International Journal of Power Electronics and Drive System (IJPEDS)

Vol. 9, No. 2, June 2018, pp. 571~578

ISSN: 2088-8694, DOI: 10.11591/ijpeds.v9.i2.pp571-578  571

Artificial Intelligence Control Applied in Wind Energy

Conversion System

Arama Fatima Zohra1, Bousserhane Ismail Khalil2, Laribi Slimane3, Sahli Youcef4, Mazari Benyounes5
1,2
Department of Electrical and Computer Engineering, University of Béchar Bp 417, Algeria
1
Department of Technical Sciences, University of d'ADRAR, Algeria
3,4
Unité de Recherche en Energies Renouvelables en Milieu Saharien, URERMS, Centre de Développement des Energies
Renouvelables, CDER, 01000, ADRAR, Algéria
5
University of Science and Technology Mohamed Boudiaf, Aljazair

Article Info ABSTRACT

Article history: The objective of this paper is to study the dynamic response of the wind
energy conversion system (WECS) based on the Doubly Fed Induction
Received Jan 4, 2018 Generator (DFIG). The DFIG rotor is connected to the grid via a converter.
Revised Mar 9, 2018 The active and reactive power control is realized by the DFIG rotor variables
Accepted Mar 23, 2018 control, using the field oriented control (FOC). The vector control of DFIG is
applied by the use of tow regulators PI and the neural network regulator
(NN). The generator mathematical model is implemented in Matlab/
Keyword: Simulink software to simulate a DFIG of 1.5 MW in order to show the
efficiency of the performances and robustness of the studied control systems.
Doubly fed induction The simulation obtained results shows that the robustness and response time
Field oriented control of the neural network regulator is better than those obtained by the PI
Generator PI regulator classical regulator.
The neural network regulator
Wind energy conversion system
Copyright © 2018 Institute of Advanced Engineering and Science.
All rights reserved.

Corresponding Author:
Arama Fatima Zohra,
Department of Electrical and Computer Engineering,
University of Béchar Bp 417, Algeria
Email: fatimaarama@yahoo.fr

1. INTRODUCTION
Wind power is one of the renewable energies which has a use growth in the world due to clean and
non-polluting nature of this technology [1] [2]. Several machines were used in WECS, but the range of wind
speed was limited in classical machines, the advanced technology created DFIG which solves this problem
and makes these systems more powerful. Several control methods of the DFIG are appeared, among them,
the vector control [3]. The principle of this control is to make DFIG similar to separate excitation DC
machine.
Zerzouri et al. [4]have tried to improve the performance of a DFIG wind energy conversion system,
they have proposed the vector control with the PI controller to decouple the active and reactive power of the
stator. They have used a single PI in each control loop, but this control has oscillations, exceedances, and the
decoupling is not fully maintained. Poitiers et al. [5] have used the RST polynomial controller, they have
found that the RST control is more robust than the PI control compared to the rotor resistance variation, but
the oscillations remain apparent.
Hamane et al. [6] have investigated the decoupling control of active and reactive DFIG powers by
the Sliding Mode (SM) regulator which shows superiority over PI during robustness testing, but this method
had a slow response time. Viswanadha et al. [7] have commanded the DFIG by two methods; the DPC
command and the power transfer matrix model, they have concluded that the power transfer matrix method
has a better result than the DPC command. Giannakis et al. [8] have presented a study of the decoupled

Journal homepage: http://iaescore.com/journals/index.php/IJPEDS

572  ISSN: 2088-8694

control of P-Q without mechanical sensor for DFIG-based WECS with intelligent controllers. They are
focused on the implementation of fuzzy logic controllers combined with conventional PI for the adjustment
of active and reactive power.
In these last years, a big interest is given to the use of neural network in identification and control of
the nonlinear systems [9] [10]; this is mainly due to their capacities of training and generalization. Mishra et
al [9] have developed an indirect vector control by implementing intelligent controllers (Fuzzy and Neural
netwerk) to control the speed of an induction motor (IM), they have concluded that the controller based on
the neural network has a more robust than the PI and fuzzy logic controllers.
This paper presents a comparison between the WECS performance using PI and NN controllers. The
PI regulator is simple and easy in implementation and gives acceptable performances, but it hasn't robustness
in case of parameter variations.The neural network controller is proved to be an interesting method for the
design of controllers and applied in many fields because of its excellent properties, such as insensitivity to
external disturbances and parameter variations, It can present also fast dynamic responses if the switching
devices support a high frequency. The studied system is presented in (Figure 1).

