06657747

1082 IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 28, NO. 4, DECEMBER 2013

New Overall Control Strategy for Small-ScaleWECS in MPPT and Stall Regions With Mode

Transfer ControlZakariya M. Dalala, Student Member, IEEE, Zaka Ullah Zahid, Student Member, IEEE,

and Jih-Sheng (Jason) Lai, Fellow, IEEE

AbstractThis paper presents a new overall control strategyfor small-scale wind energy conversion systems (WECS). The pur-pose of the proposed strategy is to utilize a maximum power pointtracking control (MPPT) to maximize the captured energy whenthe wind speeds are below the rated speed. For high wind speedregion; two stall controllers are developed: The constant speed stallcontroller which will limit the rotational speed of the generator toits rated value, and the constant power stall controller, which willregulate the captured power to be within the system rating. Forthe MPPT control, a modified perturb and observe (P&O) algo-rithm is utilized, where the dc side current is used as a perturbationvariable and the dc-link voltage slope information is used to en-hance the tracking speed and stability of the algorithm. For thespeed and power regulation operation during high wind speeds,the system is controlled in the stall region to limit the rotationalspeed and the power of the generator. A stabilizing control loopis proposed to compensate for the stall region instability. A newmode transfer control strategy is developed to effectively controlthe transition between different modes of operation while ensuringthe system stability without using any preknowledge of the systemparameters. A 1 kW hardware prototype is developed and testedto validate the proposed new control strategy.

Index TermsConstant power stalling, maximum power pointtracking (MPPT), perturb and observe (P&O) algorithm, windenergy conversion system (WECS).

I. INTRODUCTION

AMONG renewable energy sources, wind energy is thefastest growing source so far; because of the wide avail-ability of the wind and the technical progress associated withthe advanced and low cost manufacturing of the wind tur-bines [1][3]. Large-scale wind energy farms have attainedmost of the attention in the last decades resulting in their matu-rity, while small-scale wind energy conversion system (WECS)need further investigation and performance optimization [4].Permanent-magnet synchronous generators (PMSGs) are pre-ferred in small-scale WECS because of their simple structure,higher efficiency and energy density, and ease of control [5], [6].

Manuscript received July 3, 2013; revised October 1, 2013; accepted October16, 2013. Date of current version November 20, 2013. Paper no. TEC-00376-2013.

The authors are with the Future Energy Electronics Center, Virginia Polytech-nic Institute and State University, Blacksburg, VA 24061-0111 USA (e-mail:[email protected]; [email protected]; [email protected]).

Color versions of one or more of the figures in this paper are available onlineat http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/TEC.2013.2287212

Fig. 1. Ideal power versus wind speed trajectory.

Full power conversion schemes have been adopted for windpower extraction and maximum power point tracking (MPPT)by utilizing different acdc converter topologies [6][8]. A sim-ple and low-cost alternative is to use a uncontrolled three-phaserectifier and a chopper circuit, where the chopper circuit is con-trolled such that the MPP is achieved and the generator speed isindirectly controlled [9][12].

Different control theories and algorithms have been appliedto WECS. The control objective depends on the application andoperating conditions. Fig. 1 shows the ideal aerodynamic windpower as function of the wind speed [13]. Below the cut-in speedVmin , the generation is halted because there is not enough powerto drive the generation system. In region I, between Vmin andVp , the MPPT operation should be realized where the maximumpower captured in that region is less than the rated system power.Between Vp and Vmax (region II), the maximum aerodynamicpower exceeds the system rating, and thus the controller shouldlimit the power captured by either using aerodynamic controlleror soft control technique. Above Vmax , which is the cut-offspeed, the wind power is very high, demanding to shut off thewind turbine by mechanical means to protect the mechanicalparts.

According to the published literature [5], [8], [9], [14][25],most control strategies have been developed to realize MPPTcontrol in region I, while few research has been carried out toimplement the control in region II [4], [13], [26][32] and evenfewer effort has been carried out to control the transfer betweenthese two regions during changing wind speed conditions.

Among those reported MPPT techniques in literature, theperturb and observe (P&O) is found to be the simplest andthe most used one due to its reliability. In P&O techniques, anoptimum power relation with other system variable is used totrack the MPP. The maximization variable is perturbed and thecaptured power is observed. Based on the power variation with

0885-8969 2013 IEEE

DALALA et al.: NEW OVERALL CONTROL STRATEGY FOR SMALL-SCALE WECS IN MPPT AND STALL REGIONS 1083

the perturbation introduced, the next step size and direction isdetermined [18], [23], [33]. P&O MPPT algorithms have theadvantages of being simple implementation algorithms, do notneed wind speed measurements, and there is no need to a priorknowledge of the system parameters. The main disadvantagesof the P&O algorithms include their slow response to rapid windspeed fluctuations and the tracking performance depends on thestep size of the algorithm [33].

As the MPPT is important to increase the system throughputwhile the wind speed is in normal conditions and below the rated,it is vital as well to realize the control objectives in the aboverated wind speed operating conditions. In region II, the availableaerodynamic power is excessive if the MPPT algorithm is set toaction. So, to protect the system hardware, the power should belimited below rated. For large turbines, aerodynamic control isused to limit the turbine power. Blades pitch control is usuallyimplemented to reduce the turbine power coefficient Cp , andhence, limit the power and speed of the turbine [1], [16], [26],[34], [35]. Pitch control increases the system complexity andcost and is only justified for large wind turbines applications,while small-scale wind turbines systems are in favor of usingsimpler and lower cost solutions. Soft stall control has beenproposed for small-scale WECS to regulate the shaft speed andpower in the above rated wind speed conditions [13], [20], [27],[31], [36]. The objective of the controller is to reduce the rotorspeed in the high wind speed conditions and stalling the turbine,where the power and speed will be limited. Constant speed softstalling is introduced as a control solution [20], [36], but theproblem with this method is that even the power is limited, itis still increasing with increasing wind speeds and the systemshould be rated accordingly. On the other hand, soft stallingwith power regulation can limit the speed and the power at thesame time by driving the generator into the deep stall region[13], [26], [27].

