See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/304566123

Comparision of ‘perturb & observe’ and ‘incremental conductance’,

maximum power point tracking algorithms on real environmental conditions

Conference Paper · April 2016

DOI: 10.1109/ICECUBE.2016.7495244

CITATIONS READS

22 2,173

2 authors:

Muhammad Ali Baig Muhammad Hammad Uddin

NED University of Engineering and Technology, Karachi NED University of Engineering and Technology, Karachi
9 PUBLICATIONS 71 CITATIONS 9 PUBLICATIONS 99 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Muhammad Hammad Uddin on 27 November 2018.

The user has requested enhancement of the downloaded file.

 Comparision of ‘Perturb & Observe’ and
‘Incremental Conductance’ Maximum Power Point
Tracking algorithm on real environmental conditions
Muhammad Hammad Uddin1, IEEE member, Muhammad Ali Baig2,IEEE Member, Muhammad Ali3, IEEE member,
1, 2 & 3
Department of Electrical Engineering, NED University of Engineering & Technology, Karachi, Pakistan.
[email protected], [email protected], [email protected]

Abstract—The trend and need of shifting towards renewable dynamic variables throughout the day timings. In order to extract
energy sources for electricity generation, has attracted researchers maximum power from the PV panel, MPPT techniques are
towards photovoltaic (PV) systems for last few decades. Since PV applied [12-14].
generator has nonlinear behavior, and its output power is a
function of solar irradiance and ambient temperature, therefore Nuclear Wind
need of developing maximum power point tracking (MPPT) 3% 1%
techniques remains the interest. Among various MPPT
techniques, Perturb & Observe (P&O) and Incremental- Hydel,
Conductance (IC) are dominant. This paper describes these two 29.2%
algorithms and compares them in terms of their computational
complexity and oscillations near maximum power point (MPP)
using MATLAB® & Simulink®. Also testing conditions are based
on the real environmental data collected at NED University,
Karachi.
Thermal,
Keywords—Maximum Power Point (MPP), Perturb & Observe 67.14%
(P&O), Increamental Conductance (IC), Duty control
Fig. 1. Energy mix of Pakistan for electricity production [2]
I. INTRODUCTION
TABLE 1
Increase in energy demand is obvious, with the growth in
Solar Cell efficiency measurement under standard testing conditions[10]
world’s population. It is expected that world primary energy
demand will increase by 60% from 2002 to 2030, globally[1]. Open Current
Fig.1, shows the energy mix of Pakistan, which reflects a Efficiency Area Circuit Density
dominant portion contributed by thermal sources and fractional Classification
(%) (cm2) Voltage Jsc
contribution of renewables i.e. wind power [2]. PV is considered Voc (V) (mA/cm2)
one of the most important resources of energy for future[1, 3]. Si (crystalline) 25.6±0.5 143.7 0.74 41.8
By 2011, almost 30 GW of electricity had been produced by PV
solar cell globally [4]. Currently Pakistan lags the capability of Si (multi crystalline) 20.8±0.6 243.9 0.6626 39.03
harnessing solar energy, though it has a great potential of
generating energy through it [4, 5]. In this regard, several Si (thin film mini
10.5 ±0.3 94 0.492 29.7
module)
projects are initiated that could utilize solar and other renewable
energy resources available in Pakistan [5-8].
Efficiency of PV cell plays important role for the feasibility
evaluation of power generation with PV. The efficiency of solar +
panel is low and its capital cost is high when compared with Photovoltaic DC-DC
Load
DC

Converter
other power generating techniques[9, 10]. Feasibility evaluation Module - Battery
is based on a standard testing condition (STC), in which PV cell
efficiency is observed by applying 1000 Watt/m2 irradiance, at I
I V
250C Temperature and air mass of 1.5 [10-12]. [10], summarizes
highest confirmed efficiency of commercially employed solar
cell, which as listed in Table 1. Duty
MPPT Cycle
Besides lower efficiency, PV solar panel has nonlinear Controller
characteristics in terms of current verses voltage (I-V) and power
verses voltage (P-V). Also the output power of PV cell is Fig. 2. Typical PV System components[15]
dependent on irradiance and ambient temperature, which are

