IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 27, NO.

3, JULY 2012

1485

Dynamic Model and Control of DFIG Wind Energy Systems Based on Power Transfer Matrix
Esmaeil Rezaei, Ahmadreza Tabesh, Member, IEEE, and Mohammad Ebrahimi
AbstractThis paper presents a power transfer matrix model and multivariable control method for a doubly-fed induction generator (DFIG) wind energy system. The power transfer matrix model uses instantaneous real/reactive power components as the system state variables. It is shown that using the power transfer matrix model improves the robustness of controllers as the power frame of reference. The waveforms are independent of a sequential loop closing technique is used to design the controllers based on the linearized model of the wind energy system. The designed controller includes six compensators for capturing the maximum wind power and supplying the required reactive power to the DFIG. A power/current limiting scheme is also presented to protect power converters during a fault. The validity and performance of the proposed modeling and control approaches are investigated using a study system consisting of a grid-connected DFIG wind energy conversion system. This investigation uses the time-domain simulation of the study system to: 1) validate the presented model and its assumptions, 2) show the tracking and disturbance rejection capabilities of the designed control system, and 3) test the robustness of the designed controller to the uncertainties of the model parameters. Index TermsDoubly fed induction generator (DFIG), dynamics modeling, instantaneous power, multivariable control, wind energy systems, wind power control, wind turbine generator.

I. INTRODUCTION

IND ENERGY conversion systems are currently among economically available and viable renewable energy systems which have experienced rapid growth in recent years. Increasing the penetration level of wind farms highlights the grid integration concerns including power systems stability, power quality (PQ), protection, and dynamic interactions of the wind power units in a wind farm [1][3]. Wind energy systems based on doubly fed induction generators (DFIGs) have been dominantly used in high-power applications since they use power-electronic converters with ratings less than the rating of the wind turbine generators [4][8]. The scope of this paper is dynamic modelling and control of DFIG wind turbine generators. Modeling and control of DFIGs have been widely investigated based on well-established vector control schemes in a

Manuscript received July 31, 2011; revised April 02, 2012; accepted April 13, 2012. Date of publication May 30, 2012; date of current version June 20, 2012. Paper no. TPWRD-00653-2011. The authors are with the Department of Electrical and Computer Engineering, Isfahan University of Technology, Isfahan 84156, Iran (e-mail: rezaei58@ec. iut.ac.ir; a.tabesh@cc.iut.ac.ir; mebrahim@cc.iut.ac.ir). Digital Object Identier 10.1109/TPWRD.2012.2195685

stator eld-oriented frame of reference [7][9]. The vector control is a fast method for independent control of the real/reactive power of a machine. The method is established based on conframe of reference using an trol of current components in a transformation. Since the components are not physically available, the calculation of these components requires a phase-locked loop (PLL) to determine synchronous angle [10], transformations are often ig[11]. The dynamics of nored in the procedure of control design. Thus, any control design approach must be adequately robust to overcome the uncertainties in estimation of machine parameters as well as unaccounted dynamics of the overall system. The proposed power transfer matrix model for DFIG in this paper presents an alternative modeling and control approach which is independent of transformations. Direct torque control (DTC) and direct power control schemes (DPC) have been presented as alternative methods which directly control machine ux and torque via the selection of suitable voltage vectors [12][14]. It has been shown that DPC is a more efcient approach compared to modied DTC [15][17]. However, the DPC method also depends on the estimation of machine parameters and it requires a protection mechanism to avoid overcurrent during a fault in the system. This paper presents a modelling and control approach which uses instantaneous real and reactive power instead of components of currents in a vector control scheme. The main features of the proposed model compared to conventional models in the frame of reference are as follows. 1) Robustness: The waveforms of power components are independent of a reference frame; therefore, this approach is inherently robust against unaccounted dynamics such as PLL. 2) Simplicity of realization: The power components (state variables of a feedback control loop) can be directly obphase voltage/current quantities, which tained from simplies the implementation of the control system. Using power components instead of current in the model of the system, the control system requires an additional protection algorithm to prevent overcurrent during a fault. Such an algorithm can be simply added to the control system via measuring the magnitude of current. The sequential loop closing technique is adopted to design a multivariable control system including six compensators for a DFIG wind energy system. The designed control system captures maximum wind power via adjusting the speed of the DFIG and injects the required reactive power to the system via a grid-side converter.