The intended purpose of this paper is to propose a new over-all control strategy for small-scale WECS in a wide wind speedrange, and to emphasize the difficulty in optimizing the controltransfer between different operating regions without preknowl-edge of the system parameters. This paper extends the authorswork in [12] and [37]. In region (I), the MPPT operation isrealized by a modified P&O control algorithm. The authors pro-posed a MPPT algorithm in [12] and it will be adopted in thispaper. In region (II), the power is regulated using cascaded loopdesign concept while ensuring that the WECS is driven to workin the stall region. Two stall controllers will be considered inthis paper: the first one is the constant power stall in region (II),where the system is controlled to capture the MPP up to the max-imum wind speed VP and then turns to the constant power stallmode after that. The second controller will consider a constantspeed region before the wind speed reaches its maximum. Thiscontrol mode is used by certain loads to regulate the voltage toa constant value and to relief the controller design [31]. Due tothe nonlinear speedpower characteristics, the system dynamicsare unstable in the stall region. In this paper, the stall region isinvestigated and a modeling approach is presented, then a stablecontroller design is accordingly derived. The proposed controlstrategy also deals with the control mode transfer between the

Fig. 2. WECS configuration.

MPPT and the stall region control. A mode transfer structure isproposed to ensure continuous generation and effective dynamicresponse against fast wind speed changing conditions withoutthe need to a previous knowledge of the system parameters.Finally, the proposed control strategy is verified experimentallyusing a 1 kW hardware prototype.

II. WECS CONFIGURATIONThe WECS schematic is shown in Fig. 2, where it consists of a

wind turbine coupled to a PMSG. The three-phase diode bridgeis used to rectify the generated ac voltage. The boost converteris used to boost the voltage across the load Ro . The load Rocan be replaced by a unity power factor inverter that supplies astandalone ac load or connected to the utility grid [10], [24], [38].The PMSG model in dq domain can be found in [5], [39].

The mechanical power of the wind Pwind can be expressedas [16]

Pwind =12R2V 3w (1)

where is the air density, R is the turbine radius, and Vw isthe wind speed at the turbine blades. The power captured by theblades of the turbine Pblade is

Pblade =12R2V 3w CP () (2)

where CP () is the turbine power coefficient and is a nonlinearfunction of the tip speed ratio and the pitch angle and canbe expressed as in (3) [40]

CP (, ) = 0.5176(

1161i 0.4 5

)21 1i

+ 0.0068

(3)1i

=1

+ 0.08 0.035

3 + 1(4)

= Rr/Vw (5)where r is the rotational speed of the wind turbine. TypicalCP () curve is shown in Fig. 3. The available turbine mechan-ical torque (Tm ) can be expressed as

Tm =12R2V 3w CP () /r . (6)

Fig. 4 shows the turbine rotor speed versus power for differentwind speeds, and Fig. 5 shows the speed versus turbine torquecharacteristics. The simulation parameters are listed in Table I.

1084 IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 28, NO. 4, DECEMBER 2013

Fig. 3. Typical power coefficient as function of the tip speed ratio curve.

Fig. 4. Turbine power as function of the shaft rotating speed for different windvelocities.

Fig. 5. Turbine torque as function of the shaft rotating speed for different windvelocities.

TABLE IPARAMETERS OF THE MACHINES, WIND TURBINE, AND POWER ELECTRONICS

INTERFACE

Fig. 6. Experimental power as function of the shaft rotating speed for differentwind velocities.

The wind turbine characteristics are emulated experimentallyusing a PM motor drive that is controlled to generate the windprofile and the experimental powerspeed characteristics areshown in Fig. 6. The emulated turbine characteristics will beused to drive the generator unit and to test the proposed controlstrategy in this paper.

III. MPPT CONTROLWhile the wind speed is below the rated value, the controller

objective is to follow the MPP trajectory which is representedby the path ABCD in Fig. 6. In the figure, it is assumedthat the maximum rated wind speed is 11 m/s. To implementthe MPPT control in this paper, the authors proposed a modifiedP&O method in [12] and it will be presented here briefly. Inthis reported method, the dc side current (inductor current) isused as a perturbation variable and the dc-link voltage slopeinformation is used to determine the step size and directionof the next iteration. The generator electromechanical torqueequation is shown as follows [39], [41]:

Te =3

3

P

2f iL = KT iL . (7)

From Fig. 4, Fig 5 and eq. (7), it can be concluded that,changing the electromechanical torque Te through the inductorcurrent control, will change the rotating speed, which, in turn,will change the absorbed power from the wind turbine accord-ing to Fig. 4; thus, the MPP can be achieved if the optimumvalue of the inductor current is set. Moreover, the commandedtorque through the inductor current (7) should not exceed themaximum available turbine torque (6) for the system to continuegeneration. Otherwise, the generator will decelerate under thetorque difference and stop at the end. The mechanical systemcan be described by (8) with neglecting the friction

Tm Te = J ddt

(8)

where J is the system inertia. According to (8), the generatorwill accelerate or decelerate depending on the torque differenceapplied to its shaft. The machine acceleration (d/dt) can bechanged by either changing Te or Tm as can be seen from(8). Te can be varied by controlling the inductor current, while

DALALA et al.: NEW OVERALL CONTROL STRATEGY FOR SMALL-SCALE WECS IN MPPT AND STALL REGIONS 1085

Tm depends on the wind speed and generator rotating speed,so it is uncontrollable. The step size in the inductor currentperturbation is small for fine tracking of the MPP, thus (d/dt)does not change much by the current step change, while in thecase of wind speed change, Tm changes considerably leading toa larger change in (d/dt).

