review megha goyal, yuanyuan fan, arindam ghosh* and ...€¦ · megha goyal, yuanyuan fan, arindam...

Review

Megha Goyal, Yuanyuan Fan, Arindam Ghosh* and Farhad Shahnia

Techniques for a Wind Energy System Integrationwith an Islanded Microgrid

DOI 10.1515/ijeeps-2015-0139Received August 14, 2015; revised February 4, 2016;accepted February 5, 2016

Abstract: This paper presents two different techniques ofa wind energy conversion system (WECS) integration withan islanded microgrid (MG). The islanded microgrid oper-ates in a frequency droop control where its frequency canvary around 50 Hz. The permanent magnet synchronousgenerator (PMSG) based variable speed WECS is consid-ered, which converts wind energy to a low frequency acpower. Therefore it needs to be connected to the microgridthrough a back to back (B2B) converter system. One wayof interconnection is to synchronize the MG side converterwith the MG bus at which it is connected. In this case, thisconverter runs at the MG frequency. The other approach isto bring back the MG frequency to 50 Hz using the iso-chronization concept. In this case, the MG side converteroperates at 50 Hz. Both these techniques are developedin this paper. The proposed techniques are validatedthrough extensive PSCAD/EMTDC simulation studies.

Keywords: Microgrid, diesel generator, droop control,isochronization, WECS

1 Introduction

The notion of the smart grid is the incorporation of alltechnologies, topologies and approaches that benefit uti-lities and customers. A microgrid (MG) is a key compo-nent of smart grid. A MG is defined as the interconnectionof several distributed generators (DGs) such as dieselgenerators (DGENs), micro turbines, gas turbines, windfarms, PV arrays and battery storage systems. A micro-grid can be connected to or separated from the electricitygrid [1].

The most common way to supply electricity to remotecustomers is with diesel generators [2]. However, powergeneration based on diesel fuel can be offset by windenergy. Wind resource is the most promising energychoice due to its enormous availability. The developmentof new technologies for renewable sources offers attrac-tive economic and environmental merits for energy sup-port [3]. In Canada, integration of wind energy indistribution network to meet energy requirement inrural and remote areas is growing rapidly [4]. In Rameawind-diesel project, a wind generator is integrated withthe diesel generators to supply an islanded system [4].Various designing aspects are reported for diesel genera-tors operating in conjunction with wind turbines andenergy storage in [5]. The design methodology and ana-lysis approach for unit sizing of an islanded wind-dieselsystem is developed in [6].

This research discusses the integration techniques ofa variable speed wind system with an islanded microgrid.In [7], an integration of a doubly-fed induction generator(DFIG) based wind generator within microgrid is intro-duced. It focuses on variable droop control for DFIG toadjust the output power according to available windpower. The case study of Rhodes Island power system isinvestigated in [8]. It introduces a frequency controller forPMSG and DFIG types of wind turbines to regulate fre-quency in the islanded microgrid. In frequency droopcontrol, frequency varies according to load variation. Inliterature, various kinds of droop control strategies havebeen proposed [9–11]. In [12], isochronous load sharing isdiscussed for inverter based distributed generation. Inthis, the inverter operates at constant set reference fre-quency regardless of load. However, load power mea-surement is required to achieve this.

The frequency regulation in an islanded microgridusing a wind power system is discussed in [13], where thefrequency of the microgrid is maintained constant irrespec-tive of load variation through wind system. The controllerregulates power from the wind system according to loadvariation in the microgrid. Thus, the wind system is usedonly to maintain the microgrid frequency constant ratherthan using available renewable power to support the load.