Page 1 of 5
Block diagram of a typical PV system in terms of components is TABLE II
shown in Fig. 2. In principle, the MPPT controller derives the Parameters of solar array KC200GT at STC
DC-DC converter, which serves as an interface between load Maximum Power Pmax (W) 200±10%/−5%
and PV cell, by controlling its duty cycle to extract maximum
Current at MPP Impp (Ampere) 7.61
power out of the PV cell based on environmental conditions. The
output of DC-DC converter can be used to operate DC load Voltage at MPP Vmpp (Volts) 26.3
directly and can also be utilized to charge the battery. Short circuit current Iscc (Ampere) 8.21
Open circuit voltage Voc (Volts) 32.9
In this paper, comparison between P&O and IC is made under
Number of cell in series Ns 54
real weather condition at NED University Karachi. The work
focuses on the implementation of the two techniques in
MATLAB® and Simulink®, also describes their ability to
handle varying environmental conditions. The organization of
the paper is as follow: Section II, describes the modelling of
photovoltaic system in Simulink®. In Section III, dynamics of
P&O and IC algorithm in terms of MPP tracking are compared
and Section IV, has conclusion with results and discussion.
II. PV SYSTEM MODELLING IN SIMULINK
A. PV Solar panel Fig. 4. PV Curve of modeled solar panel @ STC constant condition

Literature shows techniques of modeling solar cell in

Fig.5, shows power dependency of solar panel on irradiation
simulations, with a prominent difference of either using single
condition. The PV curve at various irradiance i.e. at 1000 W/m2,
diode solar model or double diode solar model. They exhibit
800 W/m2, 600 W/m2, 400 W/m2 and 200 W/m2 are shown in
real behavior of device with inclusion of losses [12, 16-20]. In
Fig. 5. While the temparature of panel kept constant at 250C.
this paper single diode model is used and simulated similarly
Clealry the MPP varies in proportion with irradiance.
done in [12]. The equivalent model of single diode solar cell is
shown in Fig. 3. The current equation for equivalent solar
model is given in “1”.
𝑉+𝑅𝑠 𝐼
( ) 𝑉+𝑅𝑠 𝐼
𝐼 = 𝐼𝑝𝑣 − 𝐼0 [𝑒 𝛼 𝑉𝑡 − 1] − ……………………….. (1)
𝑅𝑝

Here Ipv = photovoltaic current (Ampere)

I0 = Saturation current (Ampere)
N kT
Thermal voltage = Vt = s (Volts)
q
N𝑠 = number of cells connected in series
k = Boltzmann constant (Joule/Kelvin)
T= Temperature of solar cell (Kelvin)
q = The elementary charge on an electron. (Coulomb) Fig. 5. PV Curve at different irradiance
Rs = equivalent series resistance of the PV array (Ohms)
Rp = equivalent parallel resistance of the PV array (Ohms) B. MPPT Algorithm
NP = Number of cells connected in parallel The Purpose of MPPT algorithm in PV system is to keep the
PV panel voltage closes to MPP voltage. Several authors has
Based on single diode model the solar array having model contributed towards novel MPPT techniques and compared
number as KC200GT has been simulated in this paper[12]. The these techniques [21-26]. Many Authors have compared two
parameters of the solar array been given in Table 2 under STC. fundamental MPPT techniques i.e. P&O and IC in literature
Characteristics of modeled solar cell can be measured in terms [27, 28]. With the best of knowledge, no author has done the
of power verses voltage (P-V) curves and is given in Fig. 4. qualitative comparison showing MPPT algorithm performance
I in MPP tracking. Furthermore duty control dynamics also not
shown in MPP tracking. The qualitative comparison of both
Rs algorithm is explained here with the real weather conditions at
I0
NED University, Karachi.
Rp V
Ipv 1) Perturb & Observe MPPT Algorithm
In this method the solar panel voltage is first perturb and output
power is calculated. If the power increases on increasing
voltage, then voltage is further increased. If the power decreases
on increase in voltage then the voltage of panel is reduced to
Fig. 3. Equivalent Model of Single Diode Solar cell[12] search for higher value of output power. The variations in