0885-8977/$31.00 2012 IEEE

1486

IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 27, NO. 3, JULY 2012

is approximately constant and derivatives of assumption, currents will be proportional to the derivatives of power based on (2) and (5). B. Model of DFIG Using Instantaneous Power Components The voltage and ux equations of a doubly fed induction machine in the stator voltage synchronous reference frame can be summarized as [18]
Fig. 1. Schematic diagram of the DFIG-based wind generation system.

(6) (7)

II. MODEL OF A DFIG WIND ENERGY SYSTEM USING INSTANTANEOUS POWER COMPONENTS A. Denitions and Assumptions The schematic diagram of a DFIG wind turbine generator is depicted in Fig. 1. The power converter includes a rotor-side converter (RSC) to control the speed of generator and a grid-side converter (GSC) to inject reactive power to the system. Using a passive sign convention, the instantaneous real and reactive and , power components of the grid-side converter, in the synchronous reference frame, are [18]

(8) and are the stator and rotor resistances, and is where the synchronous (stator) frequency. Subscripts and signify and are the stator, the stator and rotor variable, and rotor, and magnetization inductances, respectively. The comand represent the voltage, current, plex quantities and ux vectors, and is the slip frequency dened as

(1) and are components of the stator voltages where and GSC currents in the synchronous reference frame, respectively. Solving (1) for and , we obtain (9) where is the rotor speed of the induction machine. To obtain and , the rotor ux and a model of DFIG in terms of current are obtained from (8) as

(2) where (3) Similarly, the instantaneous real/reactive power components of DFIG can be obtained in terms of stator currents as (4) (11) and the stator current components are given by (5) The negative sign in (5) complies the direction of the stator power ow on Fig. 1. The exact dynamic model of an induction machine is conventionally expressed by voltage and torque equations [18]. Herein, we develop a simplied model for the DIFG-based wind turbine of Fig. 1 by substituting currents in the exact model in terms of instantaneous real and reactive power. The key assumption to simplify the model is assuming an approximately constant stator voltage for DFIG. This assumption can be only used under a steady-state condition where the grid voltage at the point of common coupling (PCC) varies in a narrow interval, typically less than 0.05 p.u. Using this Using (5) to replace components of rearranging the equation, we obtain in (11) and by (10) . Then, by substituting for where and from (10) in (7) and then by solving (6) and (7) , we obtain for

(12) (13) where

(14)

REZAEI et al.: DYNAMIC MODEL AND CONTROL OF DFIG WIND ENERGY SYSTEMS

1487

Fig. 2. Equivalent circuit of the grid-side lter.

The state equation of the stator ux can be obtained by substiand from (5) in (6). Solving the stator voltage tuting for yields equations for (15) (16)

C. Grid-Side Converter and Filter Model Fig. 2 shows the representation of the grid-side converter and model of its lter in the synchronous reference frame. The the grid-side converter and lter is (22)

The electromechanical dynamic model of the machine is [18] (17) where and are the number of pole pairs, inertia of the rotor, and mechanical torque of the machine, respectively. The electric torque is given by [18]

where and are the resistance and inductance of the lter, respectively, and subscript signies the variables at the gridside converter [19]. Substituting for from (2) in (22) yields (23) where

(18) (24) In (17), the mechanical torque is input to the model and , based on (18), can be expressed in terms of instantaneous real and from (5) in (18) and reactive power. Substituting for in (17), we deduce and then replacing (19) where (25)

