International Journal of Power Electronics and Drive Systems (IJPEDS)

Vol. 14, No. 3, September 2023, pp. 1684~1693

ISSN: 2088-8694, DOI: 10.11591/ijpeds.v14.i3.pp1684-1693  1684

Optimized fuzzy PI controller for variable speed wind turbine

using DE algorithm

Khaddouj Ben Meziane1, Faiza Dib2, Nabil Benaya2, Ismail Boumhidi3

1
Department of Engineering, Higher Institute of Engineering and Business, Fez, Morroco
2
Department of Physics, Faculty of Sciences and Techniques, Abdelmalek Essaadi University, Tetouan, Morroco
3
Department of Physics, Faculty of Sciences, Sidi Mohamed Ben Abdellah University, Fez, Morroco

Article Info ABSTRACT

Article history: To design an optimal fuzzy proportional-integral (PI) controller for the
variable speed wind turbine systems, a new differential evolution (DE)
Received Sep 19, 2022 algorithm approach is developed in this paper. We have investigated a fuzzy
Revised Dec 28, 2022 PI controller, in which fuzzy rules are applied to adapt the parameters of the
Accepted Jan 7, 2023 PI controller founded on the error and its first-time derivative. The fuzzy PI
controller's inputs and outputs are tuned using the DE optimization method.
The superiority of the suggested (DE fuzzy PI) controller has been proved by
Keywords: comparing the results with (fuzzy PI) and only the PI controller applied to
the wind turbine system. In comparison to a fuzzy controller with parameters
DE algorithm selected by a human operator, the numerical validation results of the
Fuzzy logic suggested approach (DE fuzzy PI) have shown good performance in terms of
PI controller robustness, pursuit, and response time.
Stability
Wind energy This is an open access article under the CC BY-SA license.
Wind turbine system

Corresponding Author:
Khaddouj Ben Meziane
Department of Engineering, Higher Institute of Engineering and Business
Fez, Morroco
Email: [email protected]

1. INTRODUCTION
The fastest production energy source is Wind energy is a clean, renewable energy source that has
been used for centuries. Wind turbine systems generate electricity for utilities, homeowners, and remote
villages [1]. A considerable investment was made in the scientific field has been made in the development
and enhancement of wind turbines. These wind systems could serve as an innovative and cost-effective
replacement for many energy sources [2].
Variable speed wind turbines are becoming more prevalent than conventional fixed-speed
turbines [3]. This type of system is characterized by its capacity to increase energy recovery, decrease
disturbance, and enhance the quality of the energy produced [4]. It can also alter the shaft speed as the wind
speed varies. Power systems are incorporating wind energy more and more, the analysis of the stability of
wind turbine systems is becoming a significant problem for planners and for operators of power systems.
Due to the variable-speed wind turbine system's (VS-WTS) increased complexity and nonlinearity,
an effective and convenient control mechanism is required [5]. However, it is challenging to guarantee
effective stability and effective control performance when utilizing linear and classical control methods;
Alternatively, recent technological improvements have produced intelligent controls for nonlinear, complex
systems ,using a variety of techniques, which include optimal feedback control [6], dynamic control founded
on nonlinear prediction models [7], control using T-S systems [8], wide-area fuzzy controller with time
delay [9], and sliding mode control by using type-2 fuzzy system [10].

