International Journal of Renewable Energy Resources 2 (2012) 33-39

MAXIMUM POWER POINT TRACKING ALGORITHMS FOR WIND ENERGY SYSTEM:

A REVIEW

M.A. Abdullah, A.H.M. Yatim and C.W. Tan

Department of Energy Conversion, Faculty of Electrical Engineering, Universiti Teknologi Malaysia (UTM)
Email address: aaamajid2@live.utm.my

ABSTRACT is essential to convert the variable-voltage-variable-

This paper reviews and studies the state of the art available frequency of the voltage-fixed-frequency that is suitable
maximum power point tracking (MPPT) algorithms. Due for the grid. References (Baroudi et al., 2007; Zhe et al.,
to the nature of wind which is instantaneously changing, 2009) have discussed the different possible configurations
there is only one optimal generator speed desirable at any of power electronic converters and electrical generators for
one time that ensures maximum energy is harvested from variable speed wind turbine systems.
the available wind. Therefore, including a controller that is
able to track the maximum peak regardless of any wind Among the electric generators, permanent magnet
speed is essential. The available maximum power point synchronous generator (PMSG) is preferred due to its high
tracking (MPPT) algorithms can be classified according to efficiency, reliability, power density; gearless
the control variable, namely with and without sensor, and construction, light weight, and self-excitation features (Li
also the technique used to locate the maximum peak. A et al., 2010; Molina et al., 2010; Muyeen et al., 2010;
comparison has been made on the performance of the Mena 2007). Controlling the PMSG to achieve the
selected MPPT algorithms based on various speed maximum power point (MPP) can be done by varying its
responses and the ability to achieve the maximum energy load. In this regard, a boost converter is one of the possible
yield. The tracking performance is performed by solutions, where, by controlling the duty cycle of the
simulating wind energy system using the converter the apparent load seen by the generator will be
MATLAB/Simulink simulation package. Besides that, a adjusted and thus, its output voltage and shaft speed. In
brief and critical discussion is made on the differences of addition to that, operating the boost converter in
available MPPT algorithms for wind energy system, discontinuous conduction mode (DCM) and applying a
followed by a conclusion. power factor correction (PFC) technique contributes in
total harmonic distortion (THD) reduction and increases
Keywords: MPPT; Wind energy system; PMSG; Boost the power factor (PF) of the wind power generator
converter (Kawale and Dutt 2009; Carranza et al., 2010).

1. INTRODUCTION In order to determine the optimal operating point of the

Wind energy systems as one of the renewable energy wind turbine, a maximum power point tracking (MPPT)
sources have gained popular demand over the past decade algorithm is essential to be included in the system. Several
due to many factors such as the possibility of depletion of MPPT algorithms have been proposed in the literature.
conventional energy sources, its high costs, as well as Reference (Raza et al., 2010) has reviewed and criticized
having negative effects on the environment. Wind energy many published MPPT algorithms and concluded that the
is preferred because it is clean, pollution-free, two methods described in (Hui and Bakhshai 2008) and
inexhaustible and secure. Therefore, a wind energy (Kazmi et al., 2011) are the best solutions due to their
generation system could be one of the significant adaptive tracking and self-tuning capability. References
candidates as an alternative energy source for the future. (Mirecki et al., 2004; Brahmi et al., 2009; AJ Mahdi et al.,
The amount of mechanical energy that can be extracted 2010) have compared some of the available MPPT for
from the wind is not solely dependent on the wind speed, PMSG-based wind energy conversion system. This paper
but also governed by the ratio of the rotational speed to reviews the fundamentals of the available MPPT
wind speed. There is a specific optimal ratio for each wind algorithms for wind energy system. In addition, a
turbine, which is called the optimal tip speed ratio (TSR) comparison of simulation results is made on the three
or opt , at which the extracted power is maximum. As the selected MPPT techniques. Finally, a critical discussion is
made, and a conclusion is drawn.
wind speed is instantaneously varying, it is essential for
the rotational speed to be variable to maintain the equality 2. SYSTEM OVERVIEW
of the TSR to the optimal one at all times. In the operation Figure 1 illustrates the schematic diagram of the proposed
of variable speed condition, a power electronic converter wind turbine system. The system supplies a resistive load