Page 2 of 5
operating voltage is controlled by a constant value. Since the Start

algorithm always keeps searching for MPP, the P&O algorithm

Measure V(n), I(n)
keeps the panel output power oscillating around MPP. This
method is simple because, it uses calculation of maximum 3 dI=In-Ib
steps to track MPP. The oscillation of output power in P&O is dV=Vn-Vb

natural and also the weakness of algorithm. In literature Yes

dV=0?
modified versions of P&O are given that improve MPP tracking
performance but no proof is presented for better performance in Yes Yes
dI/dV= -I/V ? dI=0 ?
variable irradiation condition [29-31]. The flow chart in Fig. 6,
No No
summarizes the P&O algorithm.
Yes dI>0 ?
Start dI/dV>-I/V ?
No Yes

No
Measure V(t), I(t)
V(r) = Vr-∆ V V(r) = Vr-∆ V V(r) = Vr-∆ V V(r) = Vr+∆ V

Vb=Vn
Calculate Power, P(t)
Ib=In

No Yes Return
P(t)>P(t-1)
Fig. 7. IC MPPT Algorithm[28]
No Yes No Yes Inductor
V(t) > V(t-1)
V(t) > V(t-1)
Vout < Vin
Mosfet
Diode
D(t) = D(t-1)+∆ D D(t) = D(t-1)-∆ D D(t) = D(t-1)-∆ D D(t) = D(t-1)+∆ D Vin Capacitor Load
Duty
Fig. 6. P&O MPPT algorithm[28] Cycle

2) Incremental Conductance MPPT Algorithm

Fig. 8. Equivalent Buck Converter circuit [38]
This method addresses the deficiency in P&O algorithm under
variable irradiance conditions. In order to reach at MPP this III. SIMULATION RESULTS ON MPPT COMPARISION
method uses the slop of differential voltage dV and current dI.
1) At STC
The relationship of IC algorithm is shown in “(2)”, “(3)” and
To show dynamics of PV system using P&O and IC at STC, a
“(4)” [32].
dI I dP simulation for 0.5 sec was performed and results of MPP
=− ( = 0). At MPP …………… (2) achieved with duty control at MPP tracking are shown in Fig. 9
dV V dV
dI I dP and 10 respectively. These results assume that irradiation of
>− ( > 0). At the left of MP …... (3)
dV V dV constant value 1000 Watt/m2 was kept at temperature 250C. At
dI I dP
<− ( < 0) At the right of MPP.. (4) this condition system move from zero-initial condition to the
dV V dV
steady-state value.
The perturbation in the voltage is applied until “(2)” is
achieved. And the direction of perturbation can be determined 2) At real environmental conditions at NED University
by equation “(3)” and “(4)”. The flow chart of IC algorithm is Since system simulation had to consider the dynamics of
shown in Fig.7. irradiance and temperature for the complete day, instead of
running simulation for 3600 seconds on Matlab ®. Because of
The disadvantage of P&O algorithm under rapidly changing memory constraints, the parameter dynamics are scaled for 67
irradiance condition is overcome by using IC algorithm [33]. seconds only. After monitoring weather conditions for a month
P&O oscillate at MPP, while IC can find its MPP and started from 23rd Dec 2015 to 23rd Jan 2016, it is concluded that
perturbation is clogged. There is a disadvantage of IC as weather pattern looks same throughout the month. Therefore a
compared to P&O is its complexity level. The maximum 7 step typical day profile can be chosen to test the simulation. The real
calculation takes place in IC [34, 35]. solar irradiance and ambient temperature of a typical day 23 rd
Dec 2015 has been selected. The solar irradiance and ambient
C. DC-DC Converter
temperature for a day is shown in Figs. 11 and 12 respectively.
In this work, the buck converter is employed because it has been Considering real atmospheric conditions shown, simulation of
proved in literature, that the buck converter has better efficiency PV system using P&O and IC MPPT algorithm has been run.
at all light level among other types of DC-DC converters [36, The results been shown in Figs. 13 and 14 respectively.
37]. The Inductor of 50 mH and capacitor of 1mF have been
used. The equivalent circuit is shown in Fig. 8.