The dc-link model can be deduced from the balance of real power at the converter dc-link node as given by (26) where is the real power that the converter delivers to the rotor and represents the total power loss, including converter switching losses and copper losses of the lter. The delivered real power to the rotor is [18] (27) Using (10) and (5), can be expressed as

(20) The simplied model of the induction machine is presented in (12)(16) and (19) which is summarized as

(28) In the high-power converter, the power loss is often less than 1% in (26) can of the total transferred power, and the impact of in (26), be neglected. Substituting the model of the dc link is deduced as follows: (29) The model of DFIG in (21) is a nonlinear dynamic model since the coefcients of the state variables are functions of the state variables. Using (28), the right-hand-side quantities in (29) can be expressed in terms of the state variables .

(21)

1488

IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 27, NO. 3, JULY 2012

D. Wind Turbine Model The captured mechanical power by a wind turbine can be expressed with the algebraic aerodynamic equation as [1] (30) are the wind turbine radius, air mass density, where is the wind turbine power coand wind speed, respectively. efcient which is a function of the tip speed ratio and the pitch angle of the turbine blades, . For a high-power wind turbine, the maximum mechanical power captured at ranges from 6 to 8. Theoretically, it can be shown that 0.6 is about 0.5 for high-power and practically at wind turbines [1]. III. LINEARIZED DYNAMIC MODEL OF A DFIG WIND TURBINE GENERATOR A. DFIG and Wind Turbine Model For a high-power machine, the stator resistant is small; therefore, based on (6), a constant stator voltage under normal operation yields slow-varying ux components. Thus, the components of the stator ux of a DFIG in a eld-oriented frame of 0 can be obtained from (15) and (16) as reference with (31) Substituting for from (31) in (12), (13), and (19), then by linearizing the equations about an operating point, the smallsignal model of DFIG can be expressed as (32) (33) (34) where denotes small-signal quantities, and (35) In the linearized model, superscript 0 denotes the quantities at , the power torque equation an operating point. To calculate is linearized by assuming a constant wind speed as

Transferring the linearized dynamic model of DFIG and wind turbine in the Laplace domain yields

(37) where (38) Using (37), the dynamic model of DFIG and the wind turbine in Laplace domain can be expressed based on a power transfer function as (39) where and can be readily obtained from the solution of (37) for .

B. Model of the Grid-Side Filter and DC Link The model of the grid-side lter in Laplace domain can be obtained by transferring (23) into the Laplace domain as (40) where (41) Solving (40) for Laplace domain is and , the grid-side lter model in the

(42) where (43)

(44)

Using (29), the linearized model of dc link can be obtained as

(36) where is obtained via linearizing in (30) as given by where

(45)

(46)

REZAEI et al.: DYNAMIC MODEL AND CONTROL OF DFIG WIND ENERGY SYSTEMS

1489

the SLC design method [20], the multivariable system is stable if all of the designed subsystems during the sequential controller design procedure are stable. B. Design of the Machine-Side Controllers 1) Stator Real and Reactive Power Controllers: Considering as the rst pair in (39) and, thus, imposing , we obtain the rst SISO subsystem for controller design as (48) The rst controller to be designed is
Fig. 3. Schematic diagram of the feedback control system for the machine-side and grid-side converters.

(49) Substituting from (49) in (48), the closed-loop model of the rst subsystem in Laplace domain is (50) must be designed so that all poles of (50) remain in Thus, can be simply perthe left-half plane (LHP). The design of formed via SISO system design methods, such as frequency refor reactive power control, sponse or root locus. To design is considered as a part of the system, then the rst controller and by substituting for in (39), the closed-loop model of the second subsystem is obtained (51) where