33
and consists of wind turbine rotor, PMSG, rectifier and a maximum point, denoted by the opt , where the C p is
boost converter.
maximum. Continuous operation of wind turbine at this
point guarantees the maximum available power can be
harvested from the available wind at any speed, as shown
in Fig. 3.
DC-DC Boost

Load
PMSG Uncontrolled
Rectifier Converter

Figure 1 A brief block diagram of the proposed PMSG

wind energy system

Wind turbine converts the wind energy at its input to a

mechanical energy at the output, which in turn, runs a
generator to generate electrical energy. The mechanical
power generated by wind turbine can be expressed as
(Freris 1990):
1 Figure 2 The characteristic of the power coefficient as a
Pm R 2V 3 C p( , ) (1) function of the tip speed ratio
2
where m 3 ), R is the turbine
is the air density ( kg
rotor ( m ), Vw is the wind speed ( m / s ), and C p is the
coefficient of performance. The turbine power
coefficient, C p describes the power extraction efficiency
of the wind turbine (Grimble and Johnson 2008). It is a
nonlinear function of both tip speed ratio, and the blade
pitch angle, . While its maximum theoretical value is
approximately 0.59, it is practically between 0.4 and 0.45
(Zhe et al., 2009). The tip speed ratio is a variable
expressing the ratio of the linear speed of the tip of blades
to the rotational speed of wind turbine (Freris 1990).
m R
(2)
Vw
Where m is the mechanical angular velocity of the rotor
measured in rad/s. There are many different versions of Figure 3 Characteristics of turbine power as a function of
fitted equations for C p made in the previous studies. This the rotor speed for a series of wind speeds

paper defines C p as (Mena 2007): 3. MPPT TECHNIQUES

21
1
i (3) A. Tip Speed Ratio Control
C p ( , ) 0.5
116 0.4 5
e
i The optimal TSR for a given wind turbine is constant
1 1 0.035 regardless of the wind speed. If the TSR is maintained
(4)
i 0.08 1 3 constantly at its optimal value, this ensures that the energy
extracted is in its maximum operating point too.
In this paper, due to the assumption of a fixed pitch rotor,
Therefore, this method seeks to force the energy
the is set constant. Hence, the characteristics of the C p conversion system to work at this point continuously by
mainly depend on the only. Fig. 2 presents the C p as a comparing it with the actual value, represented in (2), and
feeding this difference to the controller. That, in turn,
function of the . Based on the figure, there is only one changes the speed of the generator to reduce this error.

34
The optimal point of the TSR can be determined lookup table in (Quincy and Liuchen 2004). According to
experimentally or theoretically and stored as a reference. (Raza et al., 2010), there is no difference between the PSF
This method is simple; however, it requires the and the OT method in terms of the performance and the
measurement of wind speed consistently and accurately, complexity of implementation.
which complicates its use in reality, as well as increases
the system cost (Patel 1999; Barakati 2008; Wang, 2003).
B. Optimal Torque Control
As mentioned earlier, maintaining the operation of the
wind turbine system at the opt ensures that the
maximum exploitation of the available wind energy be
converted into mechanical energy. For the turbine power
to be determined as a function of the and m ,
equation (2) is re-written as the following equation in
order to obtain the wind speed (Nakamura et al., 2002;
Morimoto et al., 2005; Shirazi et al., 2009; Pucci and
Cirrincione, 2011).
m R
Vw (5)

