International Journal of Recent Technology and Engineering (IJRTE)

ISSN: 2277-3878, Volume-8, Issue-2, July 2019

PV Array Emulator based on Modified Hybrid

Referencing Technique with Nonlinear
Lyapunov Controller
ALAOUI Mustapha, MAKER Hattab, MOUHSEN Azeddine

 the photovoltaic modules cannot be then maintained at the

Abstract: This work proposes a new design and development of desired value, which makes its use in research and
photovoltaic (PV) array emulator which is highly requested in development less flexible especially for testing energy
laboratories and manufacturers where the need of testing the management algorithms and grid-connected photovoltaic
proper functioning of PV systems and devices such as solar
inverters that require repetitive tests at the same atmospheric
connected inverters has increased, this solar emulator can
replicate faithfully the solar characteristics without using real PV conditions[2]. Therefore, the development of equipment able
modules. The developed system is based on simplified hybrid to imitate the electrical behavior of photovoltaic modules
control strategy of referencing in order to ensure high accuracy without depending on atmospheric conditions becomes a
and speed, as well as a nonlinear Lyapunov based controller so necessity, this emulation system will then faithfully
that the closed loop of the overall system remains asymptotically reproduce the electrical behavior of a photovoltaic panel by
stable for any operating point and transition mode of the DC-DC
only implementing its characteristic in the emulator and
buck converter regardless of the load variation and weather
parameters change. choose numerically the desired atmospheric conditions of the
experiment. This innovative system will efficiently save a lot
Index Terms: Buck Converter, Control Strategy, Lyapunov of investment money in buying photovoltaic panels for
Based Controller, PV Array Emulator, V-I Characteristic laboratories and facilitate their operation and study [3].
According to the literature review, the switched mode power
supply is largely utilized in PV emulators thanks to its
I. INTRODUCTION efficiency compared to the linear regulators [3], the strategy
With the great interest given to the environment and of control is developed in such manner that the output current
sustainable development nowadays, no one can deny that and voltage of the PV emulator track continuously the V-I
conventional sources of energy have a negative impact on characteristic of the chosen PV module to emulate, it means
ecosystems and nature as well as they are exhaustible and that for any given load, the PV emulator current and voltage
non-renewable sources, hence, the use of clean and renewable should be the same as the output PV current and voltage that
energies becomes more and more crucial and unavoidable, would be generated if the load had been connected directly to
the scientific research was directed then towards the study, a real PV module.
the analysis and the modeling of these sources such as sun, The direct referencing method has been mostly utilized as a
wind, biomass , etc. besides the development of ecological control strategy for tracking the V-I characteristic, however,
energy supply systems that do not emit greenhouse gases and in spite of its simplicity and popularity, it doesn’t ensure the
contribute in reducing energy consumption and climate required performances in regards to the stability and accuracy,
change [1]. Photovoltaic solar energy is one of these especially in some zones of the V-I curve [2]. In this work,
renewable energies in which many researchers and scientists we propose a simplified hybrid control strategy with
are interested, its modeling, optimization and efficiency have nonlinear Lyapunov based controller in order to ensure good
been highly improved. However, this energy suffers from its performances and mimic accurately the PV sources.
dependence on the atmospheric parameters variation such as
temperature, irradiance as well as geographical position, II. SYSTEM DESCRIPTION
inclination, dust, shading...which directly influence its The proposed PV emulator circuit is based mainly on two
efficiency and performance [1]. As these atmospheric parts: power part which comprises a DC-DC buck converter
parameters are difficult to control, the energy delivered from to which the load is connected, this one can be either a