Journal homepage: http://ijpeds.iaescore.com

Int J Pow Elec & Dri Syst ISSN: 2088-8694  1685

To improve the system's performance and efficiency by employing traditional controls such as the
PID controllers. It's all about finding the right balance. Some programs will accept an overrun to speed up
stabilization time; therefore, it all depends on the requirements. Each of the parameters (KP, KI, and KD) has
an effect on the system's response. KP and KI must be reduced in aim to reduce the static error. If KP or KI
declines or KD increases, the overshoot is lowered. If KP or KI increases or KD drops, the rising time
lowers [11]. However, implementing this standard type of control (PID) provides several challenges in
assuring robust performance due to the numerous issues in the wind turbine system (WTS), such as
uncertainties, nonlinearity of the WTS, parameter change, and unidentified disturbances [12].
The main goal of this research is to create an enhanced intelligent technique based on the
fuzzification of a PI controller (FLC-PI) and optimization techniques using the DE algorithm to optimize the
inputs and outputs of the fuzzy controller. The goal of this technique is to develop a reliable a powerful and
effective controller for rotor speed tracking to ensure maximizing wind energy capture. Fuzzy logic has been
shown to be a powerful control tool. It enables the creation of an intelligent controller using linguistic rules
without knowledge of the plant's mathematical model. The use of fuzzy logic to establish PI controllers
considered as an effective approach that surpasses traditional PI, particularly in the presence of uncertainties,
because fuzzy logic does not require a formal model and may include expert knowledge and experience to
achieve an intelligent control [13]. Many studies have shown that by adjusting the controller's parameters,
The fuzzy logic controller can manage any operating point changes, but there is no explicit mathematical
method for selecting the right fuzzy parameters, such as the membership function, input-output scaling
factors, and the rules [14]. These settings are chosen using the trial technique, which means they are not ideal
and have an effect on the fuzzy logic controller's performance. Many optimization methods, such as the
genetic algorithm [15] and particle swarm optimization [16], have been integrated with the fuzzy logic
controller. The DE technique is widely used in scientific and professional domains for tackling the bulk of
optimization issues [17]. This algorithm employs real number coding as part of its stochastic population-
based search strategy. It is a straightforward, dependable, and strong technique for global optimization. In
this study, the fuzzy PI controller's input and output parameters are optimally tuned applying the differential
evolution algorithm (DE), which improves system performance and significantly decreases tracking errors.
In the face of uncertainties and variations in the parameters, the efficiency of the suggested approach
(DE-fuzzy-PI) is confirmed and compared to the fuzzy PI and the traditional PI. The simulation results
demonstrated that the suggested technique is robust and superior, as a result, oscillations were dampened
more effectively, and the relevant data were tracked accurately and quickly. The following is how the paper
is structured. Section 2 gives a description of the mathematical model of a WTS, section 3 presents the
proposed (DE-fuzzy-PI) control approach, and section 4 offers simulations to show the effects of the
suggested approaches. Section 5 has the conclusion.

2. THE MATHEMATICAL MODEL OF WTS

The following [12], [18] the aerodynamic power that the WTS is capable of producing:
1
𝑃υ = ρAν3 (1)
2

where ρ is the density of air (kg/m3), A = πR2 and ν is the speed of wind system (m/s). The following
equation [19] represents the power that the rotor is able to capture:

Pa = Cp (λ, β)Pυ (2)

1
Pa = ρπR2 Cp (λ, β)ν3 (3)
2

where, 𝐂𝐩 is the power performance coefficient, and λ indicates the tip speed ratio.
The aerodynamic torque is produced when the aerodynamic power is transformed into mechanical
power, as seen in [4]:

1 Cp (λ,β)
Ta = ρπR3 ν2 (4)
2 λ

A general model that may be used for wind turbines is the two-mass model given in Figure 1.

Optimized fuzzy PI controller for variable speed wind turbine using … (Khaddouj Ben Meziane)
1686  ISSN:2088-8694

Figure 1. The two-mass WTS model [20]

We may use the following to represent the WTS:

ω̇ t c11 c12 c13 ωt b11 b12

ω̇ c
[ g ] = [ 21 c22 c23 ] [ωg ] + [b21 ] Ta + [b22 ] Tem (5)
Ṫls c31 c32 c33 Tls b31 b32

where:

Kr 1 Kg 1 K ls K r
c11 = − ; c12 = 0; c13 = − ; c21 = 0; c22 = − ; c23 = ; c31 = (Bls − )
Jr Jr Jg ng Jg Jr
1 Kls Kg Jr +n2
g Jg 1 1 Kls
c32 = ( − Bls ); c33 = −K ls ( ) ; b11 = ; b12 = 0; b22 = − ; b31 = ;
ng Jg n2
g Jg Jr Jr Jg Jr
K ls
b21 = 0; b32 =
ng Jg

we define x1 = ωt , x2 = ω̇ t , x3 = ωg , and u = Tem , the set of equations in (6) may be reformulated in the
state space as follows [12]:

ẋ 1 = x2
ẋ 2 = f(x, t) + g(x, t). u + ξ(x, t)
c (6)
ẋ 3 = c22 x3 + 23 (x2 − c11 x1 − b11 Ta ) + b22 u(x, t)
c13
{ y = x1

where x = [x1 , x2 , x3 ]T ϵR3 is the system's state vector, which is supposed to be measured, uϵR and yϵR are
respectively input of system, the output of system, and f(x, t), g(x, t) are the system's nominal form provided
by (7) and (8):

2
𝑓(𝑥, 𝑡) = 𝑥1 (𝑐11 + 𝑐13 𝑐31 ) + 𝑇𝑙𝑠 (𝑐11 𝑐13 + 𝑐13 𝑐33 ) + 𝑇𝑎 (𝑐11 𝑏11 + 𝑐13 𝑏31 ) + 𝑏11 𝑇𝑎̇ (7)

𝑔(𝑥, 𝑡) = 𝑐13 𝑏32 (8)

ξtx, t) are the external disturbances and the uncertainties.

3. PROPOSED CONTROLLER DESING

3.1. PI controller for variable speed wind turbine system (PI-VS-WTS)
The designed controller's main objectives are to minimize errors, get the elimination of oscillations,
and keep track of the desired rotor speed 𝜔𝑡𝑜𝑝𝑡 . Let's define tracking errors as follows:

𝑒(𝑥, 𝑡) = 𝜔𝑡 (𝑥, 𝑡) − 𝜔𝑡𝑜𝑝𝑡 (𝑥, 𝑡) (9)

in terms of the error, the conventional PI controller is given by [9]:

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1684-1693
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1687

𝑈𝑃𝐼 (𝑥, 𝑡) = 𝐾𝑃 𝑒(𝑥, 𝑡) + 𝐾𝐼 ∫ 𝑒(𝑥, 𝑡)𝑑𝑡 (10)

𝐾𝑃 , 𝐾𝐼 represent the classical value of proportional gain, and integral gain. The (11) presents the structure of
the traditional control:

𝑢(𝑥, 𝑡) = 𝑢𝑒𝑞 (𝑥, 𝑡) + 𝑢𝑃𝐼 (𝑥, 𝑡) (11)

The traditional PI or PID controllers are frequently ineffective and unsatisfactory, especially when higher
level operations involve a temporal delay, or the systems have considerable uncertainty [12]. The usage of
the Fuzzy logic System (FLS) was proposed in this study to increase the performance and the robustness of
the classical PI controller and to adapt any adjustments to system parameters or operational conditions.

3.2. Fuzzy PI controller for variable speed wind turbine system (fuzzy PI-VS-WTS)
The wind turbine system's (WTS) rotor speed oscillations can be damped using fuzzy logic based on
PI Controller. The input signal could be the error of the generator's rotor speed deviation e(t) and its
derivative 𝑒̇ (𝑡). Disturbances in a wind turbine system led to electromechanical oscillations. To keep the
system stable, these oscillations must be adequately dampened. The Fuzzy PI Controller's output signal is
employed as an additional input UFuzzy-PI in the VS-WTS control law. The two signals' input ranges
{𝑒(𝑡), ∆𝑒(𝑡)} are from [-0.01, 0.01] are given in Figure 2.

Figure 2. Membership functions for input parameters

The input of fuzzy variables is created for the rule base [21]:

{𝑒(𝑡), ∆𝑒(𝑡)} = {NB, NM, NS, Z, PS, PM, PB}

The fuzzy logic controller based on PI replaces the traditional PI controller in this article. The fuzzy logic
toolbox in the MATLAB software is used to create the FLC. For the output and the two inputs U fuzzy-PI, the
Gaussian membership function is employed. For the controller's best performance, a total of 49 rule bases are
built, as given in Table 1.