By substituting (5) into (1), the expression yields
5 m
3
1
Pm R 3 C p (6)
2 Figure 4 The torque-speed characteristic curve for a series
If the rotor is running at the opt , it will also run at of wind speeds

the C p max . Thus, by replacing opt and

Lookup Table
C p C p max into (6), yields the following expression: Generator
i..a Speed
1 C ii..b
Pm-opt R 5 P3max m3 K poptm3 (7) Controller Wind Energy System
opt
. . Optimal
2 . . Power
Considering that Pm mTm , the Tm can be plotted as in
. .
x.z
Fig. 4 and re-arranged as follows:
1 C
Tm-opt R 5 P3max m2 K optm2 (8)
2 opt Turbine Power
In general, this method is simple, very fast and efficient.
However, the efficiency is lower as compared to the TSR
control, since it does not measure the wind speed directly,
which wind changes are not reflected instantaneously and
significantly on the reference signal (Raza et al., 2010). Figure 5 The block diagram of a wind energy with the
power signal feedback control technique
C. Power Signal Feedback Control
The block diagram of a wind energy system with power D. Perturbation and Observation Control
signal feedback (PSF) control is shown in Fig. 5. Unlike
the OT control, in this method the reference maximum The perturbation and observation (P&O) or hill-climb
power curves of the wind turbine, Fig. 3, should be searching (HCS) method is a mathematical optimization
obtained first from the experimental results. Then, the data technique used to search for the local maxima points of a
points for maximum output power and the corresponding given function. It is widely used in wind energy systems to
wind turbine speed must be recorded in a lookup table get the optimal operating point that maximizes the
(Tan and Islam 2004; Barakati 2008; Barakati et al., extracted energy. This method is based on perturbing a
2009). Instead of using the wind turbine maximum power control parameter in small step-size and observing the
versus shaft speed curve in obtaining the lookup table as resulting changes in the target function, until the slope
(Barakati 2008), the maximum DC output power and the becomes zero. As shown in Fig. 6, if the operating point is
DC-link voltage were taken as input and output of the to the left of the peak point, the controller must move the