Figure 1. Diagram of the studied system

2. MODELING AND CONTROL OF DFIG

2.1. DFIG modeling
The Park transformation of the DFIG electrical equations gives the following equations [11]-[14]:

V  R i  d sd  
 sd s sd ddt s sq
Vsq  Rsisq  dtsq s  sd
 d rd (1)
Vrd  Rr ird  dt r  rq
V  R i  d rq  
 rq r rq dt r rd
The fields are given by:

 sd  Lsisd  M srird
  sq  Lsisq  M srirq
 rd  Lr ird  M srisd
(2)

  rq  Lr irq  M srisq
The electromagnetic torque is given by:

C em  P
Ms
Ls
r

 sqird  sd irq  (3)

d mec (4)
J  C em  C r  f . mec
dt

Int J Pow Elec & Dri Syst, Vol. 9, No. 2, June 2018 : 571 – 578
Int J Pow Elec & Dri Syst ISSN: 2088-8694  573

2.2. Power control

To easily control the electrical power produced by the DFIG, an independent control of the active
and reactive powers is applied by FOC of stator. The principle of this method consists in aligning stator field
along the d axis of Park reference frame in order to eliminate the coupling between the powers. (Figure 2)
[3] [15]. We have a quadratic stator flux null ( sq  0 ), then the direct stator flux is given by sd  s . The
equation systems (1) and (2) can be simplified as the following form:

 Vsd  Rs isd

 Vsq  Rs isq  s  s

Vrd  Rr ird  ddtrd r  rq (5)

 d
Vrq  Rr irq  dtrq r  rd
For the high power machines, the resistance of the stator windings are neglected. Equation 6.

Figure 2. Position of the stator flux [16] [17].

 V 0
sd
 V V  
 sq s
d
s s

V  R i  rd   (6)
 rd r rd ddt r rq
V  R i  rq   
 rq r rq dt r rd

  s  Ls isd  M srird
 0 Ls isq  M srirq
  L i  M i (7)
 rd  Lr ird  M sri sd
 rq r rq sr sq
Msr (8)
Cem   P s irq
Ls
The active and reactive stator power in the Park reference, are written as:
 P vsd isd  vsqisq
Q v i v i (9)
 sq sd sd sq
According to FOC, the equation systems (9) can be simplified as:
 P vsisq
Q v i (10)
 s sd
 V
s Msr
isd   L  L .ird
 s s
Msr
s (11)
 isq   .irq
 L
s

Artificial Intelligence Control Applied in Wind Energy Conversion System (Arama Fatima Zohra)
574  ISSN: 2088-8694

 V M
s sr
 P   L irq
 s
 VM V2
Q   s sr i  s (12)
 L
s
rd L 
s s
  M2  M V
 rd  Lr  sr ird  sr s
  L   L
  s  s s

  M 2  (13)
   L  sr i
 rq  r L  rq
  s 

  M 2  di  M2 
 V  R i   L  sr  rd  g  L  sr i
r rd  r L  dt s  r L  rq
 rd
 s   s 
 (14)
V  R i   L  M sr  dirq  g  L  M sr i  g M srVs
2 2

 rq r rq  r L  dt s  r L  rd L
  s   s  s

2.3. Control of active and reactive powers of DFIG

The decoupling of the active and reactive powers is due to voltages and currents which are evaluated
using transient equations of the machine [18]. This method is favored with microprocessors, but it is very
sensitive to parameter variations of the machine. The voltages are calculated using the power equation.

 
L 
L 
  R  L  LL  
2
m
2
m

Vdr g 
 r
L 
* P   s
 p * Q   R V   L  L
r r
s 2  Vs 
p
  L  
r s m

 
s r
 s Lm 
 L   
VL s m V Lm s L s m s

 L s
  s
(15)

 R  L  L    L  LL 
 L   2 2

 m m

 Vqr  V Lm  p * Pg  V L * Qg  L  L

r r r
s  L s 2  Vs
  L
m
s s r

  L  s s m s  s m

  s
L s

In this used method, the power is controlled using two cascade controllers, the first is the power
controller and the second is the current controller. Figure 3.

Figure 3. Control diagram of DFIG.