Table 1. The fuzzy controller's rule set [22]

de
NB NM NS Z PS PM PB
e
NB NB NB NB NM NM NS Z
NM NB NB NM NM NS Z PS
NS NB NM NS NS Z PS PM
Z NM NM NS Z PS PM PM
PS NM NS Z PS PS PM PB
PM NS Z PS PM PM PB PB
PB Z PS PM PM PB PB PB

Optimized fuzzy PI controller for variable speed wind turbine using … (Khaddouj Ben Meziane)
1688  ISSN:2088-8694

The 𝜃1,.., 𝜃𝑛 represent the centroids of M membership functions which are allocated to the controller
output 𝑈𝐹𝑢𝑧𝑧𝑦−𝑃𝐼 . Thus, the controller's output for M rules is [23]:

∑𝑀
𝑖=1 𝛽𝑖 𝜃𝑖
𝑈𝐹𝑢𝑧𝑧𝑦−𝑃𝐼 = 𝐾𝑜𝑢𝑡 ∑𝑀
= 𝐾𝑜𝑢𝑡 𝜃 𝑇 𝜉 (12)
𝑖=1 𝛽𝑖

𝛽
where 𝜉 = [𝜉1 , 𝜉2 , … , 𝜉𝑀 ]; 𝜉𝑖 = ∑𝑀 𝑖 and 𝜃 𝑇 = [𝜃1 … 𝜃𝑀 ]; 𝛽𝑖 is the degree of activation of each 𝑖 𝑡ℎ rule.
𝑖=1 𝛽𝑖
Fuzzy sets can effectively cope with model uncertainties, but choosing appropriate membership function
parameters is difficult. The DE algorithm is employed in this paper to identify the best settings for the fuzzy
system's membership functions.

3.3. Differential evolution algorithm

The differential evolution algorithm (DE) is one of the simplest and most efficient evolutionary
algorithms. Storn and Price first proposed the notion in 1997 [17]. This technique belongs to the class of
stochastic meta-heuristic global optimization methods, and it is distinguished by its ease of use, resilience,
and fast convergence, making it an ideal tool for solving challenging problems. Starting with a randomly
generated initial population, the algorithm's (DE) goal is to enhance it through an evolutionary loop that is
separated into three basic operations: selection, mutation, and crossing.

3.3.1. Initialization
By assigning each variable a random value, the population of solution candidates is created at the first
stage of the optimization DE algorithm. These values, which must lie within the user-specified According to the
problem's nature, lower and upper limits for the control variables can be generated as (13) [24]:

xi,j,0 = xi,min + randj (0,1) × (xj,max − xi,min ) (13)

3.3.2. Mutation
A mutation operator is in charge of adding new parameters to the population. To accomplish so, the
mutation operator disrupts a random vector to produce mutant vectors 𝑉𝑖,𝐺 . 𝑋𝑟 1 unlike the two vectors
𝑖
(𝑋𝑟 2 and 𝑋𝑟 3 ) according to the (14), a random selection made from the population:
𝑖 𝑖

Vi,G = Xr1i,G + F. (Xr2i,G − Xr3i,G ) (14)

The mutation factor F is defined in the interval [0 1] and is used to adjust the size of the differential variation
(Xr2i,G − Xr3i ,G ) to improve convergence.

3.3.3. Crossover
Build a test individualUj,i,G using the crossover operator out of a combination of a mutant Vj,i,G and a
target individual Xj,i,G in accordance with probability distributions. To increase populationdiversity, the
crossover operator is utilized, while also preventing the population from converging to a local minimum.
This is accomplished by following the steps outlined in the equation [25]:

Vj,i,G if randj ≤ CR or j = jrand

Uj,i,G = { (15)
Xj,i,G if randj > 𝐶𝑅𝑜𝑟𝑗 ≠ jrand

CR is the crossover probability and rand𝑗 ∈ [0,1]is the jth random number index.

3.3.4. Selection
By comparing mainly, the values of the two competitors' objective functions, the selection operator
enables you to choose the most suitable individual for the population of the following generation from the
target individual Xi,G and the test individual Ui,G . The selection process is described in the following manner:

Ui,G if J(Ui,G ) < 𝐽(Xi,G )

Xi,G+1 = { (16)
Xi,G otherwiswe

J(X) is the objective function to be minimized and i ∈ [1, Np ].The flowchart in Figure 3 depicts the several
stages of the differential evolution algorithm.

