This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

Global Maximum Power Point Tracking

Method for Photovoltaic Arrays Under
Partial Shading Conditions
Alireza Ramyar, Hossein Iman-Eini, Member, IEEE,
and Shahrokh Farhangi, Member, IEEE

 maximum achievable power. Many maximum power point

Abstract—The power-voltage characteristic of tracking (MPPT) methods have been proposed in the literature
photovoltaic (PV) arrays displays multiple local maximum [1]-[3]. Popular MPPT methods like Perturbation and
power points when all the modules do not receive uniform Observation (P&O), Hill climbing (HC), and Incremental
solar irradiance, i.e. under partial shading conditions Conductance (IC) methods are shown to be effective when the
(PSCs). Conventional maximum power point tracking solar irradiance condition is uniform for all PV modules.
(MPPT) methods are shown to be effective under uniform
solar irradiance conditions. However, they may fail to
Since, the tracking becomes more complicated under partial
track the global peak under PSCs. This paper proposes a shading conditions (PSCs), i.e. when all the modules do not
new method for MPPT of PV arrays under both PSCs and receive uniform solar irradiance, these basic methods fail to
uniform conditions. By analyzing the solar irradiance track the GP. Though in uniform solar irradiance conditions
pattern and using the popular Hill Climbing method, the the P-V characteristic of PV array has just one peak, the P-V
proposed method tracks all local maximum power points. characteristic of PV array displays multiple peaks under PSCs.
The performance of the proposed method is evaluated Hence, several MPPT methods are proposed which are
through simulations in MATLAB/SIMULINK environment. applicable in PSCs. These methods can be categorized into
Besides, the accuracy of the proposed method is proved two groups: hardware-based methods and software-based
using experimental results.
methods [4].
Index Terms—Current-voltage (I-V) characteristic,
In [5] and [6], a controller is assigned for each module.
maximum power point tracking (MPPT), partial shading These hardware-based methods can resolve the problem, since
condition (PSC), photovoltaic (PV) array, power-voltage (P- the P-V characteristic of a module (with just one bypass
V) characteristic. diode) has always a single peak. These methods, however, are
not cost-effective and require much more devices in
comparison to software-based algorithms.
Authors of [7] have proposed an effective MPPT method
I. INTRODUCTION for PV systems based on particle swarm optimization (PSO)
algorithm. This method is too complex to be applied to the
N OWADAYS, solar energy as a clean and free available
renewable energy resource is too important for reducing
the dependency on conventional sources. Photovoltaic (PV)
commercial appliances, since some parameters have to be set
by the user. In [8], artificial neural network (ANN) algorithm
has been opted. The main problem of ANN-based methods is
systems produce electric power by directly transforming the
that the ANN’s accuracy under different conditions is highly
inexhaustible solar energy into electricity. However, the dependent to the amount of available training data. In addition,
relatively high cost, low conversion efficiency of electric they need to be retrained when the PV array is changed. In [9]-
power generation, dependency on environmental conditions 12] genetic algorithm, flashing fireflies, artificial bee colony,
(e.g., solar irradiance and temperature), and nonlinearity of the and Simulated Annealing are used in PV applications,
power-voltage (P-V) and current-voltage (I-V) characteristic respectively. These methods have good performances but
of PV arrays are the main challenges in utilization of PV similar to the aforementioned issues for the PSO and ANN
arrays. methods, the implementation complexity of these methods is
Tracking the Global Peak (GP) of a PV array in all their major problem; since they involve complex calculations
conditions is significantly important to guarantee the and several parameters have to be set by user.
In [4], the HC method has been improved. It can efficiently
Manuscript received July 1, 2016; revised September 20, 2016 and detect the shading condition. Then, by measuring power in
October 9, 2016; accepted October 11, 2016. This work was supported suitable points, it chooses the highest one and performs the
by the Iran National Science Foundation.
A. Ramyar, H. Iman-Eini, and S. Farhangi are with the School of HC around this point. However, it does not have an acceptable
Electrical and Computer Engineering, College of Engineering, accuracy for tracking the GP, since it compares the power of
University of Tehran, Tehran, Iran (e-mail: alireza.ramyar@ut.ac.ir; points near the LPs instead of the LPs themselves. In [13], a
imaneini@ut.ac.ir; farhangi@ut.ac.ir). modified P&O method has been introduced which benefits

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

from a unique characteristic that has been observed in the P-V akTN s ,m  I pv ,array  I   N s 
curves. Although it has a great performance, since almost two V  Ns( ) ln     Rs I (1)
measurements are done for each LP, the tracking speed is low. q  I 0,array   N p 
In [14], it is claimed that the GP is around the intersection of where V and I are the output voltage and current of PV array,
the I-V characteristic of PV arrays and a certain line. It respectively. IPV,array is the output current of PV array. I0,array is
depends on short circuit current of array which is problematic the equivalent saturation current. q is the electron charge
[1]. This problem is almost resolved by updating this value (1.60217646 × 10-19 C), k is the Boltzman constant
based on the solar irradiance. However, it is uncommon to (1.3806503 × 10-23 J/K), T is the junction temperature in
find sensors that measure solar irradiance levels [1]. In [15], a Kelvin, and a is the diode ideality constant. Rs is the PV
relationship is defined between the PV power and a control module's series resistance.
signal to track the P-V curve and find the GP. Although its For IPV,array and I0,array we can have [4]:
accuracy is high, it is slow because it searches almost all the I sc ,m
range of the P-V curve. [16] uses the critical observations
reported in [13] in a different way, but it does not have any I pv ,array  N p  I scn ,m  K I T  GG (2)
procedure for detecting whether there is an LP near the target n