From (45), the dc bus model in the Laplace domain is (47) Equations (39), (42), and (47) represent the linearized multivariable model of a DFIG wind turbine generator. IV. MULTIVARIABLE CONTROLLER DESIGN FOR A DFIG WIND TURBINE GENERATOR A. Controller Design Scheme Fig. 3 depicts the suggested multivariable feedback control system for the machine- and grid-side control schemes. In this scheme, the control inputs of the linearized model of the system to control real/reactive power of the rotor; and are to adjust the dc-link voltage and injected reactive power to the system. The outputs (feedbacks) of the system are the rotor speed, dc-link voltage, and the instantaneous real/reactive power of the rotor- and grid-side converters. The feedback control system includes six compensators which are used in two , and nested loops. The inner loops consist of where the required reactive power of the machine and grid and control loops as shown are directly controlled via for regulating the Fig. 3. The outer control loops include rotor speed and for adjusting the dc-link voltage level. The sequential loop closing (SLC) method [20] is adopted to design six controllers based on the multivariable model of the system developed in Section III. In the SLC method, based on physical relevance of the inputs and outputs, the input-output pairs are determined. Then, a controller is designed for the rst pair of the input-output by treating the system as a single-input single-output (SISO) system. The second controller is designed for the next pair of input-output variables using the rst controller as an integral part of the system. Based on the theory of

Thus, must be designed so that the second subsystem in (51) remains stable. 2) Rotor Speed Controller: Speed control of the turbine-generator rotor is performed via control of the real power of the uses as the constator. Therefore, the speed controller trol input. Using the control scheme of Fig. 3, is (52) Embedding and controllers in the model of the system, the transfer function of rotor speed can be calculated as (53) where

Substituting for

from (52) in (53) yields (54)

1490

IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 27, NO. 3, JULY 2012

Thus, must be designed so that the subsystem in (54) remains stable. C. Grid-Side Controller 1) Grid-Side Real and Reactive Power Controllers: The controller design procedure for and is quite similar to that of the rotor-side converter since both controllers have the and can be simply obsame structure. Therefore, tained by repeating the design procedure as explained in (48) with (51). The only modication is replacing . Also, both subscripts and should be replaced with subscript . For brevity, the details of the design procedure have been omitted. , 2) DC-Link Voltage Controller: Substituting for into (46), we obtain and (55) where detailed expressions for and are given in the can regulate Appendix. Based on (47) and (55), at its reference value using the dc-link controller in . Therefore, the closed-loop system for is deduced as

Fig. 4. Schematic diagram of the study system.

TABLE I STUDY SYSTEMS WIND TURBINE GENERATOR DATA

(56) where detailed expressions for and are given in the must be designed to stabilize the Appendix. Finally, dc-link closed-loop system in (56). D. Current Limiting During a Fault The target of the controller design procedure is to improve performance of wind energy conversion while maintaining the stability of the system under normal operating conditions. Therefore, the design procedure mainly deals with stability, tracking performance for capturing maximum wind power, disturbance rejection, and robustness against uncertainties and unaccounted dynamics. During a fault and/or sever transients, additional protection algorithms, such as fault ride through (FRT) and startup algorithms, must be added to the control system. Various algorithms, including active crowbar [21], series dynamic restorer [22], and dynamic voltage restorer [23] have been suggested for FRT. These algorithms are independent of the control approach during the normal operation; therefore, they can be used with the proposed transfer power matrix method herein as well. In addition to FRT algorithms and to mitigate overcurrent during a transient, an extra feedback loop can be used to sense the converter currents and reduce the power reference commands during transients. This extra loop only requires the magnitude of the current and it merely becomes operational during a fault condition. An example of such a current loop for the protection of the converter is elaborated in [19] and [23]. This loop does not impact the performance of controllers

during the normal operation of the system and, therefore, it will not be included in the design procedure of the controllers. V. MODEL VALIDATION AND PERFORMANCE EVALUATION OF THE MULTIVARIABLE CONTROL SYSTEM Fig. 4 shows the schematic of a study system for validation of the proposed modelling and control approaches. The study system includes a 1.5-MW DFIG wind turbine-generator connected to a grid. The electrical and mechanical parameters of the turbine generator are adopted from [24] and summarized in Table I. Using the proposed designed method, the following per-unitized controllers were designed for the study system (57)

(58)

(59) (60) The performance of these controllers was investigated based on time-domain simulations of the study system using the Matlab/ Simulink software tool. A. Tracking and Disturbance Rejection Capabilities Fig. 5(a) and (b) shows a trapezoidal pattern for wind speed and a step change in the reactive reference which are applied to

REZAEI et al.: DYNAMIC MODEL AND CONTROL OF DFIG WIND ENERGY SYSTEMS

1491

Fig. 5. Reference commands for wind and the stator reactive power.