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1684-1693
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1689

Figure 3. Flow chart of the differential evolution algorithm

3.4. Design of proposed optimized fuzzy PI controller by differential evolution algorithm (DE fuzzy PI)
This paper's key contribution is to use the differential evolution method (DE) to optimize the fuzzy
PI controller's scaling factor parameters, which improves system performance and considerably lowers
tracking errors. Finally, we obtain the proposed control's structure (DE-fuzzy-PI), which is given by (17):

𝑢(𝑥, 𝑡) = 𝑢𝑒𝑞 (𝑥, 𝑡) + 𝑢𝐷𝐸−𝐹𝑢𝑧𝑧𝑦−𝑃𝐼 (𝑥, 𝑡) (17)

𝐹(𝑥,𝑡)
𝑢(𝑥, 𝑡) = − + 𝑈𝐷𝐸−𝐹𝑢𝑧𝑧𝑦−𝑃𝐼 (𝑥, 𝑡) (18)
𝐺(𝑥,𝑡)

where F(x, t) and G(x, t) are the nominal representation of the system WTS. The two inputs (the error and its
derivative) and the output of the optimized fuzzy logic control based on PI controller, which is shown in
Figure 4. The suggested design process optimizes the fuzzy PI controller's kin1, kin2, and k out input-output
scaling factors. The membership functions in this instance are unchanged over a single universe of discourse
in the range [-1, 1].

DE algorithm

yr
+ Kin1
Fuzzy logic
- y
based on PI Wind turbine
- controller Kout system
Kin2

Figure 4. DE-Fuzzy-PI controller block diagram for WTS

Optimized fuzzy PI controller for variable speed wind turbine using … (Khaddouj Ben Meziane)
1690  ISSN:2088-8694

4. RESULTS AND DISCUSSION

The system under study is simulated by the MATLAB software presented in Figure 4 to prove the
efficacy of the proposed controllers in this work. This essay chooses one type of wind turbine from the two-
mass model. Table 2 [26] lists the wind turbines specifications. To verify the effectiveness and resilience of
the created and suggested approach (DE-fuzzy-PI), simulation experiments in MATLAB software are
suggested. We take into account the electromagnetic torque with a 2000 N.m continuous disturbance. Table 3
lists the differential evolution algorithm's parameters that were employed in this study.

Table 2. Parameters of VS of WTS Table 3. Factors of the DE algorithm

Parameters Value Parameters Value Parameters Value
𝑅 21.65 𝑚 𝐵𝑙𝑠 2.691 × 105 𝑁. 𝑚/𝑟𝑎𝑑 NP 30
𝜌 1.29 𝑘𝑔/𝑚3 𝐾𝑟 27.36 𝑁. 𝑚/𝑟𝑎𝑑/𝑠 CR 0.8
𝐽𝑟 34.4 𝑘𝑔. 𝑚2 𝐾𝑔 0.2 𝑁. 𝑚/𝑟𝑎𝑑/𝑠 F 0.6
𝐾𝑙𝑠 9500 𝑁. 𝑚/𝑟𝑎𝑑/𝑠 𝑛𝑔 43.165 G 30
D 3

- Scenario 1
The first scenario's application of the wind speed profile with variations given in Figure 5. Based on
Figures 6, 7, and 8, it can be noticed that the optimized fuzzy PI by the DE algorithm gives better
performance compared to the fuzzy PI and PI controller, the proposed approach leads to efficient
performance compared to the other controllers. The effectiveness and robustness of the suggested controller
device (DE-fuzzy-PI) can therefore be inferred from this simulation's results. The outcomes show that this
clever controller is able to reduce tracking errors and eliminating oscillations. The wind turbine's rotor speed
can be intelligently controlled in order to produce better results and faster response times when compared to
other controllers.