35
operating point to the right to be closer for the MPP, and solved by using neural network technique to estimate the
vice versa if the operating point is on the other side. In wind speed depending on actual machine torque and speed
literature, some authors perturb the rotational speed and (Lee et al. 2009; Pucci and Cirrincione 2011). The
observe the mechanical power. There are also others who proposed control structure, Wilcoxon radial basis function
monitor the electrical output power of the generator and network (WRBFN)-based with HCS MPPT strategy and
perturb the inverter input voltage (Quincy and Liuchen modified particle swarm optimization (MPSO) algorithm,
2004), or one of the variables of the converter; namely in (Lin and Hong 2010) diminish the effect of the wind
duty cycle, d (Koutroulis and Kalaitzakis 2006; Patsios et turbine inertia on HCS method performance.
al., 2009; Hua and Cheng 2010), input current, I in
Hybrid method is the combination of two methods from
(Neammanee et al., 2006), or input voltage, Vin (Kesraoui the aforementioned ones; to exploit the advantages of one
et al., 2010). In methods that used electrical power technique to overcome the disadvantage of the other. An
measurement, the mechanical sensors are not required, and example of this method is that in (Kazmi et al. 2011)
thus, they are more reliable and cost less. where OTC method is merged with HCS to solve the two
problems associated with the conventional HCS, the
Since the P&O method does not need a prior knowledge of speed-efficiency trade-off and the wrong directionality
the wind turbine characteristic curve, it is independent, under rapid wind change. Another example is combining
simple and flexible. However, it fails to reach the PSF control and HCS in (Quincy and Liuchen 2004) to
maximum power points under rapid wind variations if it is develop a sensor less and flexible method which is also
used for large and medium inertia wind turbines. applicable to all wind turbine levels.
Moreover, the problem of choosing an appropriate step-
size is not an easy task; where larger step-size means 4. SIMULATION RESULTS AND DISCUSSIONS
faster response and less efficiency, on the other hand, The performance of three MPPT control methods has been
smaller step-size improves the efficiency but slows the simulated and compared using the MATLAB/Simulink
convergence speed (Ching-Tsai and Yu-Ling 2010; Hong simulation package. The studied MPPT methods are:
and Lee 2010; Kazmi et al. 2011). OTC, P&O of the duty cycle of the boost converter, and
P&O of the input voltage of the boost converter. All the
simulations were carried out with system parameters as
(Mena, 2007). The load resistance, R is 20 for all
simulations. The step-sizes in P&O of the duty cycle and
3
the input voltage were fixed at 0.5 10 and 0.001,
respectively. The obtained performance with the different
methods is shown in Fig. 8 and the results are also
summarized in Table 1. According to the plot and results
analysis, the OTC controller is the fastest in achieving the
steady-state and also in the recovery time upon wind speed
change. In addition, the OTC method can reach the highest
value of C p and maintained the same value after the wind
speed change. It is followed by the P&O in input voltage
method, which took approximately double the time to
reach the steady-state, with the C p average of 0.4607. The
Figure 6 Wind turbine output power and torque slowest and less efficient one is the P&O in duty-cycle
characteristics with MPP tracking process (Neammanee et method, where the response time is eight times the first
al., 2006) method, 0.02142. After being 0.46 before the wind speed
step change, C p max decreased to 0.42 when the step
E. Other methods
change occurred. Since the used perturbation and
Many of the problems associated with the aforementioned observation methods are the conventional ones, with a
methods have been solved by means of artificial
fixed step-size, the ripples of C p changed under wind
intelligence control and hybrid methods. According to
(Simoes et al., 1997), fuzzy logic control methods have the speed variations. In Fig. 9, the generators output power
advantages of fast convergence, parameter insensitivity, for each method is depicted. While the generators output
and accepting noisy and inaccurate signals. They can also power for the first two methods stabilized at the same
be used to obtain an optimal step size for conventional time, 0.025 sec., it needed 0.175 sec more time for the
HCS method, as in (Trinh and Lee 2010). Wind speed third one. Taking the maximum mechanical input energy
measurement and its associated drawbacks have been of the generator as a reference and measuring the electrical
36
energy output of the generator under the selected methods,
the efficiencies can be calculated, as listed in Table 1.

Table 1 Simulated Results: Power Coefficient Average

Values, Response Times, Recovery Times; Energy and
Efficiency
Respo Recov
Efficienc
nse ery Energy
Method Media y
time time (W)
n (%)
(sec.) (sec.)
Max.
theoretical
0.48 -- -- 734.5 --
value
(reference)
0.0248
OTC 0.4789 0.0006 665.9 90.66
8 (b)
P&O of
0.4607 0.053 0.0014 645.9 87.94
input voltage
P&O of
0.3956 0.2142 0.022 597.4 81.33
duty-cycle

(c)
Figure 8 The power coefficient with: (a) OTC method
(b) P&O of input voltage (c) P&O of duty cycle
Figure 7 The wind speed

(a)
(a)