Fig. 7. RMS values of the stator voltage and currents.

Fig. 6. Tracking performance of real and reactive stator powers.

Fig. 8. Robustness of the controllers to variations in L .

the controllers of the study system. The trapezoidal pattern was selected to examine the system behavior following variation in the wind speed with both negative and positive slopes. The selected wind speed pattern spans an input mechanical wind power from 0.7 to 1 p.u. (70 to 100% of the turbine-generator rated power). The reactive power command is a step change of 0.25 3 s when the real power is about 0.6 p.u. p.u. and occurs at Fig. 6 compares real/reactive power quantities of the DFIG against their command signals. Due to the coupling phenomenon, the variation of each power quantity can be considered as a disturbance to the other one. For instance, the effect of cou3 s, where the step compling can be seen in Fig. 6(a) at mand in reactive power causes a small deviation in real power. However, as Fig. 6 shows, both real and reactive power quantities accurately track their command signals which means the controllers successfully mitigate the impact of coupling effect in the tracking of commands signals. Fig. 7(a) and (b) depicts the dc-link voltage and the rms values of the machine voltage/current quantities. These gures show that the stator and rotor currents are changing as the real/reactive power changes whereas the dc link and stator voltages remained xed as expected from and current curves the control strategy. Specically, the 3 s, corresponding to the 0.25-p.u. show a step change at step command in the reactive power. Fig. 7 shows that as the

Fig. 9. Robustness of the controllers to a 40 error in the PLL angle.

power reference commands are within the rated power of the turbine generator, the voltage/current of the machine and converter will remain within their limits. B. Control System Robustness Fig. 8 shows the tracking and disturbance rejection performances of real/reactive power when the leakage inductance of

1492

IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 27, NO. 3, JULY 2012

the machine is changed using the same reference commands as shown in Fig. 5. Since Fig. 8 shows the responses accurately and , therefore, the designed track the commands for controller is robust to a variation of this parameter. Fig. 9 compares tracking performance of the proposed control system with the conventional vector control method as described in [25]. The PI controllers of the vector control method 0.1 and 2. were rst tuned for best performance at Then, the synchronous signal of the phase-locked loop (PLL) was deviated via biasing the PLL angle with 40 . As Fig. 9 shows, the proposed method accurately follows the reference commands for real and reactive power whereas the vector control method fails to track the commands. The reason is that the frame vector control method is signicantly sensitive to the of reference whereas the proposed control system is less independent to the reference frame.