Figure 5. Profile of wind speed Figure 6. Rotor speed's reaction

Figure 7. The reaction of aerodynamic torque Figure 8. Tracking error comparison

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1684-1693
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1691

- Scenario 2
The profile of wind speed with substantial variance used in this situation after 20 seconds given in
Figure 9. With the elimination of steady-state oscillations at the speed level of the rotor and the aerodynamic
torque, the results obtained given in Figures 10, 11 and 12 show flawless static and dynamic performance,
better tracking of the regulated setpoint (speed), and a preferable response time supplied by the DE-Fuzzy PI
controller. In addition, the error converges to zero after 3 seconds, meaning that oscillations are fully
dampened in comparison to the other two control strategies (fuzzy PI and PI), or oscillations are still present
and the error converges to zero after 15 seconds.

Figure 9. Profile of wind speed Figure 10. Rotor speed's reaction

Figure 11. The reaction of aerodynamic torque Figure 12. Tracking error comparison

The effectiveness and reliability of the suggested controller (DE-fuzzy-PI) can be seen from this
simulation instance, and the results show that it is able to minimize tracking errors and remove oscillations
when compared to other controllers. The suggested controller performs admirably when it comes to removing
disturbances and pursuing the appropriate rotational speed. Even when the wind profile is changed, the
suggested controller (DE-fuzzy-PI) swiftly converges to the ideal state with reduced oscillation. The response's
output of (DE-fuzzy-PI) reaches the appropriate values more quickly than other controllers (fuzzy-PI),
according to all of these figures (PI). As a result, the simulation results demonstrate that the described approach
enables response time minimization with satisfactory convergence despite fluctuations in wind speed.

5. CONCLUSION
In the aim of maximizing the energy produced by the wind and enhancing the stability of the WTS,
the optimized fuzzy PI controller for a VS-WTS is introduced and applied in this work. The method entails
applying the DE algorithm to optimally tune the scaling factors of a fuzzy PI controller. The three approaches
(DE-fuzzy-P), (fuzzy-PI), and (PI) were compared in order to show that the proposed method (DE-fuzzy-PI)
offered the greatest performance. Regarding robustness in the face of change in the system's parameters, as

Optimized fuzzy PI controller for variable speed wind turbine using … (Khaddouj Ben Meziane)
1692  ISSN:2088-8694

well as in regards of pursuing the desired values, since it guarantees a very high performance for speed. The
new method DE-fuzzy-PI gives a quicker response than normal PI, and simulation results show that it can
achieve the optimal tracking results with the least amount of tracking error and oscillations.