*Corresponding author: Arindam Ghosh, Department of Electricaland Computer Engineering, Curtin University, Perth 6102, WesternAustralia, E-mail: [email protected] Goyal, Yuanyuan Fan, Farhad Shahnia, Department ofElectrical and Computer Engineering, Curtin University, Perth 6102,Western Australia

Int. J. Emerg. Electr. Power Syst. 2016; 17(2): 191–203

Brought to you by | Murdoch University LibraryAuthenticated

Download Date | 10/9/17 5:05 AM

In this paper, an optimal power control (OPC) schemeis used for maximum power point track (MPPT) in WECS.The PMSG based WECS is considered, which is connectedto the microgrid through a back to back (B2B) voltagesource converter (VSC). This B2B converter is required toconnect low frequency generator of the WECS to themicrogrid which is operating around the set referencefrequency of 50 Hz. This paper mainly focuses on inte-gration techniques of WECS with an islanded microgrid.The islanded microgrid operates in frequency droop con-trol. To integrate the WECS at the set reference frequencywith the islanded microgrid, modified droop control withisochronous concept is proposed to maintain the systemfrequency constant with the load fluctuation and windpower variation. Moreover, another integration techniqueusing synchronization concept is also presented in whichisochronization is not required. Extensive digital compu-ter simulation results using PSCAD/EMTDC are presentedto validate the proposals.

2 System structure

The system considered has a microgrid with two dieselgenerators (DGENs) and a WECS that operates in MPPT(Figure 1). The diesel generator set consists of a 4-strokeinternal combustion (IC) engine coupled to a synchronousgenerator. The IC engine speed is controlled through thefuel input rate by a speed governor. A PID controller isused in the governor to maintain output speed. The syn-

chronous generator contains an exciter and an automaticvoltage regulator (AVR). The AVR controls the field supplyof the generator to maintain the required terminal voltage.Note that the reference speed (for governor) and generatorterminal voltage (for AVR) are obtained from the frequencydroop control. The WECS is connected to the microgridthrough a back to back converter. The system data usedare given in the Appendix. The BTB converter consists oftwo conventional voltage source converters (VSC-1 andVSC-2) connected through a common dc link capacitor.VSC-1 is used to control the power flow from the windturbine, while VSC-2 is used to control the dc link capa-citor voltage (Vdc) constant. The microgrid frequency var-ies within the droop limits according to the loadvariations. VSC-2 connected to the microgrid, should oper-ate at the same frequency of the microgrid. To achievethis, two techniques are proposed: (1) synchronization ofthe VSC-2 voltage at the same frequency of the microgridand (2) modifying the frequency droop control such thatthe microgrid always operates at the set referencefrequency.

3 Wind energy conversion system

3.1 Wind turbine with PMSG model

The wind power that is captured by the blade and con-verted into mechanical power can be calculated by [14]

P1 Q1 P2 Q2

Load

Z23

E2∠δ2E1∠δ1

DGEN-1

Z12

DGEN-2

PMSG

Wind Turbine

WPMicrogrid

VSC-1 VSC-2Vdc

+PID-Controller AVR

FieldExcitation

Fuel Control

+ -

Engine

ωg

VT

Vω +

Generator

Figure 1: System structure.

192 M. Goyal et al.: Techniques for a WECS Integration with an Islanded MG



PM =12ρAv3wCp (1)

where ρ is the air density, A is the cross-sectional areathrough which the wind passes, vw is the wind speed andCp is the power coefficient of the blade. The rotor effi-ciency Cp is calculated from [15]

Cp =0.5176116λi

−0.4β− 5� �

e− 21λi +0.0068λ (2)

where β is the pitch angle, λ is the tip speed ratio (TSR),which is defined as the blade tip moving speed dividedby the wind speed, and λι is given by [16]

1λi

=1

λ+0.08β−0.035

β3 + 1(3)

The wind turbine operates at the generator control modewhen the wind speed is below the rated wind speed, andworks under the pitch control mode when the wind speedexceeds the rated value [17]. The parameters of the windturbine and PMSG used are given in the Appendix.

3.2 PMSG operation with varied windspeeds

To verify the modelling and control strategy applied inthis paper, the varied wind speeds are written as thefollowing equations:

vw =vm t < t1vm + k � ðt − t1Þ t1 < t < t2vm + k � ðt2 − t1Þ t > t2

8<: (4)

vES = k � ðt2 − t1Þ (5)

where vw is the wind speed available to the turbine, vm isthe mean wind speed, k is the ramp change in the windspeed, t1 and t2 are the starting and ending time respec-tively of the wind speed ramping duration, vES is theexternal wind speed of the wind source.

3.3 OPC based MPPT for PMSG

The schematic diagram of the PMSG with B2B voltagesource converters is shown in Figure 2. The kinetic energyof the wind is converted into mechanical energy by thewind turbine and then transmitted to the generator. VSC-1controls the active power through MPPT, while VSC-2maintains the DC capacitor voltage constant.