Int J Pow Elec & Dri Syst, Vol. 9, No. 2, June 2018 : 571 – 578
Int J Pow Elec & Dri Syst ISSN: 2088-8694  575

3. NEURAL NETWORK CONTROL

The use of an analogical controller leads to the performance degradation in nonlinear and several
processes [19], [10]. Many intelligent control methods have been applied on DFIG as fuzzy control, neural
networks [20], [10].
The use of neural networks is a technique for controlling complex systems can be justified by its
simplicity of implementation (preliminary mathematical analysis) [21], [10]. In this work, we consider the
process as a black box that has the ability to control the minimum of process informations. The idea is to
replace the four PI regulators of FOC by neural regulators (NN) simple. For the training of the neural weights
from PI regulators, we use an algorithm of back-propagation called the algorithm of Levenberg-Marquardt
(LM) [22] [23].
Each neural network has a function that is well defined depending on selected architecture (hidden
layers number and neural number in each hidden layer). The problem is to find the NN regulator that gives
the better results. In our case, we take a structure of neural network with only one hidden layer that
containing three neurons using the sigmoid transfer function (Figure 4).

Figure 4. Configuration of the multilevel perceptrons: (1-3-1).

The global diagram of FOC control with neural network regulators is presented in (Figure 5).

Figure 5. Global diagram of FOC control with neural network regulators

4. SIMULATION RESULTS
To prove the robustness and effectiveness of proposed control system, we apply the previous
proposed controller to control of the deoubly-fed induction generator with wind energy conversion system.
The generator and conversion system parameters are given in appendix. The configuration of the overall
control system is shown in Fig.05. It mainly consists of a doubly-fed induction generator, a ramp comparison
current-controlled pulse width modulated (PWM) inverter, an inverse park, a field oriented control and an
outer power feedback control loops. Two different operating conditions are simulated, reference tracking and

Artificial Intelligence Control Applied in Wind Energy Conversion System (Arama Fatima Zohra)
576  ISSN: 2088-8694

robustness against DFIG parameters variation, to illustrate the operation of the PI controller, RST controller
and the proposed NN regulators.

4.1. The reference tracking

In this test, the simulation has been done with the rated parameters of the DFIM. This test consists
of step changes in active and reactive powers in order to observe the system behavior with the proposed
controller. The active and reactive power changes are made at 0.35sec and 0.45sec, respectively.
The performance of wind energy conversion system with DFIG under this operating points test is
shown in figure 6. This figure shows the active and reactive power responses for the PI, RST and NN
controllers. The obtained results that are shown in figure 6 prove that the proposed controller has good
performances in active and reactive power tracking. It is clearly observed that the oscillations decrease and
the proposed NN controller present minimal rising time, no overshoot and negligible steady-state error
compared to PI and RST regulators.

5
x 10
5
Psref
Active power (W)

0
Ps(PI)
-5
Ps(NN)
-10 Ps(RST)
-15

-20
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Time (s)
6
x 10
1
Reactive power (var)

0.5
Qs ref
0 Qs(PI)
Qs(NN)
-0.5
QS(RST)
-1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Time (s)

Figure 6. Simulation results for reference tracking using PI, RST and NN regulators.

4.2. Robustness test aginst parameter variation

The effectiveness of the proposed WECS with DFIG for the different controller has been tested
under DFIG parameters variation. This second test consists of change on rated values of rotor resistance,
stator inductance, rotor and mutual inductances. The applied variation rotor resistance is 50% of its nominal
value whereas the stator and rotor inductance has been increased by 10% to their rated values. Finally, a
decrease of 10% on the mutual inductance has been done.The performances of WECS with DFIG under this
operating points test are depicted in figure 7-10. Figure 7 shows the simulation results of the system using the
different controllers for rotor resistance variation. We notice here that the same reference of active and
reactive power has taken for this test. From the simulation results, it is clearly observed that the system is not
sensitive to this parameter variation and the NN controller shows good performance compared to another
controller (PI and RST regulator). The performance of the system in the presence of the rotor and stator
inductances are depicted in figure 8 and 9. Form the simulation results, it can be seen that the variation in
rotor or stator inductances has an important effect on the control performance of WECS for the PI and RST
regulators. However, the NN regulator presents the best performance in active and reactive responses
compared to classical PI and RST regulators.
Finally, the dynamic responses of the wind energy conversion system equipped whit DFIG system
under mutual inductance variation are illustrated in figure 10. The obtained results showed that mutual
inductance has an important effect on the system responses for the classical regulators. However, the NN
regulator can keep the same characteristics against this variation with a small effect on the rise time response.