REFERENCES
[1] S. Ribbing, and G.Xydis, “Renewable energy at home: a look into purchasing a wind turbine for home use—the cost of blindly
relying on one tool in decision making,” Clean Technol., 3, pp. 299–310, 2021, doi: 10.3390/cleantechnol3020017.
[2] S. Abolhosseini, A. Heshmati, and J. Altmann, “A review of renewable energy supply and energy efficiency technologies,” IZA
Discussion, No. 8145, April 2014, doi: 10.2139/ssrn.2432429.
[3] D. Devaraj, and R. Jeevajyothi, “Impact of fixed and variable speed wind turbine systems on power system voltage stability
enhancement,” IET Conference on Renewable Power Generation, pp. 1-9, 2011, doi: 10.1049/cp.2011.0199.
[4] Y. Mousavi, G. Bevan, I. B. Kucukdemiral, and A. Fekih, “Sliding mode control of wind energy conversion systems: Trends and
applications,” Renewable and Sustainable Energy Reviews, vol. 167, 2022, doi: 10.1016/j.rser.2022.112734.
[5] S. Simani, and P. Castaldi, “Robust control examples applied to a wind turbine simulated model,” Applied sciences, vol. 8, 2018,
doi: 10.3390/app8010029.
[6] R. Gaamouche, A.Redouane, I. El Harraki, B. Belhorma, and A. El Hasnaoui, “Optimal feedback control of nonlinear variable-
speed marine current turbine using a two-mass model,” Journal of Marine Science and Application, vol. 19, pp. 83–95, 2020, doi:
10.1007/s11804-020-00134-6.
[7] R. Saravanakumar, and D. Jena, “Nonlinear estimation and control of wind turbine,” IEEE International Conference on
Electronics, Computing and Communication Technologies, pp. 1-6, 2013, doi: 10.1109/CONECCT.2013.6469307.
[8] H. Moodi, and D. Bustan, “Wind turbine control using T-S systems with nonlinear consequent parts, ” Energy, vol. 172, pp.
922−931, 2019, doi: 10.1016/j.energy.2019.01.133.
[9] Y. Bostani, and S. Jalilzadeh, “A new approach based on wide-area fuzzy controller for damping of sub synchronous resonance in
power system including DFIG,” Journal of Operation and Automation in Power Engineering, vol. 11, no. 1, pp. 61−68, 2023, doi:
10.22098/JOAPE.2023.9565.1667.
[10] Z. Lahlou, K. Ben Meziane, and I. Boumhidi, “Sliding mode controller based on type-2 fuzzy logic PID for a variable speed wind
turbine,” Int. J. Syst.Assur. Eng.Manag., vol 10, pp. 543–551, 2019, doi: 10.1007/s13198-019-00767-z.
[11] O. Apata, and D.T.O. Oyedokun, “An overview of control techniques for wind turbine systems,” Scientific African, vol. 10, 2020,
doi: 10.1016/j.sciaf.2020.e00566.
[12] K. Ben Meziane, R. Naoual, F. Dib, and I. Boumhidi, “intelligent sliding mode control based on optimized pid by de algorithm for
wind turbine system,” The Fifth International Conference On Intelligent Computing in Data Sciences (ICDS), pp 1-6 , 2021, doi:
10.1109/ICDS53782.2021.9626743.
[13] S. Karad and R. Thakur, "Genetic Algorithm and Particle Swarm Optimization tuned Fractional Order Pitch Angle Control," 2021
International Conference on Computational Performance Evaluation (ComPE), Shillong, India, 2021, pp. 921-927, doi:
10.1109/ComPE53109.2021.9752437.
[14] F. Dib, I. Boumhidi, N. E. Akchioui, and N. Benaya, “Sliding mode control optimized by teaching learning-based optimization
algorithm for variable speed wind turbine system,” International Conference on Wireless Technologies, Embedded and Intelligent
Systems (WITS), pp. 1-6, 2019, doi: 10.1109/WITS.2019.8723804.
[15] E. A. Sid Ahmed, and A. Mohamed, “Fuzzy sliding mode control applied to a doubly fed induction generator for wind turbines,”
Turk. J. Electric. Eng.Comput. Sci., vol. 23, pp. 1673 – 1686, 2015, doi: 10.3906/elk-1404-64.
[16] H. J. Asl, and Y. Jungwon, “Power capture optimization of variable-speed wind turbines using an output feedback controller,”
Renew.Energy., vol. 86, pp. 517–525, 2016, doi: 10.1016/j.renene.2015.08.040.
[17] R. Storn, and K. Price, “Differential evolution- a simple and efficient heuristic for global optimization over continuous spaces,” J.
Global Optim., vol. 11, no. 4, pp. 341–359, 1997, doi: 10.1023/A:1008202821328.
[18] X. Kong,L. Ma, X. Liu, M.A. Abdelbaky, and Q. Wu, “Wind turbine control using nonlinear economic model predictive control
over all operating regions,” Energies, vol. 13, no. 1, 2020, doi: 10.3390/en13010184.
[19] Y. Mousavi, G. Bevan, I. B. Küçükdemiral, and A. Fekih, “Maximum power extraction from wind turbines using a fault-tolerant
fractional-order nonsingular terminal sliding mode controller,” Energies, vol. 14, no.18, 2021, doi: 10.3390/en14185887.
[20] L. Wang, L. Cao, and L. Zhao, “Non-linear tip speed ratio cascade control for variable speed high power wind turbines: a
backstepping approach,” IET Renew. Power Gener., vol. 12, no. 8, pp. 968−972, 2018, doi: 10.1049/iet-rpg.2017.0698.
[21] K. Eltag, M. S. Aslam, and R.Ullah, “Dynamic Stability Enhancement Using Fuzzy PID Control Technology for Power
System,”International Journal of Control, Automation and Systems 17(1) (2019) 234-242 http://dx.doi.org/10.1007/s12555-018-
0109-7.
[22] K. B. Meziane and I. Boumhidi, “Optimized type-2 fuzzy logic PSS combined with H_tracking control for the multi-machine
power system,” inEmbedded Systems and Artificial Intelligence,Advances in Intelligent Systems and Computing, vol. 1076, 2020,
doi: 10.1007/978-981-15-0947-6_19.
[23] F. Dib, N. El Akchioui, and I. Boumhidi, “Design of sliding mode control with optimized fuzzy PSS by differential evolution
algorithm for power system SMIB,” 5th International Conference on Optimization and Applications (ICOA), IEEE, 2019, doi:
10.1109/ICOA.2019.8727664.
[24] Z. Zhiqiang M. Zhang, T. Chen, and Z. Hong, “A new selection operator for differential evolution algorithm,” Knowledge-Based
Systems, vol. 226, pp. 1−13, 2021, doi: 10.1016/j.knosys.2021.107150.
[25] M. Georgioudakis, and V. Plevris, “A comparative study of differential evolution variants in constrained structural optimization,”
Frontiers in Built Environment, vol. 6, pp. 1−14, 2020, doi: 10.3389/fbuil.2020.00102.
[26] S. Orlando, G. Henrique, M. Antonio, and C. Adriano, “nonlinear control of the doubly-fed induction generator in wind power
systems,” Renew.Energy., vol. 35, no. 8, pp.1662–1670, 2010, doi: 10.1016/j.renene.2009.12.008.