Revised Manuscript Received on July 09, 2019. MOUHSEN Azeddine, Laboratory of Radiation-Matter &
ALAOUI Mustapha, Interdisciplinary Laboratory of Applied Sciences, Instrumentation, Sciences and Technologies Faculty, Hassan First
National School of Applied Sciences, Hassan First University, Berrechid, University, Settat, Morocco.
Morocco.
MAKER Hattab, Interdisciplinary Laboratory of Applied Sciences,
National School of Applied Sciences, Hassan First University, Berrechid,
Morocco.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: XXXXXXX 1 & Sciences Publication
 Please Enter Title Name of Your Paper

resistance, battery, solar inverter… the output currents and Vpv+Rs∗Ipv Vpv+Rs∗Ipv
Ipv = Iph − Is [exp ( ) − 1] − (1)
Vt∗A Rp
voltage are sensed and sent to the second part, which
constitutes the control part, this unit has an essential element
so-called the reference generator, its role is to compute, using Where:
the electrical circuit model or the interpolation one, the Ipv: PV module current (A)
current reference which corresponds to the given load, Iph: photoelectric current (A)
temperature and irradiance. The Lyapunov controller is used Is: diode saturation current (A)
then to ensure the global asymptotic stability of the system Vpv: PV module voltage (V)
and allow tracking the current reference and following the V- Vt : junction thermal voltage (V)
I characteristic. The operating point of the system changes in Rs : PV module series resistance (Ω)
function of the load characteristic and weather conditions, the Rp : PV module shunt resistance (Ω)
user has to choose the desired PV module to emulate, the A: diode ideality factor.
temperature, the irradiance and the load so the controller The thermal voltage Vt is defined as:
Ns∗k∗T
computes the corresponding duty cycle which will be used by Vt = (2)
q
the Pulse Width Modulation (PWM) technique to generate
the control signal of the buck converter transistor as shown in Where
figure 1. T: the module temperature (K).
k: the Boltzmann constant [1.38 × 10−23J/K]
q: the electron charge [1.602 × 10−19 C]
Ns: Number of cells.
The photocurrent depends on both irradiance and
temperature, it is expressed as:
G
Iph = (Iscn + Ki(T − Tn)) (3)
Gn

Iscn is the nominal short-circuit current at the Standard

Test Condition (STC) (Tn = 25° C, Gn = 1000 W/m2 and air
mass 1.5) T and Tn are the actual and nominal module
temperatures (K), G and Gn are the actual and nominal
module irradiances, Ki is the temperature coefficient of the
short-circuit current (I/°C).
The diode saturation current Is depends on temperature
according to the following expression:
Iscn+Ki(T−Tn)
Is = Vocn+Kv(T−Tn) (4)
exp( )−1
A∗Vt

Where Vocn is the nominal open-circuit voltage and Kv is

the temperature coefficient of the open-circuit voltage (V/°C)
Figure 1: Overall descriptive diagram of the designed PV [4].
emulator
IV. CONTROL STRATEGY OF REFERENCING
III. PHOTOVOLTAIC SOURCE MODELLING
In order to act as an actual PV module, the designed PV
The one diode two resistances (1D2R) PV electrical model emulator has to track accurately the V-I solar characteristic
has been implemented inside the reference generator to and deliver for any operating point the same output current
deliver the current reference, this model has a good and voltage for given T and G values, hence, the main
compromise between accuracy and simplicity [4]. objective of the control strategy is to find as fast as possible
The characteristic Ipv=f(Vpv,G,T) is given by Eq (1) : the operating point which consists of the intersection point
between the chosen PV module characteristic and the load
one. Many control techniques have been used to follow the
V-I curve of PV emulators for any load and environmental
conditions [2], The direct referencing method is the
commonly used technique since it does not require any
additional algorithm to find the operating point of the system
as discussed in [2], unlike some strategies such as the Perturb
and Observe method (P&O), the Hill Climbing technique,
resistance comparison method and many other algorithms
whose speed and accuracy depend on the iterations number