Int J Pow Elec & Dri Syst, Vol. 9, No. 2, June 2018 : 571 – 578
Int J Pow Elec & Dri Syst ISSN: 2088-8694  577
5 5
x 10 x 10
5 5
Psref
Psref

Active power (W)

Active power (W)

0 0 Ps(PI)
Ps(PI)
-5 Ps(RST)
Ps(NN) -5
PS(NN)
-10 Ps(RST)
-10
-15
-15
-20 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Time (s)
Time (s) 5
x 10
6 2
x 10

Reactive power (var)

1
Reactive power (var)

0
Qsref
0.5
Qs ref -2 Qs(PI)
0 Qs(PI) Qs(RST)
-4
Qs(NN) Qs(NN)
-0.5
QS(RST) -6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
-1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Time (s)
Time (s)

Figure 7. Simulation results for reference tracking Figure 8. Simulation results for reference tracking
using PI, RST and NN regulators with rotor using PI, RST and NN regulators with stator
resistance variation (+50 %) inductance variation (+10 %).

6 6
x 10 x 10
1 1
Psref Psref
Active power (W)

Active power (W)

0.5 0.5
Ps(PI) Ps(PI)
0 Ps(RST) 0 Ps(RST)
-0.5 Ps(NN) -0.5 Ps(NN)
-1 -1

-1.5 -1.5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Time (s) Time (s)
6 6
x 10 x 10
1 1
Reactive power (var)

Reactive power (var)

0.5 0.5
Qs ref Qs ref
0 Qs(PI) 0 Qs(PI)
Qs(RST) Qs(RST)
-0.5 -0.5
Qs(NN) Qs(NN)
-1 -1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Time (s) Time (s)

Figure 9. Simulation results for reference tracking Figure 10. Simulation results for reference tracking
using PI, RST and NN regulators with rotor using PI, RST and NN regulators with mutual
inductance variation (+10 %) inductance variation (-10 %)

5. CONCLUSION
This paper presented a WECS based on DFIG. The studied device is composed of a DFIG with the
rotor connected to the grid via a converter. The FOC of DFIG has been presented to regulate the active and
reactive powers of the machine. The regulation is made with PI, RST regulators and with neural networks.
The architecture of the neural network corrector retained is 1-3-1. It enabled us to improve the dynamic and
static performances of the DFIG. Three different controllers are analyzed and compared. In the case of power
reference tracking, the performances of the three controllers are almost similar but with a faster response time
for the regulator neural network. A robustness test has also been investigated where the machine’s parameters
have been modified, the impact on the active and reactive powers values is important for PI controller
whereas it is almost negligible for NN regulators.

REFERENCES
[1] T. Ackermann and L.Söder, "An overview of wind energy-status 2002," Renewable and Sustainable Energy
Reviews, vol. 6, pp. 67-128, 2002.
[2] B. Belabbas et al., "Hybrid Fuzzy Sliding Mode Control of a DFIG Integrated into the Network," International
Journal of Power Electronics and Drive System (IJPEDS) , vol. 3, pp. 351-364 , December 2013.

Artificial Intelligence Control Applied in Wind Energy Conversion System (Arama Fatima Zohra)
578  ISSN: 2088-8694