point or not. As a result, it may fail in some PSCs. Moreover, and

the approach in [16] involves complex calculations (e.g.
I 0,array  N p
 I scn ,m  K I T 
calculation of square root) compared to similar methods; q (V ocn ,m  KV T ) (3)
hence, it is not as simple as other similar methods for exp[ ] 1
experimental implementation. By choosing lower and upper akTN s ,m
voltage limits, [17] narrows the searching window and tracks where Iscn,m is the short circuit current of the module in
the GP very fast. On the other hand, authors admit that the standard test condition (STC), Isc,m is the short circuit of
method may fail when two LPs have nearly equal power module in real condition, KI is the current coefficient, G is the
values. [18] maps out the solar irradiance pattern based on the solar irradiance level (W/m2), and Gn is the nominal solar
voltage of modules and chooses an appropriate voltage to irradiance level (1000 W/m2). ∆T is the temperature difference
track the GP around it. Obviously, employment of one voltage to temperature of STC. Vocn,m is the module open circuit
sensor for each module is not feasible and cost effective. In voltage in STC and KV is the voltage coefficient. Since:
[19], two methods are proposed. The first one searches the P- Q(V ocn ,m  K I T )
V curve for MPPs by means of IC. However, it skips parts of exp[ ]  1, (4)
akTN s ,m
the area based on short circuit current of the modules and the
highest local power. This method would be very slow since it (3) can be simplified as:
must scan almost all the P-V curve. Although the second
I 0,array  N p
 I scn ,m  K I T  .
method has improved the speed of tracking compared to the Q(V ocn ,m  K I T ) (5)
first one, it still uses one current sensor for each bypass diode, exp[ ]
which is not cost effective. akTN s ,m
[20] applies ramp voltage command to the converter. Substituting (2) and (5) into (1) yields:
Therefore, it avoids the oscillation of voltage and current of akTN s ,m G I 
the system in transient intervals. Hence, long delays in usual V  Ns( ) ln   
methods for correct sampling of voltage and current are not q  n
G N p scn ,m  K I T
(I ) 
needed any more. However, it searches almost all the range of V oc ,m
(6)
 Ns 
P-V curve and therefore, its tracking speed is not good.  N s (V ocn ,m  KV T )  R I
Np  s
Proposing a method which meets accuracy, convergence  
speed, simplicity, minimum needed parameters, minimum cost where Voc,m is module’s open circuit voltage in real condition.
and other important factors [1] at the same time is still of a Having Ipv,array ≈ Isc,array [21,23], (2) can be rewritten as:
great importance. In this paper, we propose a novel method for I sc ,m
MPPT of PV arrays which works effectively in PSCs and at G
the same time, has great performance in diverse factors I sc ,array  N p (Iscn ,m  K I T )  (7)
mentioned above. By measuring PV current in defined points, Gn
the method maps out the solar irradiance pattern. Based on the where Isc,array is the short circuit current of array.
mapping, it chooses appropriate points for tracking the LPs.
B. I-V Characteristic of PV Arrays under PSCs
Then, it performs HC in these points and tracks all the LPs.
Finally, by comparison of the acquired LPs, it chooses the GP. Fig. 1 shows the I-V characteristic of a sample 3×2 array
under different PSCs. The modules’ parameters are listed in
II. I-V CHARACTERISTIC OF PV ARRAYS UNDER PSCS Table I. The modules are modeled based on single diode
model in [22] and the equivalent parameters of PV modules
A. Single Diode Model are listed in Table II. Rp is the PV module's parallel resistance.
Based on the single diode model of the PV cells, if Ns As illustrated in Fig. 1, the value of the current in each step
modules (each of which consists of Ns,m series cells) are is almost constant up to the end of that step. Keeping this point
serried, and Np strings are paralleled, the voltage equation of in mind, by measuring the PV current in specific points and
the array would be as follows [21]: comparing them in a suitable manner which are presented in

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

TABLE I
10
PV MODULE’S PARAMETERS

Parameter Value
8
PMPP 35 W
Voc,n 11.15 V
6 Isc,n 4.15 A
Current (A)

Ns,m 18

TABLE II
4
EQUIVALENT PARAMETERS OF PV MODULE IN SINGLE DIODE MODEL [22] IN
STC
2 Parameter Value
0.5*Vocm 1*Vocm 2*Vocm
=5.575 V =11.15 V =22.3 V a 1.077
0 Rs 0.175 Ω
0 5 10 15 20 25 30 35 Rp 123 Ω
Voltage (V)
Fig. 1. I-V characteristic of a sample 3×2 array under different PSCs
when the module open circuit voltage and short circuit current are 1) Group 1 Analysis
based on Table I.
The value of the voltage in the starting point of the second
Subassembly 1: Subassembly 2: step of group 1 can be derived from the voltage of
1000 W/m2 500 W/m2
subassembly 1 and 2 in this point:
V beginning ,St 2G 1 V Sub1G 1 V Sub 2G 1 (8)
Group 1
where Vbeginning,St2G1 stands for the voltage of the starting point
Group 2 of the second step in group 1. VSub1G1 and VSub2G1 are the
voltages of subassembly 1 and 2 in this point, respectively. It
should be noticed that in this point, the bypass diode of the
Subassembly 1: Subassembly 2:
900 W/m2 200 W/m2 module in subassembly 2 is still on and is going to be off.
Fig. 2. A sample 3×2 PV array under PSC. Therefore, the voltage of subassembly 2 in this point derives
from the bypass diode’s forward voltage, which is 0.8 V in
8 this test:
Array Curve V Sub 2G 1   0.8 V . (9)
6 Group 1 Curve Using (7), the short circuit current of the subassembly 2 can
Current (A)