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1684-1693
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1693

BIOGRAPHIES OF AUTHORS

Khaddouj Ben Meziane is a lecturer in the Engineering Department at the

Higher Institute of Engineering and Business (ISGA), Fez, MORROCO. She received her
Ph.D. degree in Signals, systems, and computing from Sidi Mohammed Ben Abdellah
University, Fez, MORROCO, in 2016. She has been an Assistant Professor at the Higher
Institute of Engineering and Business (ISGA), Fez, MORROCO since 2017. Her research
interests include control design, power systems, fuzzy systems, sliding mode control,
renewable energy, wind turbine systems, intelligent control, and optimization algorithm. She
can be contacted at email: [email protected].

Faiza Dib is a lecturer in the Department of physics at FSTH, Abdelmalek

Essaadi University, Tetouan, Morroco. She received her Ph.D. degree in Signals, systems, and
computing from Sidi Mohammed Ben Abdellah University, Fez, Morroco, in 2016. She has
been an Assistant Professor at the FSTH, Abdelmalek Essaadi University, Tetouan,
MORROCO since 2019. Her research interests include control design, power systems, robotic
systems, fuzzy systems, sliding mode control, wind turbine systems, intelligent control, and
optimization algorithm. She can be contacted at email: [email protected].

Nabil Benaya is a doctor-engineer in Electrical Engineering. After a long career

in industry, he is today a university professor, in the Department of Physics, at the Faculty of
Sciences and Techniques (FSTH) of Al Hoceima, Morocco. He obtained his PhD degree in
Automatic and Industrial Computing, at Abdelmalek Essaadi University in 2019. He is still
interested in Automation and Control systems, Fuzzy systems, Sliding mode control,
intelligent controller, renewable energy, and dynamic reliability. He can be contacted at email:
[email protected].

Ismail Boumhidi is a professor of electronics at the Faculty of Sciences, Fez

Morocco. He received from Sidi Mohamed ben Abdellah University, Faculty of sciences his
Phd degree in 1999. His research areas include adaptive robust control, multivariable
nonlinear systems, and fuzzy logic control with applications, sliding mode control, wind
turbine systems, renewable energy, intelligent control, and optimization algorithm. He can be
contacted at email: [email protected].

Optimized fuzzy PI controller for variable speed wind turbine using … (Khaddouj Ben Meziane)