Page 3 of 5
Fig. 9. PV panel Power and Duty cycle control using P&O at STC
Fig. 14. IC simulation results on a typical day profile

IV. RESULTS AND DISCUSSION

1) Comparison at constant irradiance Condition
The PV system performance by using P&O and IC MPPT
algorithm at STC are shown in Figs. 9 and 10 respectively. Fig.
9, clearly depicts that at constant irradiation, P&O finds MPP
quickly but oscillate at MPP. While as shown in Fig. 10, IC
finds MPP after time delay. But the oscillations at constant
irradiation are less as compared to P&O.
Fig. 10. PV panel Power and Duty cycle control using IC at STC The major finding in this work is the representation of duty
control dynamics of both algorithm. As shown in Fig. 9, duty
cycle of P&O is perturb continuously and results in greater
oscillation at MPP. While in Fig. 10, duty cycle of IC is not
changed continuously, it provides fine control on duty cycle.
This fine control on duty cycle provide less oscillation at MPP
and better tracking performance at variable irradiance
conditions. If same fine control is gained by P&O then its duty
control dynamics will be improved and will results in lower
oscillations and better tracking MPP performance.

2) At real environmental conditions at NED University,

Fig. 11. A typical day solar irradiance pattern at NED University
Simulation results are then compared at real weather data
obtained at NED University weather station. A typical day
irradiance and temperature profile is shown in Figs. 11 and 12.
When the simulation under same condition run on both MPPT
algorithm. The results are shown in Figs. 13 and 14. As Fig. 13,
shows the P&O follows the irradiation pattern, but with
increased number of oscillation while tracking MPP. This
shows incompetency of P&O to find MPP at variable irradiance
condition or practical scenario. Fig. 14, shows how efficiently
IC follows irradiance pattern. As the power output has low
oscillations and better MPP tracking performance in real
Fig. 12. A typical day ambient Temperature pattern at NED University
weather conditions.
CONCLUSION
It is found that P&O MPPT algorithm has better tracking
performance at constant environmental conditions i.e. constant
irradiance and temperature which is not a good representation
of real weather conditions. But even at constant condition i.e. at
STC, duty control dynamics of P&O is poor as compare to IC.
IC has better MPP tracking performance as compare to P&O in
variable environmental conditions such as clouds moving over
panel which results in variation of irradiance. Under real
weather conditions at NED University, performance of IC
Fig. 13. P&O simulation results on a typical day profile MPPT algorithm was found to be better as compare to P&O.