be determined as following equation. In this case G=500,

4 Group 2 Curve ∆T=0 and Np=1:
I sc ,Sub2G1  2.075 A (10)
2 where Isc,Sub2G1 stands for open circuit voltage of the
subassembly 1 of group 1. Using (6), the corresponding
0 voltage of the subassembly 1 in the start point of the second
0 10 20 30 40 step can be calculated. In this case G=1000, ∆T=0, T=298.15
Voltage(V) K, Ns=2, Np=1, and Vocn,m=11.15 V. Also, the value of I in
Fig. 3. I-V characteristic of group 1, group 2 and total array in Fig.2. equation (6) equals to 2.075 A, since the starting point of the
second step is the end of the first step. So one can write:
Section III, the PSC pattern can be mapped. As a result, the V Sub1G 1  20.883 V . (11)
number of steps, their lengths, and their order in the I-V
characteristic can be detected. In addition, as it is depicted in Substituting (9) and (11) into (8) yields:
Fig. 1, voltage values in the starting points of current steps, are V beginning ,St 2G 1  20.083 V . (12)
in near left side neighborhood of certain integer multiples of As it was claimed, the starting point of the second step
Voc,m . In order to justify above claim, a sample PV array (20.083 V) is in near left side neighborhood of certain integer
consists of 2 strings each of which includes 3 series modules, multiple of Voc,m which is 2× Voc,m (22.3V) in this case.
is considered as shown in Fig. 2. The modules’ parameters are
given in Tables I and II. Since the test is executed under STC, 2) Group 2 Analysis
∆T equals to zero and Voc,m equals to Vocn,m which is 11.15 V. The analysis of this group is similar to analysis of group 1.
Also, Fig. 3 demonstrates the I-V characteristic of each group So, the value of the voltage in starting point of the second step
as well as the total characteristic of the PV array. of group 2 can be derived from the voltage of subassembly 1
and 2 in this point:
V beginning ,St 2G 2 V Sub1G 2 V Sub 2G 2 (13)

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

where Vbeginning,St2G2 stands for the voltage of the starting point Global MPPT Soubroutine
of the second step in group 2. VS1G2 and VS2G2 are the voltages Main Program
Start
of subassembly 1 and 2 in this point, respectively. In this point Start

the bypass diodes of the modules in subassembly 2 are still on Measure array current in
Perform HC integer multiples of Voc,m
and are going to be off. Then, the voltage of subassembly 2 in B
this point derives from the bypass diode’s forward voltage, Calculate ∆P for
Analyze the
insolation pattern
which is 0.8 V in this test: No each two

V Sub 2G 2  1.6 V .
consecutive
(14)
A cycles Based on the analysis ,choose appropriate
n for Vref = n× 8 .0× Voc,m to determine
The short circuit of the subassembly 2 can be calculated IS ∆P > ∆Pcrit? the farthest LP in the left side of the P-V
using (7). In this case G=200, ∆T=0 and Np=1: curve

I sc ,Sub2G 2  0.83 A (15) Yes Perform HC around

Vref and Store the
Call global information
where Isc,Sub2G2 stands for open circuit voltage of the MPPT i .e. Vmax,last and Pmax,last
subassembly 1 of group 2. Again by usage of (6), the Sobroutine

corresponding voltage of the subassembly 1 in the starting Has the algorithm Perform HC
recognized around Vmax,last
point of the second step can be calculated. In this case G=900, No
existence of another and call Main
∆T=0, T=298.15 K, Ns=1, Np=1 and Vocn,m=11.15 V. Also, the peak? Program

value of I in equation (6) equals to 0.83 A. So: Yes

V Sub1G 2  10.827 V . (16) Based on the analysis, choose
nest appropriate n for
Substituting (14) and (16) into (13) yields: Vref = n×8 .0× Voc,m
C
V beginning ,St 2G 2  9.227 V . (17) No
Perform HC around
Similar to group 1, the starting point of the second step Vref and Store the
information
(9.227 V) in group2 is in near left side neighborhood of i .e. Vmax,new and Pmax,new
certain integer multiple of Voc,m which is 1× Voc,m (11.15V) in
Is Pmax,new >
this case. Pmax,last ?