Referring to Fig. 2, the rectified dc-link voltage at the outputof the rectifier can be expressed as in (9) [42]

Vdc =3

3

Vac-amp = KvVac-amp (9)

where Vac-amp is the ac side voltage peak which is for PMmachine can be written as (10) [39]

Vac = E Iac (Rs + jLs) (10)where RS and LS are the stator resistance and inductance, re-spectively. Iac is the ac side phase current, E is the back elec-tromotive force of the machine and is equal to f , and f isthe magnet flux linkage. From (9), (10)

Vdc . (11)The proportionality in (11) is assumed to be linear, while the

true relation does not form a straight line, and it depends on theoperating point. More accurate approximation can be derived asin [43]. A discussion on the validity of (11) can be found in [27]and [43]. From (11)

dVdcdt

ddt

. (12)As (12) suggests, the machine acceleration information is

projected into the dc-link voltage slope change. The voltageslope is used then to detect wind speed change. From (8) and(12)

dVdcdt

Tm TeJ

. (13)Incorporating (6) and (7) for the mechanical and electrome-

chanical torque equations and using (13)dVdcdt

((1/2)R2V 3w CP () /r (3

3/)(P/2)f iL

)J

.

(14)From (14), we can conclude (15) and (16)

dVdcdt

V 3w (15)dVdcdt

iL . (16)Clearly, the dc-link voltage slope shows much higher sen-

sitivity against the wind speed change rather than against theinductor current change. Through the dc-link voltage slope in-formation, we can detect any possible wind speed change duringthe operation of the conversion system. If the wind speed fluc-tuation is small in magnitude and slow, the slope will be smalland less than certain threshold, in that case; normal P&O algo-rithm is applied and the step size is determined based on themeasured power increment. In the case of large magnitude andfast wind speed fluctuations, the dc-link voltage slope will be

Fig. 7. Flow chart of the proposed MPPT algorithm.

steep according to (15), and a large step in the inductor current iscommanded to compensate for the changing wind speed. Duringwind speed change condition, the step size will be a measureof the voltage slope and the direction follows the same slopesign, as the slope will be positive for increasing wind speed andnegative for the decreasing one.

The flow chart for the proposed algorithm is shown in Fig. 7.The algorithm works in two distinct modes: The normal P&Omode under slow wind fluctuation conditions where an adaptivestep size P&O is employed with the power increment used as ascaling variable to determine the next step size. The other modeis the prediction mode under fast wind speed change conditions,and this mode is responsible of bringing the operating point tothe vicinity of the new MPP during fast wind speed change andit will help preventing the generator from stalling by rapidlyadjusting the generator torque in response to sudden wind speedslow down condition. In this mode, the dc-link voltage slopeis used as a scaling variable and is used also to determine thenext step direction. The algorithm will work under this mode ifthe dc-link voltage is noticed to be higher than certain thresholdKo , which is tuned experimentally [12].

1086 IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 28, NO. 4, DECEMBER 2013

IV. STALL REGION CONTROLDue to the size and power limitations of the WECSs, it is im-

portant for their control to protect them against high wind speedconditions. Following a typical powerspeed characteristics ofthe wind turbine as shown in Fig. 6, the MPPT is ensured bya variable speed operation. The MPPT control would increasethe rotational speed for increasing wind speeds to keep trackingof the MPP following the path AE. However, after the systemreaches its power rating limit at point D (Assuming the ratedpower is 1 kW), the captured power should be limited and thiscan be done by two ways; first: increasing the rotational speedwell above the MPP speed following the path DH, thus con-verting the wind energy into kinetic energy and reducing thepumped power into the converter circuits. This would cause theproblem of over speeding the generator and increasing the me-chanical stresses. So, it is not a favorable solution. The secondoption to limit the power is to decrease the speed to the lefthalf side of the curve to ensure decreasing rotational speed forincreasing wind speeds and maintaining constant level of powerfollowing the path DF, this operation is called soft stalling. Bydoing so, the power and speed limits of the generator unit willbe met.

A. Previous Soft Stall Control AlgorithmsSoft stall control has been proposed in literature to limit the

power in the above rated wind speed region [25], [26], and iscalled constant power stall. In this control, the MPP is trackedtill the power exceeds the system ratings. After that, the stallcontrol is activated to limit the power. A little over shoot in thesystem response is expected. Other soft stall control strategiesconsidered a constant speed region between the MPP and theconstant power regions [12], [27]. In one hand, it is required forsome loads and on the other hand it reliefs the controller imple-mentation [31]. The transition between the MPP control and thestall control in these reported methods is easy to implement andis theoretically seamless, as the control trajectory is predefined.For example, in [26], the optimum current command as functionof the dc-link voltage is known for the controller, and in [42],the optimum voltage command in the MPP and stall regions,including the constant speed region is obtained first and thenused by the controller. All what the controller needs to do is totrack the optimum relation stored in the form of lockup tablesor curves. Accurate system parameters knowledge is requiredto implement these methods which make them sensitive to pa-rameters variation. The MPPT technique used in these methodsalso uses the optimum relation defined for the controller. TheP&O MPPT algorithms were not reported to be used with thesemethods because it is not easy to secure a safe and fast transitionbetween the MPP and the stall regions without prior knowledgeof the system parameters.

As far as the P&O MPPT algorithm is desired for its simplic-ity and its independence of the system parameters, this paperfocuses on implementing a soft stall control strategy in conjunc-tion with the P&O algorithm in the MPP region. The problemof control mode transfer between MPPT and stall regions isaddressed. The system parameters knowledge is not required in

the proposed method to realize the control mode transfer whichis a key advantage over the reported methods in literature.

B. Modeling in the Stall RegionTo control the system in the stall region, it is important to

derive the dynamics of the system in that region to help indesigning a robust control law. Using Taylor series analysis,the linearized mechanical (6) and electromechanical (7) torqueequations can be described as

Te =TeiL

iL Te (s) = KT iL (s) (17)

Tm =TmVw

Vw +Tmr

(18)

Tm (s) =CoVwo

o[3Cp(o) oCp(o)]vw (s)

+CoVwoR

2o[o Cp(o) Cp(o)](s)

= vw (s) + r (s) (19)where Co = (1/2)R3 . From (7), (17), and (19)

Tm (s) Te (s) = Js (s) . (20)From (17)(20), the following expression is obtained.