To extract the maximum power from the windenergy, turbine blades should change their speed as thewind speed changes. Reference [18] gives out three

methods to realize the MPPT control. Based on thePMSG model in this paper, to control the generatorpower, a method similar to Optimal Torque Control(OTC) is applied. This is called Optimal Power Control(OPC). The principle of the OTC is that the wind turbinemechanical torque T and the turbine speed ωp have thefollowing relationship at MPPT [18]

T∝ω2p (6)

Assuming that the generator power is denoted by Pg,which is given in OPC as

Pg ∝ω3p (7)

It should be noted that the values of the turbine speed andthe generator speed are equal, considering that the PMSGmodel is a direct drive. Besides, the generator mechanicaltorque is equal to its electromagnetic torque in the steadystate. Equation (7) is the basis of OPC. Suppose

Pg =Kopt ×ω3p (8)

where Kopt is calculated according to the generator ratedparameters as

Kopt = P*

ω*pð Þ3

ω*p = 2π � f *p

8<: (9)

In (9), P*, ω*p, f * and p are the rated power, rated speed,rated frequency and pole pairs of the generator.

3.4 Control of B2B Converters

To control MPPT wind power flow to the microgrid, VSC-1generates the voltage across the filter capacitor (Cfw) withan angle deviation from the PMSG output voltage. Thispower angle is calculated by a PI controller as

δ1 =KP1 Pg − P� �

+KI1

ðPg − P� �

dt (10)

where Pg is the wind power calculated from (8) and P isthe actual power from the PMSG.

The purpose of the VSC-2 is to hold the voltage (Vdc)across the DC link capacitor constant. Note that thisvoltage will remain constant only when it neither sup-plies nor absorbs any real power. Therefore the powerobtained from wind turbine should ideally appear at thegrid side. However, this is not practical since the conver-ter losses must also be supplied from the generatedpower. Therefore holding the capacitor voltage is tanta-mount to extracting the maximum possible power fromthe wind turbine after supplying the converter losses.

M. Goyal et al.: Techniques for a WECS Integration with an Islanded MG 193



For the DC voltage control another PI controller (PI-2) isdesigned, which is given by

δ2 =KP2 Vdc, ref −Vdc� �

+KI2

ðVdc, ref −Vdc� �

dt (11)

where Vdc,ref is the reference dc capacitor voltage and Vdc

is the actual dc capacitor voltage. KP2 and KI2 are propor-tional and integral gains of the controller respectively.The converters are controlled in a hysteresis band outputfeedback control as explained in [19].

To illustrate the operation of OPC based MPPT con-trol, consider the WECS that is connected to an infinitebus (Figure 2). The wind speed pattern is assumed as

t1 = 3 s, t2 = 4 s, k = 1, Vm = 10 m=s

Figure 3(a) shows the wind speed variation pattern. Inthis the mean wind speed is 10 m/s, which ramps upbetween 3 s and 4 s to 11 m/s. The power output of theWECS follows the same pattern of the wind speed, as canbe seen from Figure 3(b). The TSR versus wind speed isshown in Figure 3(c), where the TSR is kept at its optimalvalue regardless of the change in wind speed. The smallfluctuation that can be seen in TSR is due to the controlof VSC-1, which cannot respond istantaneously becauseof the dynamics of its LC filters. The rotor efficiency (Cp)against wind speed is plotted in Figure 3(d). It is alsoconstant follwing the TSR. Figures 3(c) and (d) demon-strate that a stable maxium efficiency is being achievedwhen wind speed is varying. This indicates the effective-ness of the applied MPPT control. The WECS is connectedwith the utility. Therefore, VSC-2 operates at the fix fre-quency (ωg) of the utility. When the WECS gets integratedwith the MG, ωg cannot remain constant. Therefore mea-sures have to be taken to eliminate any system oscillationdue to frequency mismatch.

4 Integration of WECS withmicrogrid

Since the microgrid operates in the frequency droop con-trol, the frequency can vary within the limits of a fre-quency band. Therefore, VSC-2 must be synchronizedwith the microgrid frequency. As mentioned before, twotechniques are proposed for the interconnection. Theseare discussed below.