3) Array Analysis Yes

Update Vmax,last
Since the curves of each group consist of steps in which the and Pmax,last :
values of current are almost constant until the next step, the Vmax,last=Vmax,new
Pmax,last=Pmax,new
summation value would also have this characteristic. On the
other hand, since the starting points of the total curve are the Fig. 4. Flowchart of the proposed method.
starting points of the steps in group 1 and 2, the value of the
voltage for each start point is in near left side neighborhood of B. Analysis of the Solar Irradiance Pattern
a certain multiple of Voc,m. The proposed analysis is valid for
In section B, the method measures the current in the
every structure.
multiples of Voc,m . It was proved in Section II-B that the
starting point of the current steps in the I-V characteristic are
III. PROPOSED METHOD FOR MPPT
in the near left side neighborhood of certain integer multiples
Fig. 4 shows the flowchart of the proposed method. In of Voc,m. In addition, the value of the current for each step is
steady state conditions, the HC method performs around the nearly constant up to the next step. Therefore, by comparing
last GP which has been detected by the proposed method. the measured currents against each other, the number of steps,
A. Detecting the Solar Irradiance Changes their lengths, and their order in the I-V characteristic can be
easily determined by the following procedure.
In order to recognize the sudden changes of the solar If IVk is the measured current in K×Voc,m and IV(k-1) is the
irradiance condition, the power difference between each two measured current in (K-1) ×Voc,m ,the proposed method checks
consecutive cycles (ΔP) is calculated and compared against a the validity of the following inequality:
certain critical power variation (ΔPcrit) as illustrated in Fig. 4.
IV ( k 1)  IVk
If ΔP is higher than ΔPcrit, variation of the solar irradiance  I crit (18)
condition is detected and the global maximum power point IV ( k 1)
tracking starts. Generally, the sudden changes in solar If this inequality is satisfied, the proposed method
irradiance are small in magnitude (smaller than 27 W/m2) recognizes that there is no new step in the neighborhood of
[13]. So, ΔPcrit can be equal to the change in output power of K×Voc,m ; otherwise the new step is detected by the method. It
array, for the condition that the solar irradiance changes by 27 should be noticed that if the steps were ideal, i.e. the I-V
W/m2 [13]. Or, it might be set to an appropriate percentage of characteristic was constructed from rectangular sections, ∆Icrit
array nominal power [13]. In this paper, this threshold is set to must be zero. However, the steps are not ideal. Since the
5% of the nominal power, based on trial and error observation current source part of the I-V characteristic of the PV array
from simulation. Once the solar irradiance change is detected, under uniform conditions continues to maximum power point
the sections B and C in Fig. 4 start to track the new GP as (in which current is about 0.9× Isc [13], [20]), an appropriate
explained in the following. value for ∆Icrit is (Isc-0.9×Isc)/Isc which equals to 0.1.

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

10 module to map out the PSC pattern, the new method maps out
P3 P2 the PSC pattern with just one current sensor.
8
C. Searching for Maximum Power Points
P1
Current (A)

6 Based on [13], the LPs are in neighborhood of integer

multiples of 0.8×Voc,m. So according to the analyzed solar
4
L1 L2 irradiance pattern, the method allocates a certain multiple of
2 0.8×Voc,m , to each LP. Thus by operation of the HC method in
0.5*Vocm 1*Vocm 2*Vocm its neighborhood, the corresponding LP is tracked. Finally, by
=5.575 V =11.15 V =22.3 V
0 comparison among the LPs, the GP is determined.
0 5 10 15 20 25 30 35
Whether each LP is tracked or not is recognized by
Voltage (V)
(a)
checking the slope of the P-V curve. After determining the
largest LP as the GP, the HC is performed around it. When the
200
P7 GP is tracked, the duty cycle is fixed to prevent the
P5 P6 perturbations, as discussed in [15]. However, it is not
150 P4 necessarily an issue for the proposed method. Besides, a
Power (W)

variable step HC can be used to decrease the perturbations

100 around the GP.
To better understand the above procedure, the previous
50 example is considered for finding the GP. After obtaining the
2*0.8*Vocm 3*0.8*Vocm PSC pattern, the HC is performed around P 4(17.84V) in which
=17.84 V =26.76 V voltage equals to L1×0.8×Voc,m(2×0.8×Voc,m), and tracks
0
0 5 10 15 20 25 30 35 P5(138.9 W), as it is illustrated in Fig. 5(b). Then, based on the
Voltage(V) obtained PSC pattern, the algorithm goes to the neighborhoods
(b) of the other LP, i.e. P6(26.76V) in which voltage equals to
Fig. 5. (a) I-V characteristic, and (b) P-V characteristic of a sample
5×5 array.
(L1+L2)×0.8×Voc,m(3×0.8×Voc,m). The HC method is
performed around this point and the other LPs are tracked,
Generally, lower values of ∆Icrit lead to higher accuracy, but it which are P7(720 W). Finally, by comparison among these
increases the time required to track the GP. Since the boost LPs, the GP which is P6 (159.3 W) is determined.
converter is used, it is better to keep distance from zero Since the proposed method is a search-based tracking
voltage point and measure the current of 0.5×Voc,m instead of algorithm and just initiates the HC method in the
the current of zero voltage point. However, measuring the neighborhood of 0.8×Voc,m , it does not really depend on the
current in a small voltage makes the boost converter to work open circuit voltage. But the voltage can be updated every 10
in a relatively large duty cycle and it is a drawback. minutes by the following equation [4]:
It is helpful to describe this procedure in a sample case. The V oc ,m V ocn ,m  KV T (21)
corresponding I-V and P-V curves of a 3×2 PV array whose
parameters are listed in Tables I and II, under a sample PSC and after each update, the proposed method can be performed.
are shown in Fig. 5(a) and Fig. 5(b), respectively. For As described, the proposed algorithm has modified the
analyzing the PSC pattern, the current of PV array is measured conventional HC method to work properly in PSCs. Therefore,
in three points (2×Voc,m, 1×Voc,m and 0.5×Voc,m). As it is it is simple for experimental implementation. Also, as
depicted in Fig. 5(a), these 3 points are P1(22.3 V, 6.13 A), presented in subsequent sections, the accuracy and
P2(11.15 V, 8.2 A) and P3(5.575 V, 8.24 A), respectively. convergence speed of the proposed method are better than
(18) is checked for corresponding currents as follows: existing methods.
8.24  8.2
 0.005  0.1 (19) IV. SIMULATION RESULTS
8.24
In this section several simulation results will be presented.
8.2  6.13
 0.25  0.1 (20) The simulated PV system is a 3×2 PV array, whose parameters
8.2 are listed in Tables I and Table II.
Hence, based on the described procedure, the algorithm The PV array is connected to a boost DC-DC converter
recognizes that there are two steps in the I-V curve, with the which tracks the maximum power point. There are 3 series
lengths of L1 (2×Voc,m) and L2 (1×Voc,m). Actually, (18) is true connected 12-V batteries in the output side. The parameters of
in the case that k=1 which means that P3 and P2 are in the system under study are listed in Table III. Also the schematic
same step. Since (18) is not satisfied in the case that K=2, it of the system is shown in Fig. 6.
means that P1 and P2 are not in the same steps. Thus, as it was During adoption of HC method, the duty cycle of the boost
claimed, the proposed method detects the number of steps, converter is calculated directly based on the last duty cycle as
their lengths, and their order in the I-V characteristic with just follows:
measuring the array current. This new idea is very simple and Dk  Dk 1  d k (22)
yet so useful for MPPT in PSCs. It should be mentioned that, where Dk and Dk-1 are the calculated duty cycles in previous
although methods like [18] use one voltage sensor for each (i.e. (K-1)-th) and present (i.e. K-th) cycles, respectively. dk is
a number which it’s value is constant, but it’s sign may change