(s)[sJ CoVwoR

2o

[oCp (o) Cp (o)

]]

= KT iL (s) + CoVwoo

[3Cp(o) o Cp(o)]vw (s). (21)

From (21), the current to rotor speed transfer function (22)and the wind speed to rotor speed transfer function (23) arederived(

(s)iL (s)

)vw (s)=0

=

KTsJ (CoVwoR/2o)

[o Cp(o) Cp(o)

] = KTsJ

(22)((s)vw (s)

)iL (s)=0

=

CoVwoo

[3Cp(o) o Cp(o)

]

sJ (CoVwoR/2o)[oCp (o) Cp (o)

]

=CoVwo

o

[3Cp(o) oCp(o)

]sJ . (23)

=CoVwoR

2o

[o Cp(o) Cp(o)

]. (24)

And it is assumed that vdc = Kv [27], so (25) is derivedvdc (s)iL (s)

= KT KvsJ (25)

DALALA et al.: NEW OVERALL CONTROL STRATEGY FOR SMALL-SCALE WECS IN MPPT AND STALL REGIONS 1087

Fig. 8. as function of . = C o Vw o R2o

[oCp (o ) Cp (o )] =(CoVw oR) ().

Fig. 9. Nyquist plot of the loop gain of the closed voltage loop. Operatingpoint at = 4.22, Vw = 15 m/s.

where o is the tip speed ratio at the operating point and Cp (o)is the slope of the power coefficient curve at the operating point.From (25), it can be seen that the system has a pole at s =/J , where is positive in the left hand side of the powercoefficient curve as can be seen in Fig. 8. The pole in the righthalf plane makes the system dynamics unstable under currentcontrol. So, the voltage loop should be closed to stabilize thesystem dynamics.

Fig. 9 shows the Nyquist plot for the designed loop gainof the closed voltage loop. The designed compensator is PIwith gains (Kp = 83.6, Ki = 880). As can be seen in thefigure, the number of encirclements of the (1 + j0) point isone in the counter clockwise direction; and the system has onepole in the RHP, meaning a stable system. The phase margin isdirectly measured at the plot and is more than 72. Moreover,with the increase of the system gain, the Nyquist plot will justexpand radially without changing the number of encirclementsmeaning an infinite gain margin system. With closing the voltage

Fig. 10. Frequency response of the closed power loop.

loop, the system is stabilizable and the RHP pole dynamics arecompensated.

To regulate the power generated, the dynamic relations as-sociated with power command are derived as well. The outputpower can be described as

Po = iL vdc = Te = KT iL. (26)Linearizing the power as function of both the current and the

rotational speed yields

Po = KT IL + KT o iL

Po(s) = KT IL(s) + KT oiL (s). (27)In case of regulating the output power through the dc-link

voltage, the plant is derived as follows:

Po(s) =KTKv

ILvdc(s) + KT oiL (s)

Po(s)vdc(s)

=KTKv

IL +KT oiL (s)

vdc(s)(28)

Po(s)vdc(s)

=KTKv

IL oKv

(sJ ). (29)

The power loop has a RHP zero leading to nonminimum phasesystem. The power loop can be stabilized with only integratorcontroller. Fig. 10 shows the bode plot of the power loop gainwith integrator compensator of gain Ki = 0.013.

The outer power loop has very slow dynamics compared tothe internal voltage loop, and that is acceptable for the WECS asthe power control is associated with mechanical state variables.The internal current loop is much faster than the power andvoltage loops and hence, its dynamics can be neglected whenconsidering the slow dynamics variables and the current can beassumed equal to its reference at all times.

V. PROPOSED CONTROL STRATEGYThe proposed overall control strategy block diagram is shown

in Fig. 11. The plant has two inputs, the wind speed and theelectromechanical torque represented by the dc side current.The internal current dynamics are much faster than the rest of

1088 IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 28, NO. 4, DECEMBER 2013

Fig. 11. Proposed overall control strategy schematic diagram.

the system dynamics, so they are neglected in the figure. Thevoltage loop is activated only in the stall region. Two soft stallcontrollers will be considered in this paper, the constant powerstall and the constant speed stall:

A. Constant Power Stall ControlIn this mode, the controller will try to follow the ideal wind

power curve shown in Fig. 1, where the constant speed regulationis not used.

Whenever the wind speed is below rated value, the MPPTcontrol is activated and the current reference is supplied by theMPPT block. During this operating condition, the power loopcompensator Gcp(s) is saturated to its maximum limit namedvmax . The limit vmax represents the maximum rated speed thatis allowed. When vrefdc is set to vmax , the voltage loop compen-sator is saturated at its maximum limit named imax . When thewind speed rises above rated value, the dc power is more thanPmax . The decision block decides to shift to stall control andpasses the current reference coming from the cascaded powerand voltage loops. At the instant of the transition, the immediatecurrent reference will be the saturated value imax (high torquevalue), which will force the system to go into the stall regionand force the speed to be reduced. During this operation, thecaptured power is reduced and the compensators start to desat-urate and power regulation takes place while in the stall regionto ensure reduced speed operation. When the wind speed comesback below rated maximum, the decision block detects the neg-ative difference between the commanded power Pmax and thecaptured power Pdc and decides to break the voltage loop andbypasses the output of the power loop compensator through thegain blocKf , where Kf is a large gain that helps to dischargethe compensator integrator very fast, resulting in a very smallcurrent reference. This small or nearly zero current referencewill release the torque from the output of the generator and willlet it to accelerate under the effect of the turbine torque only.While the generator speed increases, it will leave the stall regionand go back to the stable side (MPPT region) at minimal time.At the time the generator picks up speed, the dc-link voltage

rises and the decision block break the transition control loopand return to MPPT operation again.