4.1 Frequency droop control withisochronization

In this technique, the frequency of the microgrid is heldconstant at 50 Hz by integrating isochronization withfrequency droop control. Therefore, the VSC-2 can con-nect to the microgrid at the set reference frequency. In theislanded microgrid, the DGs operate in the conventionalfrequency droop control, given by

ω = ω* +m × ð0.5 ×P* − PÞ (12)

where the instantaneous and rated frequency of the DGsare ω and ω* respectively. The rated and measured realpowers are represented by P* and P respectively. Thedroop coefficient is denoted by m. The droop parametersof all DGs in a microgrid depend on their respectiveratings such that all of them operate at the same fre-quency in the steady state. How these droop parameterscan be calculated has been explained in [20].

Equation (12) signifies that when the system fre-quency is rated (50 Hz), the DG should supply half ofits rated power. The frequency will be below (above) 50Hz when the DG supplies more (less) than half of its ratedpower. The droop gain is calculated based on a frequency

Figure 2: Schematic diagram of wind energy conversion system.




limit of ± 0.3 Hz. In isochronous mode, the frequency ofthe microgrid must remain constant at 50 Hz. Therefore,if the droop control and isochronous mode are combined,the system can operate at the set reference frequencywhile sharing load power according to DG ratings.

To achieve this, the droop line shifts according to theload variation while preserving its slope. For constantdroop coefficient, each DG has its own droop line accord-ing to its rating with the constant droop coefficient. Thedroop line shifts in such a way that it retains the sameslope and can supply load power demand at the setreference frequency.

The droop lines of Figure 4 are considered for one ofthe DGs. Ordinarily, the DG operates in line-1. Thisimplies that when it supplies P1, the frequency shouldbe ω1. However, it is required that the DG operates at 50Hz (ω*). Therefore the frequency must be compensatedby the amount Δω=ω*– ω1. We also need to maintain thepower sharing, which depends on the slope of the drooplines. Therefore the quantity Δω must be kept zeroagainst any frequency variation due to load change. Toachieve this, a PID controller is used to set frequencyvariation to zero. This is shown in Figure 5. The outputof the PID controller is added to the output of the droopcontroller to produce ω1′, which is the input of the DGENspeed governor. Note that in steady state, ω1′=ω*. Thismeans that the droop line is now shifted to line-2, whichgives the power output of P1 at ω*.

4.2 Synchronization with microgrid

In this technique, the microgrid operates in conventionalfrequency droop control and its frequency varies withinthe frequency band according to the load power demand.Therefore, to connect the WECS with the microgrid, VSC-2must be synchronized with the microgrid to operate at the

Figure 3: Wind speed variation, output power, TSR and rotor efficiency.

Figure 4: Frequency droop control with isochronous mode.




same frequency (Figure 6). To achieve this, a simplealgorithm is used for synchronization, which is discussedbelow.

The frequency of themicrogrid is set atω. The instantaneousvoltages at the point of connection of VSC-2 are given as

va =V ×ffiffiffi2

psin ω t +ϕð Þ

vb =V ×ffiffiffi2

psin ω t − 120� +ϕð Þ

vc =V ×ffiffiffi2

psin ω t + 120� +ϕð Þ

(13)

Using symmetrical component theory [21], the positivesequence component of the instantaneous voltage of(13) will be

va1 = va + vba+ vca2� ��

3

=Vffiffiffi2

p sin ω t +ϕð Þ− j cos ω t +ϕð Þ½ � (14)

va1 =Vffiffiffi2

p α+ jβð Þ (15)

where a= ej120° and

α = sin ωt +ϕð Þ , β = − cos ωt +ϕð Þ (16)

Note that from (11) δ2 is obtained. The purpose is to addthis angle with positive sequence voltage of (14).