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

TABLE III
SYSTEM’S PARAMETERS 1000 500 500
1000 1000 1000
W/m2 W/m2 W/m2 W/m2 W/m2 W/m2

Parameter Value

L 0.6 mH 1000 500 500 1000 1000 500

Cin 34 µF W/m2 W/m2 W/m2 W/m2 W/m2 W/m2
Cout 48 µF (a) (b)
Switching Frequency 40 KHz Fig. 7. PSC patterns in the first simulation, a) from 0.3 s to 0.6 s, b)
Sampling Frequency 1 KHz from 0.6 s to 0.9 s.
Array Nominal Open Circuit Voltage 33.45 V
Array Nominal Short Circuit Current 8.3 A
10
Array Nominal Power 210 W Curve1

Current (A)
L D Curve3
5
Switch Curve2
array

Cout
Cin
PV

0
PWM

0 10 20 30 40

Battery
Voltage (V)
VPV MPPT: (a)
duty cycle calculation
IPV Vout
Curve1
Fig. 6. Schematic of the system. 200 Curve3

Power (W)
in each cycle. If the value of the power measured in the K-th Curve2
cycle is larger than the value of power measured in (K-1)-th 100
cycle, dk is calculated as:
d k  d k 1 (23)
0
On the other hand, if the power in the K-th cycle is smaller 0 10 20 30 40
than the power in (K-1)-th cycle, dk is calculated as: Voltage (V)
d k   d k 1 (24) (b)
Fig. 8. Corresponding (a) I-V and (b) P-V characteristics under first
Also, when a reference voltage (Vref) is chosen in global
simulation.
maximum power point tracking subroutine, the duty cycle (D∗)
is generated as follows [13]: 3×0.8×Voc,m. The algorithm performs HC and tracks two LPs
V with 59.5W and 107 W power which are very close to the
D *  1  ref (25) peaks of curve 2 in Fig. 8(b), i.e. 63W and 108W. So the
V out
proposed method chooses the biggest LP and continues to
A. Performance Exploration under Four Consecutive work around 107 W.
Solar Irradiance Condition Fig. 10(b) shows that, when the solar irradiance changes
again at 0.6s, the proposed method starts to measures current
In this section, the performance of the algorithm is tested
in 2×Voc,m, 1×Voc,m and 0.5×Voc,m. In this case, second and
under four consecutive solar irradiance conditions. From 0 to
third measured currents are very close and the first one differs
0.3s, the solar irradiance level is equal to 1000 W/m2 for all
from these currents. Thus, the method recognizes that there are
the modules. From 0.3s to 0.6s and 0.6s to 0.9s, the solar
two LPs near 2×0.8×Voc,m and 3×0.8×Voc,m. The algorithm
irradiances are shown in Fig. 7(a) and Fig. 7(b), respectively.
performs HC and tracks two LPs with 139 and 158 W power
Finally, from 0.9s to 1.2s the solar irradiance is equal to 1000
which are very close to the peaks of curve 3 in Fig. 8(b), i.e.
W/m2 for all the modules again. The I-V and P-V curves of
139W and 159 W. Therefore, the proposed method chooses
the PV array in these four states are shown in Fig. 8. Array’s
the biggest LP and continues to work around 159 W.
corresponding voltage, current, power and duty cycle
As it is shown in Fig. 10(c), when the PSC is removed at
waveforms are shown in Fig. 9(a), Fig. 9(b), Fig. 9(c) and Fig.
0.9s, the proposed method starts to measures the current in
9(d), respectively. Moreover, zoomed view of per unit array’s
2×Voc,m, 1×Voc,m and 0.5×Voc,m. In this case, all measured
voltage, current, power and duty cycle waveforms in 0.3s to
currents are very close. Hence, the method recognizes that
0.5s, 0.6s to 0.8s and 0.9s to 1.1s intervals are depicted in Fig.
there is just one LP near 3×0.8×Voc,m. The algorithm operates
10(a), Fig. 10(b) and Fig. 10(c), respectively.
HC and tracks the LP with 208 W power which is very close
As illustrated in Fig. 9(c), the algorithm operates properly
to the peak of curve 1 in Fig. 8(b). So the proposed method
under normal conditions and the GP is equal to 208 W which
continues to work around 208 W.
is so close to the peak of curve 1 in Fig. 8(b). It is illustrated in
Fig. 10(a) that when the solar irradiance level changes at 0.3s, B. Comparison of the New Method against Other
the proposed method measures current in 2×Voc,m, 1×Voc,m and Methods
0.5×Voc,m. Since first and second measured currents are very As it was mentioned before, although a large amount of
close and the third differs from these currents, the method studies are presented in this field, proposing a method which
recognizes that there are two LPs near 1×0.8×Voc,m and meets accuracy, convergence speed, simplicity, minimum

