Jivacate C, Chenvidhya D, et al. Construction
of tungsten halogen, pulsed LED, and combined tung-
sten halogen-LED solar simulators for solar cell I-V
characterization and electrical parameters determination.
Int J Photoenergy 2012; 2012: 527820.
5. Meng Q, Wang Y and Zhang L. Irradiance characteris-
tics and optimization design of a large-scale solar simula-
tor. Sol Energy 2011; 85: 1758–1767.
6. Sarwar J, Georgakis G, LaChance R, et al. Description
and characterization of an adjustable flux solar simulator
for solar thermal, thermochemical and photovoltaic
applications. Sol Energy 2014; 100: 179–194.
7. Codd DS, Carlson A, Rees J, et al. A low cost high flux
solar simulator. Sol Energy 2010; 84: 2202–2212.
8. Wang W, Aichmayer L, Garrido J, et al. Development of
a Fresnel lens based high-flux solar simulator. Sol Energy
2017; 144: 436–444.
9. Müller F, Poživil P, van Eyk PJ, et al. A pressurized
high-flux solar reactor for the efficient thermochemical
gasification of carbonaceous feedstock. Fuel 2017; 193:
432–443.
10. Levêque G, Bader R, Lipiński W, et al. Experimental
and numerical characterization of a new 45 kWel multi-
source high-flux solar simulator. Opt Express 2016; 24:
A1360–A1373.
11. Islamy MRF, Salam RA, Munir MM, et al. Design of
deformation monitoring system for volcano mitigation.
J Phys Conf Ser. Epub ahead of print 1 August 2016.
DOI: 10.1088/1742–6596/739/1/012084.
12. Munir MM, Salam RA, Widiatmoko E, et al. Web-based
surface level measuring system employing ultrasonic sen-
sors and GSM/GPRS-based communication. Appl Mech
Mater 2015; 771: 92–95.
13. Munir MM, Mataubenu KDF, Salam RA, et al. Devel-
opment of a wireless sensor network for temperature and
humidity monitoring. Appl Mech Mater 2015; 771: 42–45.
14. Salam RA, Islamy MRF, Munir MM, et al. Design and
implementation of wireless sensor network on ground
movement detection system. In: Proceedings of the 2015
international conference on computer, control, informatics
and its applications (IC3INA), Bandung, Indonesia, 5–7
October 2015, pp. 89–92. New York: IEEE.
15. Chen D, Liu Z, Wang L, et al. Natural disaster monitor-
ing with wireless sensor networks: a case study of data-
intensive applications upon low-cost scalable systems.
Mob Netw Appl 2013; 18: 651–663.
16. Aygün B and Gungor VC. Wireless sensor networks for
structure health monitoring: recent advances and future
research directions. Sens Rev 2011; 31: 261–276.
17. Simbeye DS, Zhao J and Yang S. Design and deployment
of wireless sensor networks for aquaculture monitoring
9
and control based on virtual instruments. Comput Elec-
tron Agric 2014; 102: 31–42.
18. Chen GZ, Zhu ZC, Zhou GB, et al. Sensor deployment
strategy for chain-type wireless underground mine sensor
network. J China Univ Min Technol 2008; 18: 561–566.
19. Philippe KK, Timmers R, Van Grieken R, et al. Photoca-
talytic disinfection and removal of emerging pollutants
from effluents of biological wastewater treatments, using
a newly developed large-scale solar simulator. Ind Eng
Chem Res 2016; 55: 2952–2958.
20. Grigalavicius M, Moan J, Dahlback A, et al. Vitamin D
and ultraviolet phototherapy in Caucasians. J Photochem
Photobiol B Biol 2015; 147: 69–74.
21. Kohtani S, Koshiko M, Kudo A, et al. Photodegradation
of 4-alkylphenols using BiVO4 photocatalyst under irra-
diation with visible light from a solar simulator. Appl
Catal B Environ 2003; 46: 573–586.
22. Oliveira HG, Ferreira LH, Bertazzoli R, et al. Remedia-
tion of 17-a-ethinylestradiol aqueous solution by photo-
catalysis and electrochemically-assisted photocatalysis
using TiO2 and TiO2/WO3 electrodes irradiated by a
solar simulator. Water Res 2015; 72: 305–314.
23. Aksoy F, Karabulut H, Cxinar C, et al. Thermal perfor-
mance of a Stirling engine powered by a solar simulator.
Appl Therm Eng 2015; 86: 161–167.
24. Kongtragool B and Wongwises S. Performance of a twin
power piston low temperature differential Stirling engine
powered by a solar simulator. Sol Energy 2007; 81: 884–895.
25. Dong X, Sun Z, Nathan GJ, et al. Time-resolved spectra
of solar simulators employing metal halide and xenon arc
lamps. Sol Energy 2015; 115: 613–620.