In the proposed control strategy in this paper, the MPP istracked by utilizing the dc-link voltage slope information whilethe system is in the MPPT mode. However, the same conceptcan be used when the system is coming back from the stall re-gion. Take the case in Fig. 6 as an example. When the systemis working under stall control at point F , and a sudden drop ofthe wind speed to 9 m/s is assumed, the operating point tends tomove to point x while the desired one is point B. In this operat-ing condition and according to the described strategy above, theelectromechanical torque is released from the generator shaft al-lowing the generator to accelerate under the effect of the turbinetorque only. Thus, it can be assumed that

dVdcdt

(Tm )J

. (30)

From (30), the dc-link voltage slope will follow the turbinetorque characteristics in the stall region as can be seen in the lefthand side of the torque characteristics in Fig. 5. The slope willincrease with increasing turbine torque, and once the voltageslope starts to decrease, the system is judged to be out of the stallregion. At that moment, the MPPT control mode is activated.And to move the operating point rapidly to the new MPP (pointB), a current step is applied. The current step is proportional tothe measured dc-link voltage slope. Higher slope means higherturbine torque (30), requiring higher electromechanical torqueto match the turbine torque which can be achieved by applyinghigher current step.

As described in the above discussion, the proposed controlstrategy manages to operate the system with MPPT control,while the wind speed is below rated. Once it is above rated, thesystem will go into the stall region and the power is regulatedto the maximum rated value. In the case when the wind speedcomes back below rated; the MPPT control is activated with anadaptive current step to ensure the system operates near the newMPP.

DALALA et al.: NEW OVERALL CONTROL STRATEGY FOR SMALL-SCALE WECS IN MPPT AND STALL REGIONS 1089

Fig. 12. Ideal power versus wind speed trajectory with constant speed region.

B. Constant Speed Stall ControlThe other stall control strategy to be considered in this paper

is the constant speed stall control, which means constant voltageregulation at the rectifier output. Constant voltage regulation isneeded by some loads, for example, when the boost converteris used to charge a battery, the input voltage should not exceedthe battery voltage to ensure converter stability. And it is neededto protect the generator from over speeding. Constant voltageregulation is realized by inserting a constant speed region be-tween the MPPT and the constant power regions. The ideal windpower curve in this case looks as in Fig. 12 [27].

In the proposed controller, the voltage loop can be used toregulate the voltage to a constant level in region II. It is requiredby the controller to know the power levels Pmax and PL toactivate the proper controller in each region. Pmax comes usuallyfrom the turbine manufacturer. However, PL where the constantspeed region begins can be defined by prior lab testing beforeinstallation or can be supplied from the manufacturer. However,PL at which the upper speed limit is hit, is not constant during thecourse of the turbine employment in the field, because of severalreasons, such as the variable losses with aging, and parametervariation due to environmental conditions. So, in this paper, itis suggested to implement a completely blind controller to themanufacturer specifications and at the same time insensitive toparameter variation.

The speed limit of the generator corresponds to a voltage limitat the output of the rectifier. An auxiliary algorithm is imple-mented such that it will record the power level at every timethe voltage hits its limit VdcL (VdcL is obtained experimentallyby pre system testing), the recorded power is PL . PL is up-dated only whenever the recorded value exceeds the previousstored one, so the auxiliary algorithm will keep tracking of thepower limit PL making the controller independent of the turbineparameters prior knowledge and adaptive to any changes.

When the wind speed is below rated, the MPPT controlleris activated. In this region, the generator speed increases forincreasing wind speed and the rectified dc-link voltage will in-crease as well. When the rectified voltage hits its limit; VdcL ,the voltage loop is activated, and voltage regulation takes placethroughout region II. The extracted power in this region in-creases with increasing wind speed as well, however, when thepower reaches its maximum Pmax , the power loop is activatedto drive the generator into the stall region and regulate the powerto a constant level. In this region, the generator speed and hence,

the dc voltage is decreased with increasing wind speed to main-tain constant power level.

While the wind speed is increasing, transitioning the controlfrom the region I to region II is done by detecting the voltagelimit VdcL . And from region II to region III, by detecting thepower limit Pmax . When the wind speed is decreasing on theother hand; moving from region III back to region II is heldby detecting when the power falls below Pmax . However, thetransition from region II back to region I, cannot be done usingthe voltage limit VdcL . The power limit PL is used insteadas this paper proposes. Whenever the power falls below thepower limit PL , the MPPT control is activated again with thesame transition controller strategy proposed for constant powerstall controller. Previous literature in the stall control used thepredefined relations to mitigate the controller transitioning task.However, in this paper, the transition is completely blind to anypredefined relation.

VI. EXPERIMENTAL RESULTS AND DISCUSSION

The schematic diagram for the designed WECS is shown inFig. 13, and the respective hardware of the system is shown inFig. 14. The wind turbine is emulated by using an IPM motordriven by a commercial inverter unit. The wind speed and shaftspeed are taken as inputs and the reference torque is generatedas control input to the IPM motor according to (5) and (6). Thegenerator is coupled to the motor and a three-phase diode bridgeis used to rectify the generated ac voltage and is followed by aboost converter. The turbine is started free of any load torqueand slowly the boost will build the torque on the generator shafttill it catches the MPP and after then the full control strategy isapplied. The shut-down procedure at very high wind speeds isdone using mechanical breakers.

Fig. 15 shows the performance of the MPPT algorithm underfast rate step changing wind profile. When the wind speed issteady or has slow rate of change, the normal P&O mode isactivated and fine tracking to the actual MPP takes place. Whenthe wind speed changes rapidly (step change in the figure), theprediction mode is activated and a large current step is appliedto compensate for the changing turbine torque. The predictionmode is responsible of bringing the operating point near theMPP and to prevent the generator from stalling under fast windspeed slow down scenario.

Fig. 16 shows the experimental verification of the proposedconstant power stall control strategy. In region 1, the wind speedis 8 m/s and the MPPT control is active resulting in maximumpower capture. In region 2, the wind speed is raised to 10 m/swhile the MPPT control is still active resulting in higher powercapture.