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

Voltage (V) 40 1

0.8 Voltage
20 Operating
0.6 HC Pow er
Measuring
0 0.4 Currents
0 0.2 0.4 0.6 0.8 1 1.2 Current
Time (s)
0.2 Duty cycle
(a)
10 0
0.3 0.35 0.4 0.45 0.5
Current (A)

Time (s)
5 (a)
1
0 Voltage
0 0.2 0.4 0.6 0.8 1 1.2 0.8 Pow er
Time (s) Current
0.6 Measuring
(b) Currents
400 Operating
0.4 HC
Power (W)

Duty cycle

200 0.2

0
0.6 0.65 0.7 0.75 0.8
0
0 0.2 0.4 0.6 0.8 1 1.2 Time (s)
Time (s) (b)
(c)
Measuring Currents
1 Pow er
1
Duty Cycle

Current
0.8 Voltage
0.5
0.6
Operating
HC
0 0.4 Duty cycle
0 0.2 0.4 0.6 0.8 1 1.2
Time (s) 0.2
(d) 0
Fig. 9. Corresponding array’s (a) voltage, (b) current, (c) power and 0.9 0.95 1 1.05 1.1
(d) duty cycle waveforms in the first simulation. Time (s)
(c)
needed parameters and other important factors is still of a
Fig. 10. Zoomed view of per unit array’s voltage, current, power and
great importance. In this section, simulations are done to duty cycle waveforms in the first simulation during a) 0.3s to 0.5s, b)
compare the new method against two highly cited methods to 0.6s to 0.8s and c) 0.9s to 1.1s intervals. Power should be multiply to
show its benefits over those. It should be considered that prior 35×6, voltage should be multiply to 3×11.15 and current should be
works like [23] has shown that some of main hypothesis in multiply to 2×4.15.
[13] are not correct, and it may fail to track the GP in some
conditions. However, still [13] is now a classic and highly 1000
W/m2
1000
W/m2
600
W/m2
cited method, and most algorithms are compared to it. For
comparing the proposed method against [13] and [17], a PSC
pattern depicted in Fig. 11(a) is applied to the PV array. The 1000 300 300
corresponding P-V curve is shown in Fig. 11(b). Also, the W/m2 W/m2 W/m2

corresponding power waveforms of the proposed method, [13] (a)

and [17] are illustrated in Fig. 12(a), Fig. 12(b), and Fig. 12(c)
respectively. 100
It is illustrated in Fig. 12(a) that the proposed method tracks
Power (W)

the GP with corresponding 97 W power within 0.093s. The

method proposed in [13] tracks the same peak (i.e. the GP in 50
Fig. 11(b) with 99 W power), but in a longer time which is
0.103 s.
Although the method in [17] is faster than two other
methods and tracks the peak within 0.077 s, it fails to track the 0
0 10 20 30 40
GP correctly. It tracks the middle LP in the P-V curve (87.5W)
instead of the GP. So, it is proved that the proposed method in Voltage (V)
(b)
this paper has good performance in both speed and accuracy Fig. 11. Corresponding a) PSC pattern and b) P-V characteristics
factors in comparison to two highly cited methods. under second simulation.
Table IV summarizes the comparison of three methods in
important factors such as accuracy, convergence speed,
implementation complexity and needed parameters. As

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

100 Measuring Current in 3 points a c

80
Power (W)

60 Operating HC for
tracking 3 LPs
40

20 0.093 s b D d

3 × 2 PV Array
0
0 0.05 0.1 0.15
Time (s)
(a)

100
Fig. 13. Experimental setup, (a) The boost converter, (b) converter’s
80
inductor, (c) STM32F4 Discovery kit with ARM Cortex-M4 32-bit core
Power (W)

60 on the interface board, (d) PV array.

40 TABLE V
20
ELEMENTS OF EXPERIMENTAL SETUP
0.103 s
Element Name
0
0 0.05 0.1 0.15 ARM Processor STM32F4 Discovery kit with ARM
Time (s) Cortex-M4 32-bit core
(b) MOSFET Switch IRFP260N
Diode BYW80-200
100 Current Sensor ACS 712
Voltage Sensor HCPL 7840
80
Power (W)