26. Grandi G, Ienina A and Bardhi M. Effective low-cost
hybrid LED-halogen solar simulator. IEEE Trans Ind
Appl 2014; 50: 3055–3064.
27. Yang C, Wang J, Guo X, et al. A multisource regular
dodecahedron solar simulator structure for uniform flux.
IEEE J Photovoltaics 2016; 6: 516–521.
28. Stuckelberger M, Perruche B, Bonnet-Eymard M, et al.
Class AAA LED-based solar simulator for steady-state
measurements and light soaking. IEEE J Photovoltaics
2014; 4: 1282–1287.
29. Jang SH and Shin MW. Fabrication and thermal optimi-
zation of LED solar cell simulator. Curr Appl Phys 2010;
10: S537–S539.
30. Irwan YM, Leow WZ, Irwanto M, et al. Analysis air
cooling mechanism for photovoltaic panel by solar simu-
lator. Int J Electr Comput Eng 2015; 5: 636–643.
31. Audwinto IA, Leong CS, Sopian K, et al. Temperature
dependences on various types of photovoltaic (PV) panel.
IOP Conf Ser-Mat Sci. 2015; 88: 012066.
32. Jiménez-Planas A, Martı́n J, Ábalos C, et al. Develop-
ments in polymerization lamps. Quintessence Int (Berl)
2008; 39: e74–e84.
33. Bala O, Ölmez A and Kalayci Sx. Effect of LED and halo-
gen light curing on polymerization of resin-based compo-
sites. J Oral Rehabil 2005; 32: 134–140.
34. Bazzi AM, Klein Z, Sweeney M, et al. Solid-state solar
simulator. IEEE Trans Ind Appl 2012; 48: 1195–1202.
35. Namin A, Jivacate C, Chenvidhya D, et al. Determina-
tion of solar cell electrical parameters and resistances
using color and white LED-based solar simulators with
high amplitude pulse input voltages. Renew Energy 2013;
54: 131–137.
10
36. Salam RA, Saputra C, Yuliza E, et al. Development of a
simple low-scale solar simulator and its light distribution.
In: Proceedings of the 2016 international conference on
instrumentation, control and automation (ICA), Bandung,
Indonesia, 29–31 August 2016, pp. 28–31. New York:
IEEE.
37. Sanjaya Y, Fauzi A, Edikresnha D, et al. Air tempera-
ture regulation in a chamber for rotary forcespinning. In:
Proceedings of the 2016 international seminar on sensors,
instrumentation, measurement and metrology (ISSIMM),
Malang, Indonesia, 10–11 August 2016, pp. 28–31. New
York: IEEE.
38. Sanjaya Y, Fauzi A, Edikresnha D, et al. Single phase
induction motor speed regulation using a PID controller
for rotary forcespinning apparatus. Procedia Eng 2017;
170: 404–409.
39. Lam JCW, Hui JCY and Jain PK. A dimmable high
power factor single-switch electronic ballast for compact
fluorescent lamps with incandescent phase-cut dimmers.
IEEE Trans Ind Electron 2012; 59: 1879–1888.
40. Lam JCW, Hui JCY and Jain PK. A multifunctional digi-
tal controller for a high-power-factor electronic ballast
dimmable with standard phase-cut dimmers. IEEE Trans
Ind Appl 2014; 50: 2359–2369.
Measurement and Control
41. Burkhart RM and Ostrodka DL. Reverse phase-
controlled dimmer for incandescent lighting. IEEE Trans
Ind Appl 1979; IA–15: 579–583.
42. Morris AS. Variable conversion elements. In: Morris AS
(ed.) Measurement and instrumentation principles.
Oxford; Boston, MA: Butterworth-Heinemann, 2001,
pp. 119–150.
43. Khairurrijal Abdullah M and Budiman M. Home-made
PIC 16F877 microcontroller-based temperature control
system for learning automatic control. Comput Appl Eng
Educ 2011; 19: 10–17.
44. Ogata K. Modern control engineering. Upper Saddle
River, NJ; Prentice Hall PTR, 2001.
45. Bates D and Malvino A. Electronic principles. New York:
McGraw-Hill Education, 2015.
46. Halliday D, Resnick R and Walker J. Fundamentals of
physics. Hoboken, NJ: John Wiley & Sons, 2010.
47. McCormack AS and Godfrey KR. Rule-based autotun-
ing based on frequency domain identification. IEEE
Trans Control Syst Technol 1998; 6: 43–61.
48. Suhendi A, Munir MM, Suryamas AB, et al. Control of
cone-jet geometry during electrospray by an electric cur-
rent. Adv Powder Technol 2013; 24: 532–536.