APPENDIX Details of , and in (55) and (56) are shown in the equations at the top of the page. REFERENCES
[1] M. Patel, Wind and Solar Power Systems: Design, Analysis, and Operation. Boca Raton, FL: CRC, 2006. [2] K. Xie and R. Billinton, Determination of the optimum capacity and type of wind turbine generators in a power system considering reliability and cost, IEEE Trans. Energy Convers., vol. 26, no. 1, pp. 227234, Mar. 2011. [3] T. Zhou and B. Franois, Energy management and power control of a hybrid active wind generator for distributed power generation and grid integration, IEEE Trans. Ind. Electron., vol. 58, no. 1, pp. 95104, Jan. 2011. [4] J. Hu, H. Nian, H. Xu, and Y. He, Dynamic modeling and improved control of DFIG under distorted grid voltage conditions, IEEE Trans. Energy Convers., vol. 26, no. 1, pp. 163175, Mar. 2011. [5] L. Fan, H. Yin, and Z. Miao, On active/reactive power modulation of DFIG-based wind generation for interarea oscillation damping, IEEE Trans. Energy Convers., vol. 26, no. 2, pp. 513521, Jun. 2011. [6] S. Muller, M. Deicke, and R. De Doncker, Doubly fed induction generator systems for wind turbines, IEEE Ind. Appl. Mag., vol. 8, no. 3, pp. 2633, May/Jun. 2002. [7] E. Tremblay, S. Atayde, and A. Chandra, Comparative study of control strategies for the doubly fed induction generator in wind energy conversion systems: A DSP-based implementation approach, IEEE Trans. Sustain. Energy, vol. 2, no. 3, pp. 288299, Jul. 2011. [8] M. Mohseni, S. Islam, and M. Masoum, Enhanced hysteresis-based current regulators in vector control of DFIG wind turbines, IEEE Trans. Power Electron., vol. 26, no. 1, pp. 223234, Jan. 2011. [9] Z. Wang, G. Li, Y. Sun, and B. Ooi, Effect of erroneous position measurements in vector-controlled doubly fed induction generator, IEEE Trans. Energy Convers., vol. 25, no. 1, pp. 5969, Mar. 2010. [10] P. Rodrguez, A. Luna, I. Candela, R. Mujal, R. Teodorescu, and F. Blaabjerg, Multiresonant frequency-locked loop for grid synchronization of power converters under distorted grid conditions, IEEE Trans. Ind. Electron., vol. 58, no. 1, pp. 127138, Jan. 2011. [11] G. Escobar, M. Martinez-Montejano, A. Valdez, P. Martinez, and M. Hernandez-Gomez, Fixed reference frame phase-locked loop (FRF-PLL) for grid synchronization under unbalanced operation, IEEE Trans. Ind. Electron., vol. 58, no. 5, pp. 19431951, May 2011. [12] R. Datta and V. Ranganathan, Direct power control of grid-connected wound rotor induction machine without rotor position sensors, IEEE Trans. Power Electron., vol. 16, no. 3, pp. 390399, May 2001. [13] S. Chen, N. Cheung, K. Wong, and J. Wu, Integral sliding-mode direct torque control of doubly-fed induction generators under unbalanced grid voltage, IEEE Trans. Energy Convers., vol. 25, no. 2, pp. 356368, Jun. 2010.

VI. SUMMARY AND CONCLUSION An alternative modeling and controller design approach based on the notion of the instantaneous power transfer matrix is described for a DFIG wind energy system. The waveforms of the power components remain intact at different reference phase voltages frames and can be easily calculated using the and currents. Therefore, this approach facilitates the implementation of the controllers and improves the robustness of the control system. Furthermore, the proposed model can be potentially used to simplify the control issues of the wind energy system under an unbalanced condition since feedback variables are independent of -components in positive, negative, and zero sequences. The proposed approach is veried using the time-domain simulation of a study system for DFIG wind energy systems. The simulation results show that the suggested model and control scheme can successfully track the rotor speed reference for capturing the maximum power and maintain the dc-link voltage of the converter regardless of disturbances due to changes in real and reactive power references.