60
TABLE VI
COMPARISON OF THE NEW METHOD WITH PROPOSED METHODS IN [13]
40 AND [17]
Tracked Tracking Achievable
20 Tracking
0.077 s Method Power Efficiency Power
Time (s)
0 (W) (%) (W)
0 0.05 0.1 0.15
Time (s) New
83 0.085 97
(c) Experimental Method
85.8
Fig. 12. Comparison of the performance of a) the proposed method, Investigation [13] 82 0.12 97
b) the proposed method in [13], and c) the proposed method in [17]. [17] 63 0.073 73
New
97 0.093 98
TABLE IV Simulation Method
99
COMPARISON OF THREE METHODS IN IMPORTANT FACTORS Investigation [13] 97 0.103 98
Convergence Implementation Needed [17] 87.5 0.077 88
Accuracy
Speed Complexity Parameters
new
Very high High Low Voc,m and Ns,m the corresponding power and current are derived. The
method
[13] Medium Medium Low Voc,m and Voc,array corresponding voltage, power, and current of array are
Voc,m , Isc,array illustrated in Fig. 14(a). As it is shown, three LPs which are
[17] High Very high Low and Voc,array in (5.6V, 5.4 A, 30.24 W), (14V, 4.65 A, 65.1 W), and (26V, 3.3
STC
A, 85.8 W) exist and the last one is the GP. Fig. 14(b), Fig.
14(c), and Fig. 14(d) show the performance of the proposed
demonstrated in Table IV, the proposed method has an method, [13], and [17] in this condition, respectively. It can be
appropriate relative performance in comparison to other two
seen that the proposed method tracks the GP with 83 W power
highly cited methods.
within 0.085s. The method proposed in [13] tracks the same
peak with 82 W power, but in a longer time which is 0.12 s.
V. EXPERIMENTAL RESULTS Similar to the second simulation, the method in [17] is
In order to check the validity of the proposed method in faster than two other methods and tracks the peak within 0.073
practice, the PV modules whose parameters are listed in S, but it fails to track the GP correctly. It tracks the middle LP
Tables I and II are used. The configuration of the array (i.e. in the P-V curve with 63 W power instead of the GP.
3×2 array), the schematic of the system and the parameters of So, as it was proved in the second simulation, the
setup under study are similar to the simulation one. In experimental results show that the proposed method in this
addition, the boost converter, interface board, and PV array paper outperforms the two other methods in both speed and
are depicted in Fig. 13. The elements used in the experimental accuracy. The performances of these three methods in the
setup are listed in Table V. The shadows are applied via dark second simulation test and experimental test are summarized
laminates in experimental tests manually. in Table VI.
In order to compare the proposed method to the methods of
[13] and [17], a test with a PSC similar to the PSC in the
second simulation is done. By changing the duty cycle
gradually, the voltage has changed from 2V to about 30V and

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

Voltage experimental results, it is shown that the current in each step

26 V
14 V of the I-V characteristic is almost constant until the beginning
10 V 5.6 V point of the next step. Besides, it is proved that the starting
85.8 W
65.1 W points of each step in the I-V curve are in near left side
neighborhood of the multiples of Voc,m.
50 W

30.24 W
The proposed method is in fact, a modified HC method
Power
5.4 A
4.65 A
which tracks the GP effectively under different conditions.
3.3 A
Thus, the implementation of this method is simple. Once the
5A

PSCs appear, the number and length of I-V characteristic's

Current
steps are recognized by measuring the current value in
multiples of Voc,m. Then, around specific multiples of
0.8×Voc,m the HC method tracks all LPs. Finally, the GP is
100 ms
detected by comparing the LPs. Simulation and experimental
(a)
results have validated the advantages of this method in terms
of accuracy and speed over two popular existing methods.
85 ms
Power
REFERENCES
Voltage [1] T. Esram and P. L. Chapman, "Comparison of photovoltaic array
50 W

maximum power point tracking techniques," IEEE Trans. Energy

Convers., vol. 22, pp. 439-449, Jun. 2007.
Operating HC
[2] K. Ishaque and Z. Salam, "A review of maximum power point tracking
techniques of PV system for uniform insolation and partial shading
10 V

condition," Renew. Sustain. Energy Rev., vol. 19, pp. 475-488, Mar.
Measuring Current
Current 2013.
[3] Liu, Liqun, Xiaoli Meng, and Chunxia Liu. "A review of maximum
5A

power point tracking methods of PV power system at uniform and

partial shading," Renew. Sustain. Energy Rev., vol. 53, pp. 1500-1507,
25 ms
Jan. 2016.
(b) [4] A. Kouchaki, H. Iman-Eini, and B. Asaei, "A new maximum power
point tracking strategy for PV arrays under uniform and non-uniform
insolation conditions," Solar Energy, vol. 91, pp.221-232, May 2013.
120 ms [5] R. C. Pilawa-Podgurski and D. J. Perreault, "Submodule integrated
Power
distributed maximum power point tracking for solar photovoltaic
Voltage
applications," IEEE Trans. Power Electron., vol. 28, pp. 2957-2967,
50 W

Jun. 2013.
[6] C. Woei-Luen, and C. Tsai. "Optimal balancing control for tracking
theoretical global MPP of series PV modules subject to partial shading."
IEEE Trans. Ind. Electron., vol. 62, no. 8, pp. 4837-4848, Aug. 2015.
10 V

[7] K. Ishaque, Z. Salam, M. Amjad, and S. Mekhilef, "An improved

Particle Swarm Optimization (PSO)–based MPPT for PV with reduced
Current
steady-state oscillation," IEEE Trans. Power Electron., vol. 27, pp.
5A

3627-3638, Aug. 2012.