The wind speed is raised to 12 m/s in region 3, the capturedpower momentarily increases, and the stall region controller isactivated because the captured power is more than the maximumpreset value of 1000 W. The current limit iLmax is applied, whichmeans large torque on the machine, forcing the generator to slowdown and enters the stall region, after that, the power loop startsto regulate the captured power to the preset value of 1000 Wwhile keeping the speed and hence the voltage, at low levels as

1090 IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 28, NO. 4, DECEMBER 2013

Fig. 13. Schematic for the designed WECS with the proposed controller strategy.

Fig. 14. Hardware of the experimental setup.

Fig. 15. MPPT control performance under variable wind speed conditions.iL (5 A/div), vdc (20 V/div), vo (50 V/div), Pdc (375 W/div), and t (1 s/div).

can be seen in the figure. In region 4, the wind speed goes furtherto 13 m/s, and accordingly the controller drives the generatordeeper in the stall region. In this region, the voltage decreasesand the current increases while the power is kept regulated atits maximum. In region 5, the wind speed is stepped down to10 m/s. At that region, the transition controller is activated and

Fig. 16. Performance of the overall proposed controller strategy under MPPTand stall region control modes. iL (10 A/div), vdc (20 V/div), vo (50 V/div),Pdc (375 W/div), and t (2 s/div).

sends a nearly zero current command (which means zero torquecommand to the generator). The generator starts acceleratingwith the absence of the load torque and leaves the stall region.The decision block detects when the voltage slope starts todecrease knowing by which that the system has left the stallregion and entered the right half side of the torquespeed curve.Then it activates the MPPT control in region 6 and starts with acurrent command that is proportional to the slope of the dc-linkvoltage to bring the operating point close to the MPP. Comparingregion 6 with region 2, it shows that the transition controller isable to move fast to the MPP whenever the wind speed comesdown below rated value.

Fig. 17 shows the performance of the other stall control strat-egy that includes the constant speed (viz. voltage) operation.The maximum voltage limit is set to 50 V and the maximumpower to 1000 W. In region 1, the wind speed is 9 m/s, and theconverter is working in the MPPT region. The power and voltageare below rated values. The wind speed rises to 10 m/s in region2. While the converter is attempting to follow the MPP, the volt-age limit is hit and thus the voltage loop is activated to regulatethe voltage at its maximum. In region 3, the wind speed goeshigher to 11 m/s. The constant voltage stall still in action, so thevoltage remains regulated while the captured power increases.

DALALA et al.: NEW OVERALL CONTROL STRATEGY FOR SMALL-SCALE WECS IN MPPT AND STALL REGIONS 1091

Fig. 17. Performance of the overall proposed controller strategy under MPPTand stall region control modes with constant voltage stall employed. iL(10 A/div), vdc (10 V/div), vo (100 V/div), Pdc (375 W/div), and t (2 s/div).

In region 4, the wind speed further increased to 13 m/s. Whilethe controller is trying to regulate the voltage not to exceed itslimit by increasing the current, the power limit of the system isreached. The constant power controller is triggered and startsto drive the generator deeper into the stall region to maintainthe power below its limit. The speed goes down and hence thevoltage as well. In region 5, the power captured is reduced as aresult of wind speed slow down to 10 m/s. The power capturedis less than the system limit Pmax but more than the power limitPL , so the controller decides to activate the constant voltageregulation controller again. It starts by decreasing the current(viz. torque) considerably to let the generator accelerates till thevoltage reaches its limit and then starts the regulation. In region6, the wind speed goes up again and constant power controltakes place again similar to region 4. In region 7, the wind speedstepped down to 8 m/s. the captured power in this case goesbelow PL , and the controller decides to go back to the MPPTcontrol. Similar transition to that shown in Fig. 16 takes placeto get the generator rapidly to the new MPP.

VII. CONCLUSIONIn this paper, a new overall control strategy for small-scale

WECS has been proposed. The proposed strategy controls thesystem under MPPT mode, while the wind speed is below therated value, and employs soft stalling control while the windspeed is above rated.

In the MPPT region, the conventional P&O technique is mod-ified and adopted. In the above rated wind speed conditions,two stall controllers have been considered and implemented:the constant power and constant voltage stall. The proposedstrategy uses the cascaded loop control concept to regulate thecaptured. The stall region has been analyzed and the dynamicshas been derived. The proposed control structure compensatesfor the instability associated with the stall region. A controlmode transfer structure is proposed to effectively transfer be-tween the MPPT and stall regions. The dc-link voltage slopeinformation are utilized during mode transfer to rapidly move

the operating point near the new MPP location when the systemtransfers from the stall mode to the MPP mode.

A lab hardware test setup has been built to verify the proposedstrategy, and various testing conditions have been applied. Theexperimental results show the effectiveness of the proposed con-troller under various operating conditions. The advantages of theproposed strategy include the simplicity and easy implementa-tion. The mechanical sensors are totally avoided which helpsreducing the system cost. And the reliability is a key advantageof the proposed controller as it is independent of the systemparameter variation.

REFERENCES

[1] C. Zhe, J. M. Guerrero, and F. Blaabjerg, A review of the state of theart of power electronics for wind turbines, IEEE Trans. Power Electron.,vol. 24, no. 8, pp. 18591875, Aug. 2009.

[2] H. Li and Z. Chen, Overview of different wind generator systems andtheir comparisons, IET Renewable Power Generation, vol. 2, pp. 123138, 2008.

[3] M. Liserre, R. Cardenas, M. Molinas, and J. Rodriguez, Overview ofmulti-MW wind turbines and wind parks, IEEE Trans. Ind. Electron.,vol. 58, no. 4, pp. 10811095, Apr. 2011.

[4] I. Serban and C. Marinescu, Sensorless control for small wind turbineswith permanent magnet synchronous generator, in Proc. 2011 IEEE Int.Symp. Ind. Electron. (ISIE), 2011, pp. 14821487.

[5] M. Chinchilla, S. Arnaltes, and J. C. Burgos, Control of permanent-magnet generators applied to variable-speed wind-energy systems con-nected to the grid, IEEE Trans. Energy Convers., vol. 21, no. 1, pp. 130135, Mar. 2006.