Page 4 of 5
 ACKNOWLEDGMENT [19] K. Ishaque, Z. Salam, and H. Taheri, "Accurate MATLAB simulink
PV system simulator based on a two-diode model," Journal of
We would like to acknowledge Alternate Energy Board Power Electronics, vol. 11, pp. 179-187, 2011.
(AEDB) Pakistan for providing real weather data of NED [20] D. Chan, J. Phillips, and J. Phang, "A comparative study of
University, in collaboration with solar energy lab of department extraction methods for solar cell model parameters," Solid-State
Electronics, vol. 29, pp. 329-337, 1986.
of Mechanical Engineering. [21] A. R. Reisi, M. H. Moradi, and S. Jamasb, "Classification and
Also would like to acknowledge the contribution of Prof. Dr. comparison of maximum power point tracking techniques for
Saad Ahmed Qazi, Dean ECE Faculty, Mr. Naveed ur Rehman, photovoltaic system: A review," Renewable and Sustainable Energy
Assistant Professor at Mechanical Engineering Department and Reviews, vol. 19, pp. 433-443, 2013.
[22] T. Logeswaran and A. SenthilKumar, "A review of maximum power
Mr. Iqbal Azeem, Lecturer at Electrical Engineering point tracking algorithms for photovoltaic systems under uniform
Department, NED University in this work. and non-uniform irradiances," Energy Procedia, vol. 54, pp. 228-
235, 2014.
[23] T. Esram and P. L. Chapman, "Comparison of photovoltaic array
REFERENCES maximum power point tracking techniques," IEEE Transactions on
[1] K. Solangi, M. Islam, R. Saidur, N. Rahim, and H. Fayaz, "A review Energy Conversion EC, vol. 22, p. 439, 2007.
on global solar energy policy," Renewable and sustainable energy [24] L. Arun Shravan and D. Ebenezer, "Maximum Power Point
reviews, vol. 15, pp. 2149-2163, 2011. Tracking (MPPT) for a Solar Photovoltaic System: A Review," in
[2] NEPRA, "State of industry report 2014," 2014. Applied Mechanics and Materials, 2015, pp. 227-232.
[3] A. W. Bhutto, A. A. Bazmi, and G. Zahedi, "Greener energy: issues [25] C. Ke, F. Xingming, L. Jueqiang, and W. Yinglong, "Review of the
and challenges for Pakistan—solar energy prospective," Renewable MPPT algorithms of photovoltaic array," Journal of Guilin
and Sustainable Energy Reviews, vol. 16, pp. 2762-2780, 2012. University of Electronic Technology, vol. 5, p. 011, 2011.
[4] K. Harijan, M. A. Uqaili, and U. K. Mirza, "Assessment of Solar PV [26] D. Hohm and M. E. Ropp, "Comparative study of maximum power
Power Generation Potential in Pakistan," Journal of Clean Energy point tracking algorithms," Progress in photovoltaics: Research and
Technologies, vol. 3, pp. 54-56, 2015. Applications, vol. 11, pp. 47-62, 2003.
[5] U. K. Mirza, M. M. Maroto-Valer, and N. Ahmad, "Status and [27] S. S. Kondawar and U. Vaidya, "A comparison of two MPPT
outlook of solar energy use in Pakistan," Renewable and Sustainable techniques for pv system in Matlab simulink," International Journal
Energy Reviews, vol. 7, pp. 501-514, 2003. of Engineering Research and Development, vol. 2, pp. 73-79, 2012.
[6] M. A. Sheikh, "Energy and renewable energy scenario of Pakistan," [28] N. Moubayed, A. El-Ali, and R. Outbib, "A comparison of two
Renewable and Sustainable Energy Reviews, vol. 14, pp. 354-363, MPPT techniques for PV system," WSEAS Transactions on
2010. Environment and Development, vol. 12, pp. 73-79, 2009.
[7] M. A. Chaudhry, R. Raza, and S. Hayat, "Renewable energy [29] L. Piegari and R. Rizzo, "Adaptive perturb and observe algorithm
technologies in Pakistan: prospects and challenges," Renewable and for photovoltaic maximum power point tracking," IET Renewable
Sustainable Energy Reviews, vol. 13, pp. 1657-1662, 2009. Power Generation, vol. 4, pp. 317-328, 2010.
[8] M. A. Sheikh, "Renewable energy resource potential in Pakistan," [30] A. K. Abdelsalam, A. M. Massoud, S. Ahmed, and P. Enjeti, "High-