[8] E. Karatepe and T. Hiyama, "Performance enhancement of photovoltaic
25 ms
array through string and central based MPPT system under non-uniform
(c) irradiance conditions," Energy Convers. Manage., vol. 62, pp. 131-140,
Oct. 2012.
[9] S. Daraban, D. Petreus, and C. Morel, "A novel global MPPT based on
73 ms genetic algorithms for photovoltaic systems under the influence of
partial shading," in Proc. Annu. Conf. IEEE Ind. Electron. Soc., Vienna,
Power
Austria, Nov. 2013, pp. 1490-1495.
50 W

[10] K. Sundareswaran, S. Peddapati, and S. Palani, “MPPT of PV systems

under partial shaded conditions through a colony of flashing fireflies,”
Voltage IEEE Trans. Energy Convers., vol. 29, no. 2, pp. 463-472, Jun. 2014.
[11] K. Sundareswarm, P. Sankar, P. S. R. Nayak, S. P. Simon, and S. Palani,
10 V

“Enhanced energy output from a PV system under partial shaded

conditions through artificial bee colony,” IEEE Trans. Sustain. Energy,
vol. 6, no. 1, pp. 198−209, Jan. 2015.
Current [12] S. Lyden, and M. E. Haque, "A simulated annealing global maximum
5A

power point tracking approach for PV modules under partial shading

25 ms conditions." IEEE Trans. Power Electron., vol. 31, no. 6, pp. 4171-4181,
(d) Jun. 2016.
Fig. 14. (a) Corresponding array power, voltage, and current in [13] H. Patel and V. Agarwal, "Maximum power point tracking scheme for
experimental test, performance of (b) the proposed method, (c) [13], PV systems operating under partially shaded conditions," IEEE Trans.
and (d) [17]. Ind. Electron., vol. 55, pp. 1689-1698, Apr. 2008.
[14] J. Qi, Y. Zhang, and Y. Chen, "Modeling and maximum power point
tracking (MPPT) method for PV array under partial shade conditions,"
VI. CONCLUSION Renew. Energy, vol. 66, pp. 337-345, Jun. 2014.
[15] E. Koutroulis and F. Blaabjerg, "A new technique for tracking the global
In this paper a novel MPPT method is proposed which has a maximum power point of PV arrays operating under partial-shading
great performance under PSC. Based on the simulation and conditions," IEEE J. Photovolt., vol. 2, pp. 184-190, Apr. 2012.

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2016.2632679, IEEE
Transactions on Industrial Electronics
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

[16] KS. Tey and S. Mekhilef, "Modified incremental conductance algorithm

for photovoltaic system under partial shading conditions and load
variation," IEEE Trans. Ind. Electron., vol. 61, no. 10, pp. 5384-5392,
Oct. 2014.
[17] M. Boztepe, F. Guinjoan, G. Velasco-Quesada, S. Silvestre, A. Chouder
and E. Karatepe, "Global MPPT scheme for photovoltaic string inverters
based on restricted voltage window search algorithm," IEEE Trans. Ind.
Electron., vol. 61, no. 7, pp. 3302-3312, Jul. 2014.
[18] K. Chen, S. Tian, Y. Cheng and L. Bai, "An improved MPPT controller
for photovoltaic system under partial shading condition," IEEE Trans.
Sustain. Energy, vol. 5, no. 3, pp. 978-985, Jul. 2014.
[19] W. Yunping, Y. Li, and X. Ruan, "High-accuracy and fast-speed MPPT
methods for PV string under partially shaded conditions," IEEE Trans.
Ind. Electron., vol. 63, no. 1, pp. 235-245, Jan. 2016.
[20] M. A. Ghasemi, H. Mohammadian Forushani, and M. Parniani. "Partial
Shading Detection and Smooth Maximum Power Point Tracking of PV
Arrays Under PSC." IEEE Trans. Power Electron., vol. 31, no. 9, pp.
6281-6292, Sep. 2016.
[21] P. Lei, Y. Li, and J. E. Seem, "Sequential ESC-Based Global MPPT
Control for Photovoltaic Array With Variable Shading," IEEE Trans.
Sustain. Energy, vol. 2, No. 3, pp. 348-358, Jul. 2011.
[22] M. G. Villalva and J. R. Gazoli, "Comprehensive approach to modeling
and simulation of photovoltaic arrays," IEEE Trans. Power Electron.,
vol. 24, pp. 1198-1208, May 2009.
[23] S. Kazmi, H. Goto, O. Ichinokura, and H. J. Guo. "An improved and
very efficient MPPT controller for PV systems subjected to rapidly
varying atmospheric conditions and partial shading," in Proc. AUPEC,
2009, pp. 1-6.

Alireza Ramyar received the B.Sc. degree in

electrical engineering from the Sharif University
of Technology, Tehran, Iran, in 2013, and the
M.Sc. degree in electrical engineering from the
University of Tehran, Tehran, Iran, in 2015. His
research interests include design, modeling and
control of power converters, photovoltaic, and
renewable energy systems.

Hossein Iman-Eini (M’ 10) received his B.S.

and M.S. from the University of Tehran, Tehran,
Iran, in 2001 and 2003, respectively, and his
Ph.D. from both the University of Tehran and
the Grenoble Institute of Technology, France, in
2009, all in electrical Engineering. He is
currently an Associate Professor in the School
of Electrical and Computer Engineering,
University of Tehran. His current research
interests include the modeling and control of
power converters, multilevel converters, and
renewable energy systems.

Shahrokh Farhangi (M’ 90) obtained the B.Sc.,

M.Sc. and Ph.D. degrees in electrical
engineering from University of Tehran, Iran, with
honors. He is currently professor of School of
Electrical and Computer Engineering, University
of Tehran. His research interests include design
and modeling of Power Electronic Converters,
Drives, Photovoltaics and Renewable Energy
Systems. He has published more than 100
papers in conference proceedings and journals.
He has managed several research and industrial projects, which some
of them have won national and international awards. He has been
selected as the distinguished engineer in electrical engineering by Iran
Academy of Sciences, in 2008.

0278-0046 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.