[6] Y. Xibo, W. Fei, D. Boroyevich, L. Yongdong, and R. Burgos, DC-linkvoltage control of a full power converter for wind generator operating inweak-grid systems, IEEE Trans. Power Electron., vol. 24, no. 8, pp. 21782192, Sep. 2009.

[7] C. H. Ng, M. A. Parker, R. Li, P. J. Tavner, J. R. Bumby, and E. Spooner,A multilevel modular converter for a large, light weight wind turbinegenerator, IEEE Trans. Power Electron., vol. 23, no. 3, pp. 10621074,May 2008.

[8] S. Morimoto, H. Nakayama, M. Sanada, and Y. Takeda, Sensorless outputmaximization control for variable-speed wind generation system usingIPMSG, IEEE Trans. Ind. Appl., vol. 41, no. 1, pp. 6067, Jan. 2005.

[9] K. Tan and S. Islam, Optimum control strategies in energy conversionof PMSG wind turbine system without mechanical sensors, IEEE Trans.Energy Convers., vol. 19, no. 2, pp. 392399, Jun. 2004.

[10] Y. Xia, K. H. Ahmed, and B. W. Williams, A new maximum power pointtracking technique for permanent magnet synchronous generator basedwind energy conversion system, IEEE Trans. Power Electron., vol. 26,no. 12, pp. 36093620, Dec. 2011.

[11] K. Nishida, T. Ahmed, and M. Nakaoka, A cost-effective high-efficiencypower conditioner with simple MPPT control algorithm for wind-powergrid integration, IEEE Trans. Ind. Appl., vol. 47, no. 2, pp. 893900,Mar./Apr. 2011.

[12] Z. M. Dalala, Z. Zahid, W. Yu, Y. Cho, and J. S. Lai, Design and analysisof an MPPT technique for small-scale wind energy conversion systems,IEEE Trans. Energy Convers., vol. 28, no. 3, pp. 756767, Sep. 2013.

[13] B. Neammanee, S. Sirisumranukul, and S. Chatratana, Control perfor-mance analysis of feedforward and maximum peak power tracking forsmall-and medium-sized fixed pitch wind turbines, in Proc. 9th Int. Conf.Control, Autom., Robot. Vision (ICARCV 06), 2006, pp. 17.

[14] R. C. Portillo, M. M. Prats, J. I. Leon, J. A. Sanchez, J. M. Carrasco,E. Galvan, and L. G. Franquelo, Modeling strategy for back-to-backthree-level converters applied to high-power wind turbines, IEEE Trans.Ind. Electron., vol. 53, no. 5, pp. 14831491, Oct. 2006.

[15] T. Ahmed, K. Nishida, and M. Nakaoka, Advanced control of PWM con-verter with variable-speed induction generator, IEEE Trans. Ind. Appl.,vol. 42, no. 4, pp. 934945, Jul./Aug. 2006.

[16] E. Hau, Wind Turbines: Fundamentals, Technologies, Application, Appli-cation, Economics. Berlin, Germany: Springer, 2005.

[17] F. Blaabjerg, C. Zhe, and S. B. Kjaer, Power electronics as efficientinterface in dispersed power generation systems, IEEE Trans. PowerElectron., vol. 19, no. 5, pp. 11841194, Sep. 2004.

1092 IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 28, NO. 4, DECEMBER 2013

[18] E. Koutroulis and K. Kalaitzakis, Design of a maximum power track-ing system for wind-energy-conversion applications, IEEE Trans. Ind.Electron., vol. 53, no. 2, pp. 486494, Apr. 2006.

[19] R. M. Hilloowala and A. M. Sharaf, A rule-based fuzzy logic controllerfor a PWM inverter in a stand alone wind energy conversion scheme,IEEE Trans. Ind. Appl., vol. 32, no. 1, pp. 5765, Jan./Feb. 1996.

[20] A. Miller, E. Muljadi, and D. S. Zinger, A variable speed wind turbinepower control, IEEE Trans. Energy Convers., vol. 12, no. 2, pp. 181186,Jun. 1997.

[21] R. Chedid, F. Mrad, and M. Basma, Intelligent control of a class of windenergy conversion systems, IEEE Trans. Energy Convers., vol. 14, no. 4,pp. 15971604, Dec. 1999.

[22] A. S. Neris, N. A. Vovos, and G. B. Giannakopoulos, A variable speedwind energy conversion scheme for connection to weak AC systems,IEEE Trans. Energy Convers., vol. 14, no. 1, pp. 122127, Mar. 1999.

[23] R. Datta and V. T. Ranganathan, A method of tracking the peak powerpoints for a variable speed wind energy conversion system, IEEE Trans.Energy Convers., vol. 18, no. 1, pp. 163168, Mar. 2003.

[24] W. Quincy and C. Liuchen, An intelligent maximum power extractionalgorithm for inverter-based variable speed wind turbine systems, IEEETrans. Power Electron., vol. 19, no. 5, pp. 12421249, Sep. 2004.

[25] F. Blaabjerg, Z. Chen, R. Teodorescu, and F. Iov, Power electronics inwind turbine systems, in Proc. CES/IEEE 5th Int. Power Electron. MotionControl Conf. (IPEMC 2006), 2006, pp. 111.

[26] A. Ahmed, R. Li, and J. R. Bumby, New constant electrical power soft-stalling control for small-scale VAWTs, IEEE Trans. Energy Convers.,vol. 25, no. 4, pp. 11521161, Dec. 2010.

[27] J. Chen and C. Gong, New overall power control strategy for variable-speed fixed-pitch wind turbines within the whole wind velocity range,IEEE Trans. Ind. Electron., vol. 60, no. 7, pp. 26522660, 2013.

[28] E. Muljadi, K. Pierce, and P. Migliore, Control strategy for variable-speed, stall-regulated wind turbines, in Proc. 1998 Amer. Control Conf.,vol. 3, pp. 17101714.

[29] F. D. Bianchi, H. De Battista, and R. J. Mantz, Optimal gain-scheduledcontrol of fixed-speed active stall wind turbines, IET Renewable PowerGeneration, vol. 2, pp. 228238, 2008.