Therefore, the positive sequence of the reference vol-tage across the filter capacitor of VSC-2 is

Vvsc2 =Vffiffiffi2

p sin ω t +ϕ+ δ2ð Þ− j cos ω t +ϕ+ δ2ð Þ½ �

Vvsc2 =Vffiffiffi2

p sin ωt +ϕð Þ cos δ2 + cos ωt +ϕð Þ sin δ2½

− j cos ωt +ϕð Þ cos δ2 + j sin ωt +ϕð Þ sin δ2�=

Vffiffiffi2

p α+ jβð Þ cos δ2 − β− jαð Þ sin δ2½ �

(17)

From (16) and (17), Vvsc2 can be calculated. The instanta-neous negative sequence is the complex conjugate of thepositive sequence. Also, since the system is assumed to bebalanced, the instantaneous zero sequence will be zero.Therefore, using inverse symmetrical component transform,the instantaneous reference voltages across the capacitorare obtained. Thus VSC-2 can track this reference outputvoltage without any explicit frequency measurement.

5 Simulation studies

Simulation studies are carried out in PSCAD/EMTDC toverify the discussed techniques above with different pat-terns of wind speed in WECS connected to the microgrid.The parameters of the considered system are given in theAppendix.

Three different case studies are considered. These are(a) Nominal operation of the islanded microgrid(b) WECS integration with the microgrid using

isochronization(c) WECS integration with the microgrid using frequency

synchronization

Case (a): The considered 11 kV microgrid consists of twoDGENs with ratings of 500 kW and 250 kW respectively andits local load. The droop coefficients for each DG are calcu-lated from their ratings. The local load is assumed to be 625kW. As shown in Figure 7(a), both DGENs share the powerin the ratio of 2:1 according to their ratings. The microgridfrequency is 49.8 Hz with the conventional droop controlmethod as shown in Figure 7(b). The synchronization tech-nique is applied at 2 s and it can be seen from Figure 7(b)that the frequency converges to 50 Hz. However the powersremain unaltered. Once the load power changes at 11 s, the

Figure 5: Isochronous controller.

Figure 6: VSC-2 connection with microgrid.




power supplied by the DGs also changes in a same ratio,which is shown in Figure 8(a). However, in Figure 8(b), itcan be seen that with the proposed method, the systemretains at the set reference value of 50 Hz. This verifies thatthe system frequency maintains at the set reference withload power sharing according to the ratings of the DGs.

Case (b): The microgrid operates in frequency droopcontrol with isochronous mode. Therefore, the microgridoperates at the set reference frequency and hence VSC-2can also operate at the same frequency. The results arediscussed below.

Case (b.1): Microgrid with constantspeed WECS

For this case, it has been assumed when the WECS sys-tem is connected to the microgrid at 2 s, the sources areoperating in the steady state. The WECS supplies its totalpower to the microgrid. The real power flow in the micro-grid is shown in Figure 9(a). It can be seen from thisfigure that the power supplied by the DGs reduces pro-portionally, maintaining the sharing ratio. The frequencyof the microgrid is shown in Figure 9(b). It can be seen

Figure 7: Real power sharing and frequency of microgrid with conventional droop and isochronization in Case (a).

Figure 8: Real power sharing and frequency with load variation in Case (a).




that the system frequency is retained at the set referencevalue of 50 Hz even after the integration of the WECS. Thedc link capacitor voltage is shown in Figure 10. It is heldconstant at 2.5 kV barring some transients by VSC-2 evenafter the WECS connects to the microgrid.

Case (b.2): Microgrid with variablespeed WECS

In this case, the wind speed of the WECS is consideredvariable. As shown in Figure 11, initially the wind speed isconstant at 10m/s. Then at 10 s, it starts ramping up to 12m/s till 12 s. The output power ofWECS also follows the patternof the wind speed as shown in Figure 12(a). Once the wind

speed increases, the output power of WECS also increasesand therefore the power supplied by the DGENs reduces.The system frequencywith the variation of the output powerof the WECS is also maintained at the set reference value of50 Hz as shown in Figure 12(b).

Case (b.3): Tripping of WECS

Usually wind turbines start to operate when the wind speedexceeds 4–5 m/s, and are shut off at speeds over 25–30 m/s[14]. If wind speed is less than the cut in speed or higher thanthe cut off speed,WECSwill be disconnected from themicro-grid. In this case, the wind speed is reduced from 10 m/s to

Figure 9: Real power flow in microgrid with WECS and microgrid frequency in Case (b.1).

Figure 10: DC capacitor voltage in Case (b.1).