Figure 2: Single diode two resistances PV model

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: XXXXXXX 2 & Sciences Publication
 International Journal of Recent Technology and Engineering (IJRTE)
ISSN: 2277-3878, Volume-8, Issue-2, July 2019

and the chosen step [2]. However, since the PV module has Update the temperature T and irradiance G values
nonlinear characteristic, it can be considered as a current
source in the constant current region where Ipv is near to Isc
while Vpv varies highly and vice versa in the constant voltage
region where the PV module is considered as a voltage source, Sense the output current and voltage (I, V) of the PV
so using solely a standard current or voltage loop may not emulator

work well for all operating points in the V-I characteristic

which affects the stability of the PV emulator especially when Evaluate the load resistance value R = V/I
the reference point changes excessively due to the
nonlinearity of the PV source, because the controller tries to
track this fluctuant reference so undesirable oscillations Estimate the maximum power point current and voltage
appear as a consequence. Some authors have proposed to (Imp, Vmp)
make the system over-damped in the constant voltage zone
for a current reference in order to avoid excessive oscillations
and improve the accuracy of the system, however, the
dynamic response becomes too late and the emulator No
Yes
performances are affected [2]. For those reasons, some
researchers have developed a new control strategy based on
hybrid referencing in such way that when the system operates
in constant current region the current reference is used with a
Calculate Vpv using PV Calculate Ipv using PV
current closed loop and current controller and vice versa at model by Newton model by Newton
the constant voltage region [5]–[7], this technique improves Raphson Method Raphson Method
the accuracy of the PV array emulator and benefits from the Vpv = f-1(I, G, T) Ipv = f (V, G, T)
advantages of direct referencing, however, it is considered
too complicated due to the use of two reference generators
based on two controllers [2]. Our contribution consists on
Update the reference Update the reference
simplifying this control strategy by using only one closed current current
loop with one controller, whether the operating point is in the Iref = Vpv/R Iref = Ipv
constant current region or voltage one, the reference is either
a current or voltage. Furthermore, in order to guaranty the
global asymptotic stability of the overall system for any
operating point and weather conditions, Lyapunov based
controller is employed. The following flow chart illustrates Yes
clearly the main idea of the proposed control strategy. T and G
changed?
It is important to note that for avoiding excessive switching
around the switch point which has been chosen as Rmp =
Vmp/Imp caused by measurement errors and ripples, an
overlap can be set up between the two regions or a hysteresis No
can be added to overcome this problem in practice as
discussed in [5]–[7]. Yes
In addition, some authors proposed to divide the V-I curve
into three regions rather than two considering the maximum R changed?
power point region in the aim of ensuring further accuracy
[5]–[7], however, it complicates more the control strategy [2]. No

Deliver the reference current Iref to the Lyapunov

Controller

Figure 3: flow chart of the proposed referencing control

strategy for PV emulator

V. NONLINEAR LYAPUNOV CONTROLLER DESIGN

Using Kirchhoff’s current and voltage laws, the model of
the DC-DC buck converter can be obtained as below:
diL
L + vc = Vdc ∗ u
dt
{ dvc vc
(5)
C∗ + = iL
dt R

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: XXXXXXX 3 & Sciences Publication
 Please Enter Title Name of Your Paper

Where iL and vc are respectively the inductance current

and the capacitance voltage of the power converter, Vdc is
the input supply value and u is the duty cycle of the buck
converter.
The state space model can be written as:

−x2 Vdc
x1̇ = + u
L L
{ x1 x2 (6)
x2̇ = − Figure 4 : DC-DC Buck Converter
C RC
Vdc
L≥ (13)
4∆Ipp∗f
Where x1 = iL and x2 = vc ∆Ipp
Let’s consider the following Lyapunov function: C ≥ (14)
8f∗∆Vpp
1 1
σ = ( 𝐿𝑖𝐿 2 − 𝐿𝑖𝑟𝑒𝑓 2 )2 (7) Where: ∆Ipp and ∆Vpp are respectively peak-to-peak
2 2 ripple current and voltage and f is the switching frequency of
the buck converter. The details of the buck converter
Where iref is the current reference delivers from the parameters values are given in section VII-A.
reference generator.
VII. SIMULATION RESULTS
σ(x0) = σ(iref, Iref ∗ R) = 0
In this section, simulation results using Matlab Simulink
σ(𝑥) > 0 ( x ≠ 0)
software are presented and discussed, the main target of this
σ̇ = 𝐿2 (𝑖𝐿 2 − 𝑖𝑟𝑒𝑓 2 )𝑖𝐿 ∗ 𝑖𝐿̇ (8)
study is to show the ability of the proposed system to mimic
the electrical behavior of the PV module by comparing its
Choosing 𝑖𝐿̇ to be as: output current and voltage with the PV modules ones for
𝑖𝐿̇ = −k(𝑖𝐿 2 − 𝑖𝑟𝑒𝑓 2 ) (k>0) (9) different operating points, the dynamic transient response is
analyzed too. Besides, the study of the developed PV
So σ̇ becomes: emulator subjected to the variation of ambient temperature
σ̇ = −𝑘𝐿2 (𝑖𝐿 2 − 𝑖𝑟𝑒𝑓 2 )2 𝑖𝐿 (10) and cell irradiance is given, Moreover, we have evaluated the
output current and voltage of the proposed circuit in the
Therefore σ̇ < 0 for any k positive and (x ≠ 0) steady state with different resistance values and compare
So the closed loop of the system will be asymptotically them with the theoretical PV modules outputs for the purpose
stable for any value of the resistance R, the accuracy and of judging the effectiveness and accuracy of the proposed
speed can be adjusted through one parameter k. emulator.
Note that 𝑖𝐿 is always positive in continuous current mode
A. Simulation Model Parameters
and null if the buck converter operates in discontinued current
The parameters of the chosen PV module to emulate have
mode and cannot be negative for the buck converter, so σ̇ is
always negative and null at steady state. been taken from ALLMAX PLUS datasheet and are shown
in table III.
Let’s find the expression of u which forces 𝑖𝐿̇ to be equal
In addition, the parameters of the DC-DC buck converter
to – k(𝑖𝐿 2 − 𝑖𝑟𝑒𝑓 2 )
and the controller are given in table I and II.
Substituting the expression of 𝑥1̇ from the equation (6) in It has to be noted that the parameters Rs, Rp and A have
(9), we find: been extracted according to the algorithm described in [1].
−vc Vdc
+ u = −k(𝑖𝐿 2 − 𝑖𝑟𝑒𝑓 2 ) (11)
L L
Table 1. DC-DC Buck Converter Parameters
Finally, the duty cycle expression will be written as: Input voltage Vdc 60 V
k∗L(𝑖𝑟𝑒𝑓 2 − 𝑖𝐿 2 )+𝑣𝑐
𝑢= (12) Capacitor C 10 µF
𝑉𝑑𝑐
Inductor L 2 mH
k is used to adjust the response speed of the system, the
Switching frequency f 100 KHz
more k is high, the more the system is fast and accurate,
however, overshoot can be observed if we excessively
increase k. Table 2. Lyapunov Based Controller Parameter

VI. DC-DC BUCK CONVERTER

K 100/L
The DC-DC buck converter as illustrated in figure 4 has
been selected as a power converter topology of the proposed
PV emulator, the values of inductance L and capacitance C
are computed based on the equations (13) and (14) in order to
guaranty a little amount of current and voltage ripples, which
should not exceed 5 % and to operate in the continuous
current mode [8].

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: XXXXXXX 4 & Sciences Publication
 International Journal of Recent Technology and Engineering (IJRTE)
ISSN: 2277-3878, Volume-8, Issue-2, July 2019

Table 3. PV Model parameters under STC conditions

Peak rated power 300 W
Open circuit voltage Voc 39.8 V
Short circuit current Isc 9.77 A
Maximum power point 32.6 V
voltage Vmpp
Maximum power point 9.19 A
current Impp
Temperature coefficient of 0.05
Isc : ki
Temperature coefficient of -0.29
Voc : kv
Temperature coefficient of -0.39
power:kp
Cell number Ns 60
Series resistance Rs 0.1906 Ω
Shunt resistance Rp 3785.50 Ω
Figure 6: The output current and voltage of the proposed
fidelity factor A 1.2452 PV emulator with PV model for R = Rmp at STC
conditions

B. Dynamic Study

1) PV Emulator with Resistive Load

As shown in the following figures 5, 6 and 7, three
operating points from the V-I curve are shown, one is located
in the constant current region which corresponds to 0.5 Ω as
a resistance load value, and the other is equal to the maximum
power point whose resistance value is Rmp = 3.547Ω, then
the third point is selected to be in the constant voltage zone
(R = 15 Ω). According to the simulation results shown below,
the proposed PV emulator generates the same load current
and voltage as the PV module.