[30] N. Rosmin, S. Samsuri, M. Y. Hassan, and H. A. Rahman, Power opti-mization for a small-sized stall-regulated variable-speed wind turbine, inProc. 2012 IEEE Int. Power Eng. Optimization Conf. (PEDCO), Melaka,Malaysia, 2012, pp. 373378.

[31] T. Ekelund, Speed control of wind turbines in the stall region, in Proc.Third IEEE Conf. Control Appl., 1994, vol. 1, pp. 227232.

[32] M. Yundong, H. Zurong, W. Junqi, L. Rixin, and X. Yan, Research onfixed-pitch wind turbine running in deep stall region, in Proc. WorldNon-Grid-Connected Wind Power Energy Conf. (WNWEC 2009), 2009,pp. 16.

[33] S. M. Raza Kazmi, H. Goto, G. Hai-Jiao, and O. Ichinokura, Review andcritical analysis of the research papers published till date on maximumpower point tracking in wind energy conversion system, in Proc. EnergyConver. Congr. Expo. (ECCE), 2010, pp. 40754082.

[34] E. Muljadi and C. P. Butterfield, Pitch-controlled variable-speed windturbine generation, IEEE Trans. Ind. Appl., vol. 37, no. 1, pp. 240246,Jan./Feb. 2001.

[35] T. K. Barlas and G. A. M. V. Kuik, State of the art and prospectivesof smart rotor control for wind turbines, J. Phys. Conf. Ser., vol. 75,p. 012080, 2007.

[36] A. E. Haniotis, K. S. Soutis, A. G. Kladas, and J. A. Tegopoulos, Gridconnected variable speed wind turbine modeling, dynamic performanceand control, in Proc. IEEE PES Power Syst. Conf. Expo., 2004, vol. 2,pp. 759764.

[37] Z. M. Dalala, Z. Zahid, and J. S. Lai, New overall control strategy forwind energy conversion systems in MPPT and stall regions, in Proc.Energy Conver. Congr. Expo., Sep. 2013, pp. 24122419.

[38] S. M. R. Kazmi, H. Goto, G. Hai-Jiao, and O. Ichinokura, A novel al-gorithm for fast and efficient speed-sensorless maximum power pointtracking in wind energy conversion systems, IEEE Trans. Ind. Electron.,vol. 58, no. 1, pp. 2936, Jan. 2011.

[39] O. W. P. C. Krause and S. D. Sudhoff, Analysis of Electric Machinery.New York, NY, USA: IEEE Press, 1994.

[40] H. B. Zhang, J. Fletcher, N. Greeves, S. J. Finney, and B. W. Williams,One-power-point operation for variable speed wind/tidal stream turbineswith synchronous generators, IET Renewable Power Generation, vol. 5,pp. 99108, 2011.

[41] T. M. U. N. Mohan and W. P. Robbins, Power Electronic: Converter, Ap-plication and Design. New York, NY, USA: Wiley, 1995.

[42] M. H. Rashid, Power Electronics Circuits, Devices, and Applications.Englewood Cliffs, NJ, USA: Prentice-Hall, 1993.

[43] A. M. Knight and G. E. Peters, Simple wind energy controller for anexpanded operating range, IEEE Trans. Energy Convers., vol. 20, no. 2,pp. 459466, Jun. 2005.

Zakariya M. Dalala (S13) received the B.S. degreein electrical engineering from Jordan University ofScience and Technology, Irbid, Jordan, and the M.S.degree in electrical engineering from the Universityof Jordan, Amman, Jordan, in 2005 and 2009, re-spectively. He is currently working toward the Ph.D.degree in electrical engineering at the Virginia Poly-technic Institute and State University, Blacksburg,USA.

His current research interests include power elec-tronic converters design for renewable energy sys-

tems, digital control, and high-performance motor drives.

Zaka Ullah Zahid (S13) received the B.S. degreein electrical and electronics engineering from NWFPUniversity of Engineering and Technology (UET),Peshawar, Pakistan, and the M.S. degree in electri-cal engineering from George Washington University(GWU), Washington, DC, USA, in 2007 and 2009, re-spectively. He is currently working toward the Ph.D.degree in electrical engineering at Virginia Polytech-nic Institute and State University, Blacksburg, USA.

His current research interests include design andcontrol of transformer isolated dcdc converter.

Jih-Sheng (Jason) Lai (S85M89SM93F07)received the M.S. and Ph.D. degrees in electrical engi-neering from the University of Tennessee, Knoxville,USA, in 1985 and 1989, respectively.

In 1989, he joined the Electric Power Research In-stitute (EPRI) Power Electronics Applications Center(PEAC), where he managed EPRI-sponsored powerelectronics research projects. In 1993, he then join theOak Ridge National Laboratory as the Power Elec-tronics Lead Scientist, where he initiated a high powerelectronics program and developed several novel high

power converters including multilevel converters and soft-switching inverters.In 1996, he joined the Virginia Polytechnic Institute and State University, wherehe is currently the James S. Tucker Professor in the Electrical and ComputerEngineering Department and the Director of Future Energy Electronics Center.He has authored or coauthored more than 300 technical papers and 2 books andreceived 22 U.S. patents. His current research interests include high efficiencypower electronics conversions for high power and energy applications.

Dr. Lai is the recipient of several distinctive awards, including a Techni-cal Achievement Award in Lockheed Martin Award Night, four prize paperawards from IEEE IAS and best paper awards from IECON-97, IPEC-05, andPCC-07. His student teams won the first place award from 2009 TI EnginousPrize Analog Design Competition and 2011 Grand Prize Award from Inter-national Future Energy Challenge. He chaired the 2000 IEEE Workshop onComputers in Power Electronics (COMPEL 2000), 2001 IEEE/DOE FutureEnergy Challenge, and 2005 IEEE Applied Power Electronics Conference andExposition (APEC 2005).

/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages false /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False

/Description >>> setdistillerparams> setpagedevice