Figure 11: Wind speed in Case (b.2).




the cut in speed as shown in Figure 13. Therefore, the powerPW fromWECSbecomeszeroafter 11 s. It canbe seen inFigure14(a) that now load power is supplied only from the DGENs.Furthermore, the system frequency is retained constant at theset reference value as shown in Figure 14(b).Case (c): The microgrid operates in conventional fre-quency droop control and the frequency of the microgridvaries within the frquency band according to loadvariation. Therefore, for connecting the WECS with themicrogrid, the synchronization algorithm is used. Thesimulation results are discussed below.

Case (c.1): Microgrid with constantspeed WECS

It is assumed that when WECS connects to the microgridat 2 s, the wind speed is constant at 10 m/s. FromFigure 15(a), it can be seen that the power suppliedfrom the DGENs reduces and the system frequency alsoincreases due to power reduction. This is shown inFigure 15(b). The dc capacitor voltage is held constantby VSC-2 during the connection of WECS with microgridas shown in Figure 16.

It can be seen in Figure 17(a) that once the load isincreased to 625 kW at 10 s, the power of DGENsincreases in a same proportion and the power of WECS(Pw) is retained constant. The frequency of the microgridalso reduces as shown in Figure 17(b).

Case (c.2): Microgrid with variablespeed WECS

In this case, it is assumed that the wind speed rampsup from 10 m/s to 11 m/s. The power flow from theWECS follows the wind speed pattern and thepowers of DGENs reduce according to the available

Figure 12: Power flow in microgridwith WECS and microgrid frequencyin Case (b.2).

Figure 13: Wind speed in Case (b.3).




Figure 14: Power flow in microgrid with WECS and frequency in Case (b.3).

Figure 15: Real power sharing and frequency of the microgrid in Case (c.1).




WECS power and load requirement. This is shown inFigure 18(a). The variation in microgrid frequency isshown in Figure 18(b).

6 Conclusions

In this paper, Optimal Power Control (OPC) scheme isused for MPPT control to draw maximum possiblepower from the WECS. The WECS is connected through

a back to back converter to supply the available max-imum power to the microgrid. Two different techniques ofWECS integration to the microgrid are proposed in thispaper. The microgrid operates in frequency droop con-trol. In the first technique, an isochronization method isused in which the droop line is shifted such that each DGsupplies its required power at 50 Hz. This has been dis-cussed in details. The other technique is to synchronizeWECS at the microgrid frequency. A simple synchroniza-tion method is proposed in which only measurements ofthe instantaneous PCC voltages are needed. Both thesetwo proposed techniques are validated through computersimulation using PSCAD/EMTDC. It has been shown thatboth of them work satisfactorily during load change,WECS connection or disconnection with the microgrid.The operation is seamless where no large transient isvisible.

Acknowledgement: The authors thank the AustralianResearch Council (ARC) for the financial support for thisproject through the ARC Discovery Grant DP110104554.

Funding: Department of Industry and Science, AustralianGovernment, Australian Research Council “DP110104554”.

Figure 17: Real power sharing and frequency of the microgrid with local load variation in Case (c.1).

Figure 16: DC capacitor voltage (Vdc) in Case (c.1).




Appendix

Figure 18: Real power sharing and frequency of microgrid in Case (c.2).

Table 3: Diesel generator set data.

System data Value

Rated voltage kVRated power DGEN kWRated power DGEN kWRated frequency HzRated speed , rpm

Reactance Value (per unit)Xd : Unsaturated d axis

synchronous reactance.

X′d : Unsaturated d axis transientsynchronous reactance

. ×-

X″d : Unsaturated d axissubtransient synchronous

. ×–

Xq : Unsaturated q axissynchronous reactance

. ×–

X″q : Unsaturated q axissubtransient synchronousreactance

. ×–

X : Negative-sequence reactance .X : Zero-sequence reactance . ×–

Time constants Value (ms)t′’do : d axis subtransient open

circuit time constant

(continued )

Table 1: Wind turbine and PMSG parameters.

Parameters names Parameter values

Rotor radius cmAir density . kg/m

Rated wind speed m/sRated apparent power . MVARated line-to-line voltage kVRated frequency HzNumber of pole pairs

Table 2: Parameters of the DGs connected in microgrid.