[3] B. Hopfensperger et al., "Stator-flux-oriented control of a doubly-fed induction machine with and without position
encoder," Proceedings: Electric Power Applications, vol. 147, pp. 241-250, July 2000.
[4] N. Zerzouri and H. Labar, "Active and Reactive Power Control of a Doubly Fed Induction," International Journal of
Power Electronics and Drive System, vol. 5, pp. 244-251, October 2014.
[5] F. Poitiers et al., "Control of a doubly-fed induction generator for wind energy conversion systems," IEEE Trans
.Renewable Energy, vol. 3, pp. 373-378, December 2001.
[6] B. Hamane et al., "Direct active and reactive power control of DFIG based WECS using PI and sliding mode
controllers," IECON Proceedings (Industrial Electronics Conference), no. 7048784, pp. 2050-2055, 2014.
[7] K. Viswanadha S Murthy et al., "A Performance Comparison of DFIG using Power Transfer," International Journal
of Power Electronics and Drive System, vol. 5, pp. 176-184, October 2014.
[8] A. Giannakis et al., "A combined control strategy of a DFIG based on a sensorless power control through,"
Renewable Energy, pp. 30052-1, 2018.
[9] A. Mishra and P. Choudhary, "Artificial Neural Network Based Controller for Speed Control," International Journal
of Power Electronics and Drive System (IJPEDS), vol. 2, pp. 402-408, December 2012.
[10] S. Laribi et al., "Impedance model for diagnosis of water management in fuelcell susing artificial neural networks
methodology," international journal of hydrogen energ, vol. 41, pp. 17093-17101, 2016.
[11] F. Poitiers et al., "Advanced control of a doubly-fed induction generator for wind," Electric Power Systems
Research, vol. 79, p. 1085–1096, 2009.
[12] N. Bouchiba et al., "Implementation and comparative study of control strategies for an isolated DFIG based WECS,"
The european phsical journal plus , pp. 132-415, 2017.
[13] B. Elnaghi et al., "Adaptation of PI controller used with combination of perturbationand observation method and
feedback method for DFIG," Electrical Engineering, pp. 0565-8, 2017.
[14] B. HAMANE et al., "Comparative Study of PI, RST, Sliding Mode and Fuzzy Supervisory Controllers for DFIG
based Wind Energy Conversion System," International Journal of Renewable Energy Research, vol. 5, pp. 1174-
1184, 2015.
[15] T. Arantxa Tapia et al., "Modeling and control of a wind turbine driven doubly fed induction generator," IEEE
Trans. on Energy Conversion, vol. 18, pp. 194-204, June 2003.
[16] Z. Boudjema et al., "A novel direct torque control using second order continuous sliding mode of adoubly fed
induction generator for a wind energy conversion system," Turkish Journal of Electrical Engineering & Computer
Sciences, vol. 25, p. 965 – 975, 2017.
[17] M. Allam et al., "A comparative study between field oriented control and sliding mode control for dfig integrated in
wind energy system," Journal of Electrical Engineering, vol. 15, pp. 205-213, 2015.
[18] S. Drid et al., "The Doubly Fed Induction Machine Modeling In The Separate Reference Frames," Journal of
Electrical Engineering, vol. 4, pp. 11-16, 2004.
[19] J. J. François, "Les Réseaux de Neurones : Principes et Application," Hermès Sciences Publicat, sept 1994..
[20] Song, Yong-Hua and Johns Allan T, "Application of fuzzy logic in power systems. Part 2. Comparison and
integration with expert systems, neural networks and genetic algorithms," Power Engineering Journal, vol. 12, pp.
185-190, August 1998.
[21] W. Qiao et al., "Design of optimal PI controllers for doubly fed induction generators driven by wind turbines using
particle swarm optimization," IEEE International Conference on Neural Networks - Conference Proceedings, pp.
1982-1987, 2006.
[22] J. Moré, "The Levenberg-Marquardt Algorithm: Implementation and Theory," Numerical Analysis, ed.G. A. Watson,
Lecture Notes in Mathematics 630,Springer Verlag, pp. 105-116., 1977.
[23] M.T. Hagan and M. Menhaj, "Training feedforward networks with the Marquardt algorithm," IEEE Trans, Neural
Networks, vol. 5, pp. 989-993, 1994.

APPENDIX
Table 1. Parameters of DFIM Table 2. Parameters of Turbine Table 3. Parameters of Feed
Symbol Value Symbol Value Symbol Value
Rated Power PN 1.5 MW Radius of the wind 35.25 m Stator rated voltage 398/690V
Stator resistance Rs 0.012 turbine R Vs Rated
Rotor resistance Rr 0.021 Gear box G 90 Frequency stator f 50 Hz
Stator inductance Ls 0.0137 H inertia J 1000 kg.m ² Rotor rated voltage Vr 225/389 V
Rotor inductance Lr 0.0136 H Surface swept by π ∙ R² m ² Rated frequency 14 Hz
Mutual inductance Lm 0 0.0135 H rotor S stator f2
The friction coefficient fr 0.0024 N.m.s1 Air density 1.22 kg/ m 3
Slip g 0.03
Pole Pairs p 2

Int J Pow Elec & Dri Syst, Vol. 9, No. 2, June 2018 : 571 – 578