37
 REFERENCES
Mahdi, A.J., Tang, W.H. et al. 2010. A Comparative
study on variable-speed operations of a wind
generation system using vector control.
International Conference On Renewable Energies
And Power Quality, The 10th International
Conference On Renewable Energies And Power
Quality,University of Granada.Granada, Spain.
Barakati, S.M. 2008. Modeling and Controller Design
of a Wind Energy Conversion System Including
Matrix Converter. Electrical and Computer
Engineering. Waterloo, Ontario, Canada, University
of Waterloo: 265.
Barakati, S.M., Kazerani, M. et al. 2009. Maximum
Power Tracking Control for a Wind Turbine System
(b) Including a Matrix Converter, IEEE Transactions
on Energy Conversion 24(3): 705-713.
Baroudi, J.A., Dinavahi, V., et al. 2007. A review of
power converter topologies for wind generators.
Renewable Energy 32(14): 2369-2385.
Brahmi, J., Krichen, L., et al. 2009. A comparative
study between three sensorless control strategies for
PMSG in wind energy conversion system, Applied
Energy 86(9): 1565-1573.
Carranza, O., Garcer, G., et al. 2010. Peak current
mode control of three-phase boost rectifiers in
discontinuous conduction mode for small wind
power generators, Applied Energy 87(8): 2728-
2736.
Ching-Tsai, P. and Yu-Ling, J. 2010. A Novel
Sensorless MPPT Controller for a High-Efficiency
Microscale Wind Power Generation System, IEEE
Transactions on Energy Conversion 25(1): 207-216.
(c)
Freris, L.L. 1990. Wind Energy Conversion System.
Figure 9 The output power response produced by the
London, U.K., Prentice Hall.
PMSG generator with : (a) OTC method (b) P&O of input
Grimble, M.J. and Johnson, M.A. 2008. Optimal
voltage (c) P&O of duty cycle
Control of Wind Energy Systems.
Hong, M.K. and Lee, H.H. 2010. Adaptive maximum
5. CONCLUSION
power point tracking algorithm for variable speed
This paper discusses and reviews the available MPPT
wind power systems, Proceedings of the 2010
algorithms. In addition, simulation and comparison of
international conference on Life system modeling
selected three control methods in terms of the efficiency
and and intelligent computing, and 2010
and speed of response were made. Simulation results
international conference on Intelligent computing
demonstrate superiority of OTC method; where it obtained
for sustainable energy and environment: Part I.
the maximum average value of C p and held it at its Wuxi, China, Springer-Verlag: 380-388.
maximum even with wind speed change. Nevertheless, its Hua, A.C.C. and Cheng, B.C.H. 2010. Design and
dependency on the wind turbine characteristics makes it implementation of power converters for wind
inflexible. On the other hand, P&O method is flexible and energy conversion system, Power Electronics
simple in implementation, but it is less efficient and has a Conference (IPEC), 2010 International.
difficulty in determining the optimum step-size. Hui, J. and Bakhshai, A. 2008. A new adaptive control
Comparing the perturbation in duty cycle, perturbation of algorithm for maximum power point tracking for
the input voltage to get a reference voltage is better, as wind energy conversion systems, Power Electronics
there is a controller to force the input voltage to track the Specialists Conference, 2008. PESC 2008. IEEE.
reference. Finding out an adaptive step-size algorithm and Kawale, Y.M. and Dutt, S. 2009. Comparative Study of
combining two or more of the available methods will Converter Topologies Used for PMSG Based Wind
improve the performance and overcome some of the Power Generation, Computer and Electrical
obstacles of the current methods.
38
 Engineering, 2009, ICCEE '09. Second Medium-Sized Fixed Pitch Wind Turbines, Control,
International Conference on. Automation, Robotics and Vision, 2006, ICARCV
Kazmi, S.M.R., Goto, H. et al. 2011. A Novel '06, 9th International Conference on.
Algorithm for Fast and Efficient Speed-Sensorless Patel, M.R. 1999. Wind and Solar Power Systems,
Maximum Power Point Tracking in Wind Energy CRC Press.
Conversion Systems, Industrial Electronics, IEEE Patsios, C., Chaniotis, A., et al. 2009. A comparison of