Renewable and Sustainable Energy Reviews, vol. 13, pp. 2696- performance adaptive perturb and observe MPPT technique for
2702, 2009. photovoltaic-based microgrids," Power Electronics, IEEE
[9] R. E. Sims, H.-H. Rogner, and K. Gregory, "Carbon emission and Transactions on, vol. 26, pp. 1010-1021, 2011.
mitigation cost comparisons between fossil fuel, nuclear and [31] A. Al-Diab and C. Sourkounis, "Variable step size P&O MPPT
renewable energy resources for electricity generation," Energy algorithm for PV systems," in Optimization of Electrical and
policy, vol. 31, pp. 1315-1326, 2003. Electronic Equipment (OPTIM), 2010 12th International
[10] M. A. Green, K. Emery, Y. Hishikawa, W. Warta, and E. D. Dunlop, Conference on, 2010, pp. 1097-1102.
"Solar cell efficiency tables (Version 45)," Progress in [32] K. Hussein, I. Muta, T. Hoshino, and M. Osakada, "Maximum
photovoltaics: research and applications, vol. 23, pp. 1-9, 2015. photovoltaic power tracking: an algorithm for rapidly changing
[11] H. S. Rauschenbach, Solar cell array design handbook - The atmospheric conditions," in Generation, Transmission and
principles and technology of photovoltaic energy conversion, 1980. Distribution, IEE Proceedings-, 1995, pp. 59-64.
[12] M. G. Villalva and J. R. Gazoli, "Comprehensive approach to [33] S. Mekhilef and M. N. Abdul Kadir, "Voltage control of three-stage
modeling and simulation of photovoltaic arrays," Power hybrid multilevel inverter using vector transformation," Power
Electronics, IEEE Transactions on, vol. 24, pp. 1198-1208, 2009. Electronics, IEEE Transactions on, vol. 25, pp. 2599-2606, 2010.
[13] M. A. De Brito, L. P. Sampaio, L. G. Junior, and C. A. Canesin, [34] A. Safari and S. Mekhilef, "Simulation and hardware
"Evaluation of MPPT techniques for photovoltaic applications," in implementation of incremental conductance MPPT with direct
Industrial Electronics (ISIE), 2011 IEEE International Symposium control method using cuk converter," Industrial Electronics, IEEE
on, 2011, pp. 1039-1044. Transactions on, vol. 58, pp. 1154-1161, 2011.
[14] M. A. Eltawil and Z. Zhao, "MPPT techniques for photovoltaic [35] H. N. Zainudin and S. Mekhilef, "Comparison study of maximum
applications," Renewable and Sustainable Energy Reviews, vol. 25, power point tracker techniques for PV systems," in Proceedings of
pp. 793-813, 2013. the 14th international middle east power systems conference
[15] M. Boxwell, Solar Electricity Handbook: A Simple, Practical Guide (MEPCON’10), Cairo University, Egypt, 2010, pp. 19-21.
to Solar Energy: how to Design and Install Photovoltaic Solar [36] C. Hua and C. Shen, "Study of maximum power tracking techniques
Electric Systems: Greenstream Publishing, 2015. and control of DC/DC converters for photovoltaic power system,"
[16] D. S. Chan and J. C. Phang, "Analytical methods for the extraction in Power Electronics Specialists Conference, 1998. PESC 98
of solar-cell single-and double-diode model parameters from IV Record. 29th Annual IEEE, 1998, pp. 86-93.
characteristics," IEEE Transactions on Electron Devices, pp. 286- [37] C. Hua and C. Shen, "Control of DC/DC converters for solar energy
293, 1987. system with maximum power tracking," in Industrial Electronics,
[17] K. Ishaque and Z. Salam, "A comprehensive MATLAB Simulink Control and Instrumentation, 1997. IECON 97. 23rd International
PV system simulator with partial shading capability based on two- Conference on, 1997, pp. 827-832.
diode model," Solar Energy, vol. 85, pp. 2217-2227, 2011. [38] M. Wens and M. Steyaert, Design and implementation of fully-
[18] K. Ishaque, Z. Salam, and H. Taheri, "Simple, fast and accurate two- integrated inductive DC-DC converters in standard CMOS:
diode model for photovoltaic modules," Solar Energy Materials and Springer Science & Business Media, 2011.
Solar Cells, vol. 95, pp. 586-594, 2011.

Page 5 of 5

View publication stats