REZAEI et al.: DYNAMIC MODEL AND CONTROL OF DFIG WIND ENERGY SYSTEMS

1493

[14] S. Chen, N. Cheung, K. Wong, and J. Wu, Integral variable structure direct torque control of doubly fed induction generator, IET Renew. Power Gen., , vol. 5, no. 1, pp. 1825, 2011. [15] D. Zhi, L. Xu, and B. Williams, Model-based predictive direct power control of doubly fed induction generators, IEEE Trans. Power Electron., vol. 25, no. 2, pp. 341351, Feb. 2010. [16] P. Zhou, W. Zhang, Y. He, and R. Zeng, Improved direct power control of a grid-connected voltage source converter during network unbalance, J. Zhejiang Univ.-Sci. C, vol. 11, no. 10, pp. 817823, 2010. [17] J. Hu, H. Nian, B. Hu, Y. He, and Z. Zhu, Direct active and reactive power regulation of DFIG using sliding-mode control approach, IEEE Trans. Energy Convers., vol. 25, no. 4, pp. 10281039, Dec. 2010. [18] P. Krause, O. Wasynczuk, S. Sudhoff, and I. P. E. Society, Analysis of Electric Machinery and Drive Systems. Piscataway, NJ: IEEE, 2002. [19] A. Tabesh and R. Iravani, Multivariable dynamic model and robust control of a voltage-source converter for power system applications, IEEE Trans. Power Del., vol. 24, no. 1, pp. 462471, Jan. 2009. [20] J. Maciejowski, Multivariable Feedback Design, ser. Electron. Syst. Eng. Ser. Reading , MA: Addison-Wesley, 1989, vol. 1. [21] G. Pannell, D. Atkinson, and B. Zahawi, Minimum-threshold crowbar for a fault-ride-through grid-code-compliant DFIG wind turbine, IEEE Trans. Energy Convers., vol. 25, no. 3, pp. 750759, Sep. 2010. [22] J. Yang, J. Fletcher, and J. OReilly, A series-dynamic-resistor-based converter protection scheme for doubly-fed induction generator during various fault conditions, IEEE Trans. Energy Convers., vol. 25, no. 2, pp. 422432, Jun. 2010. [23] C. Wessels, F. Gebhardt, and F. Fuchs, Fault ride-through of a DFIG wind turbine using a dynamic voltage restorer during symmetrical and asymmetrical grid faults, IEEE Trans. Power Electron., vol. 26, no. 3, pp. 807815, Mar. 2011. [24] R. Fadaeinedjad, M. Moallem, and G. Moschopoulos, Simulation of a wind turbine with doubly fed induction generator by fast and simulink, IEEE Trans. Energy Convers., vol. 23, no. 2, pp. 690700, Jun. 2008. [25] R. Pena, J. Clare, and G. Asher, Doubly fed induction generator using back-to-back PWM converters and its application to variable-speed wind-energy generation, in Proc. Inst. Elect. Eng., Elect. Power Appl., 1996, vol. 143, no. 3, pp. 231241, IET.

Esmaeil Rezaei was born in Isfahan, Iran in 1979. He received the B.Sc. degree in electronics and the M.Sc. degree in electrical engineering from Isfahan University of Technology (IUT), Isfahan, Iran, in 2001 and 2004, respectively, where he is currently pursuing the Ph.D. degree in electrical engineering. He was a Technical Designer with the Information and Communication Technology Institute (ICTI), Isfahan University of Technology, from 2004 to 2007. His current research interests include electrical drives and energy conversion systems for renewable energy resources.

Ahmadreza Tabesh (M12) received the B.Sc. degree in electronics and the M.Sc. degree in systems control from Isfahan University of Technology, Isfahan, Iran, in 1995 and 1998, respectively, and the Ph.D. degree in energy systems from the University of Toronto, Toronto, ON, Canada, in 2005. From 2006 to 2009, he was Postdoctorate at the Microengineering Laboratory for MEMS, Department of Mechanical Engineering, Universit de Sherbrooke, Sherbrooke, QC, Canada. Currently, he is an Assistant Professor with the Department of Electrical and Computer Engineering, Isfahan University of Technology. His areas of research include renewable energy systems and micropower energy harvesters (power MEMS).

Mohammad Ebrahimi received the B.Sc. and M.Sc. degrees in electrical engineering from Tehran University, Tehran, Iran, in 1984 and 1986, respectively, and the Ph.D. degree in power systems from the Tarbiyat Modarres University, Tehran, Iran, in 1996. Currently, he is an Associate Professor at the Isfahan University of Technology (IUT), Isfahan, Iran. His research interests include electrical drives, renewable energy, and energy savings.