Transactions on 58(1): 29-36. maximum-power-point tracking control techniques
Kesraoui, M., Korichi, N., et al. 2010. Maximum power for low-power variable-speed wind generators. 8th
point tracker of wind energy conversion system, International Symposium on Advanced
Renewable Energy 2010. Electromechanical Motion Systems & Electric
Koutroulis, E. and Kalaitzakis, K.. 2006. Design of a Drives Joint Symposium.
maximum power tracking system for wind-energy- Pucci, M. and Cirrincione, M. 2011. Neural MPPT
conversion applications, IEEE Transactions on Control of Wind Generators With Induction
Industrial Electronics 53(2): 486-494. Machines Without Speed Sensors, IEEE
Lee, C.Y., Shen, Y.X., et al. 2009. Neural Networks Transactions on Industrial Electronics 58(1): 37-47.
and Particle Swarm Optimization Based MPPT for Quincy, W. and Liuchen, C. 2004. An intelligent
Small Wind Power Generator, World Academy of maximum power extraction algorithm for inverter-
Science, Engineering and Technology 60 2009: 17- based variable speed wind turbine systems, IEEE
23. Transactions on Power Electronics 19(5): 1242-
Li, S., Haskew, T. A., et al. 2010. Conventional and 1249.
novel control designs for direct driven PMSG wind Raza K..S.M., Goto, H., et al. 2010. Review and critical
turbines, Electric Power Systems Research 80(3): analysis of the research papers published till date on
328-338. maximum power point tracking in wind energy
Lin, W.M. and Hong, C.M. 2010. Intelligent approach conversion system, Energy Conversion Congress
to maximum power point tracking control strategy and Exposition (ECCE), 2010 IEEE.
for variable-speed wind turbine generation system, Shirazi, M., Viki, A.H., et al. 2009. A comparative
Energy 35(6): 2440-2447. study of maximum power extraction strategies in
Mena, L. H. E. 2007. Maximum power tracking control PMSG wind turbine system, Electrical Power &
scheme for wind generator systems, Texas A&M Energy Conference (EPEC), 2009 IEEE.
University. Simoes, M.G., Bose, B.K., et al. 1997. Design and
Mirecki, A., Roboam, X., et al. 2004. Comparative performance evaluation of a fuzzy-logic-based
study of maximum power strategy in wind turbines, variable-speed wind generation system, IEEE
IEEE International Symposium on Industrial Transactions on Industry Applications 33(4): 956-
Electronics. 965.
Molina, M.G., Santos, E.C., et al. 2010. Advanced Tan, K. and Islam, S. 2004. Optimum control strategies
power conditioning system for grid integration of in energy conversion of PMSG wind turbine system
direct-driven PMSG wind turbines, Energy without mechanical sensors, IEEE Transactions on
Conversion Congress and Exposition (ECCE), 2010 Energy Conversion 19(2): 392-399.
IEEE. Trinh, Q.N. and Lee, H.H. 2010. Fuzzy Logic
Morimoto, S., Nakayama, H., et al. 2005. Sensorless Controller for Maximum Power Tracking in PMSG-
output maximization control for variable-speed Based Wind Power Systems. Advanced Intelligent
wind generation system using IPMSG, IEEE Computing Theories and Applications. With
Transactions on Industry Applications 41(1): 60-67. Aspects of Artificial Intelligence. D.-S. Huang, X.
Muyeen, S.M., Takahashi, R., et al. 2010. A Variable Zhang, C. Reyes Garca and L. Zhang, Springer
Speed Wind Turbine Control Strategy to Meet Berlin / Heidelberg. 6216: 543-553.
Wind Farm Grid Code Requirements, IEEE Wang, Q. 2003. Maximum wind energy extraction
Transactions on Power Systems 25(1): 331-340. strategies using power electronic converters.
Nakamura, T., Morimoto, S., et al. (2002). Optimum Canada, University of New Brunswick.
control of IPMSG for wind generation system. Zhe, C., Guerrero, J.M., et al. 2009. A Review of the
Power Conversion Conference, 2002. PCC Osaka State of the Art of Power Electronics for Wind
2002. Proceedings of the. Turbines, Power Electronics, IEEE Transactions on
Neammanee, B., Sirisumranukul, S., et al. 2006. 24(8): 1859-1875.
Control Performance Analysis of Feedforward and
Maximum Peak Power Tracking for Small-and

39