Figure 7 : The output current and voltage of the proposed

PV emulator with PV model for R = 15 Ω at STC conditions

It is worth noting that the developed PV emulator presents

good performances in terms of speed, accuracy and overshoot
limitation, the voltage and current ripples are too little,
therefore, the system performances remain as required for any
operating point and the settling time can be reduced too by
increasing the parameter k.

Figure 5: The output current and voltage of the proposed

PV emulator with PV model for R = 0.5 Ω at STC
conditions

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: XXXXXXX 5 & Sciences Publication
 Please Enter Title Name of Your Paper

2) PV Emulator under Environmental Conditions

Change
The proposed PV array emulator can also replicate the
behavior of the real PV module under the variation of the
atmospheric parameters T and G. Indeed, we have chosen a
random profile of temperature and irradiance and we have
compared the PV module outputs with the designed emulator
ones, it has been proven according to the figure 8 that
effectively the designed PV source emulator can follow
rapidly the PV module outputs with a good level of accuracy.

Figure 9: Comparison between V-I curve characteristic of

PV model and PV emulator under STC conditions.

VIII. CONCLUSION
In this paper, a new PV array emulator has been presented,
the main purpose of this novel device is to imitate accurately
the electrical behavior of the real PV module regardless of the
variation of weather conditions. A new control strategy based
on modified hybrid referencing and nonlinear Lyapunov
controller has been proposed to ensure the asymptotic
stability of the full system as well as increase the accuracy
and the transient dynamic response of the converter. It has
Figure 8: PV Emulator under temperature and irradiance been effectively shown that the developed PV source
variation emulator tracks faithfully the V-I characteristic of the PV
source and generates similar current and voltage outputs as
C. Steady State Study
an actual PV model under varying the load and the
In this part, the steady state analysis is done, various environmental condition. Further work will be dedicated to
resistance load have been utilized in order to sweep different the experimental validation of the proposed PV source
operating points in the V-I characteristic, it is obviously emulator.
shown from figure 9 that the proposed PV array emulator can
produce the same outputs as the PV module and track REFERENCES
accurately the whole V-I characteristic, so it can be efficiently
1. M. C. Di Piazza and G. Vitale, Photovoltaic sources: modeling and
used for testing MPPT algorithms and evaluating the
emulation. London: Springer, 2013.
performances of many other photovoltaic systems. 2. R. Ayop and C. W. Tan, “A comprehensive review on photovoltaic
emulator,” Renewable and Sustainable Energy Reviews, vol. 80, pp.
430–452, Dec. 2017.
3. J. P. Ram, H. Manghani, D. S. Pillai, T. S. Babu, M. Miyatake, and N.
Rajasekar, “Analysis on solar PV emulators: A review,” Renewable
and Sustainable Energy Reviews, vol. 81, pp. 149–160, Jan. 2018.
4. R. Ayop and C. W. Tan, “A comparison study of interpolation and
circuit based photovoltaic mathematical models,” in 2016 IEEE
International Conference on Power and Energy (PECon), Melaka,
Malaysia, 2016, pp. 626–631.
5. Y. Kim, W. Lee, M. Pedram, and N. Chang, “Dual-mode power
regulator for photovoltaic module emulation,” Applied Energy, vol.
101, pp. 730–739, Jan. 2013.
6. Y. Li, T. Lee, F. Z. Peng, and D. Liu, “A Hybrid Control Strategy for
Photovoltaic Simulator,” p. 5.
7. T. D. Mai, S. De Breucker, K. Baert, and J. Driesen, “Reconfigurable
emulator for photovoltaic modules under static partial shading
conditions,” Solar Energy, vol. 141, pp. 256–265, Jan. 2017.
8. K. H. Billings and T. Morey, Switchmode power supply handbook.
New York: McGraw-Hill, 2011.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: XXXXXXX 6 & Sciences Publication