System quantities Values

DG feeder impedance Rf=. Ω, Lf=. mHDG feeder impedance Rf=. Ω, Lf=. mHDGs rated power DGEN-: kW,

DGEN-: kWDroop coefficient (frequency–voltage)m . rad/MWsm . rad/kWsn . kV/MVArn . kV/MVAr




References

1. Farhangi H. The path of the smart grid. IEEE Power Energy MagJanuary-February 2010;8:18–28.

2. Hunter R, Elliot G. Wind-diesel systems, a guide to the tech-nology and its implementation. Cambridge: CambridgeUniversity Press, 2005.

3. Hier S. Grid integration of wind energy conversion systems.New York: John Wiley and sons, June 2014.

4. Abbey C, Katiraei F, Brothers C, Dignard LB, Joos G,“Integration of distributed generation and wind energy inCanada,” IEEE, Power Engineering Society General Meeting,2006.

5. Lipman NH. Overview of wind/diesel systems. RenewableEnergy August 1994;5:595–617.

6. Katiraei F, Abbey C, Diesel plant sizing and performance ana-lysis of a remote wind-diesel microgrid, IEEE, PowerEngineering Society General Meeting, 1–8, June 2007.

7. Fazeli M, Asher GM, Klumpner C, Liangzhong Y, Bazargan M.Novel integration of wind generator-energy storage systemswithin microgrids. IEEE Trans Smart Grid June 2012;3:728–37.

8. Margaris ID, Papathanassiou SA, Hatziargyriou ND, Hansen AD,Sorensen P. Frequency control in autonomous power systemswith high wind power penetration. IEEE Trans SustainableEnergy April 2012;3:189–99.

9. Planas E, Muro AG, Andreu J, Kortabarria I, Alegria IM. Generalaspects, hierarchical controls and droop methods in micro-grids: a review. Renewable Sustainable Energy Rev January2013;17:147–59.

10. Lopes JAP, Moreira CL, Madureira AG. Defining control strate-gies for microgrids islanded operation. IEEE Trans Power SystMay 2006;21:916–24.

11. Vandoorn TL, Meersman B, De Kooning JDM, Vandevelde L.Analogy between conventional grid control and islandedmicrogrid control based on a global DC-link voltage droop.IEEE Trans Power Deliv July 2012;27:1405–14.

12. Mohd A, Ortjohann E, Sinsukthavorn W, Lingemann M,Hamsic N, Morton D. Isochronous load sharing and controlfor inverter-based distributed generation. Int Conf CleanElectrical Power June 2009;2009:324–9.

13. Bunker KJ, Weaver WW. Microgrid frequency regulation usingwind turbine controls. Power and Energy Conference at Illinois(PECI), Feb-Mar. 2014;1–6.

14. Patel MR. Wind and solar power systems. Boca Raton, FL: CRCPress, 1999:124–30.

15. Jadallah AA, Mahmood DY, Abdulqader ZA. Optimalperformance of horizontal axis wind turbine for low windspeed regime. Int J Multidisciplinary Curr Res 2014;2:159–64.

16. Slootweg JG, Polinder H, Kling WL. Dynamic modeling of windturbine with doubly feed induction generator. IEEE Power Engsociety Summer Meet July 2001;1:644–9.

17. Kurohane K, Senjyu T, Yona A, Urasaki N, Goya T, Funabashi T.A hybrid smart AC/DC power system. IEEE Trans Smart Grid2010;1:199–204.

18. Wu B, Lang Y, Zargari N, Kouro S. Power conversion and control ofwind energy systems. New Jersey: IEEE John Wiley and Sons, 2011.

19. Ghosh A. Performance study of two different compensatingdevices in a custom power park. Proc IEEE Gener TransmDistrib 2005;152:521–8.

20. Goyal M, Ghosh A., A phase-locked-loop design for smoothoperation of a hybrid microgrid, Australasian UniversitiesPower Engineering Conference AUPEC 2013, Hobart, 2013.

21. Lyon WV. Transient analysis of alternating-current machinery.New York: John Wiley, 1954.

Table 4: Parameters of the isochronization (PID)controller.

System data Value

Proportional gain (KPi) .Integral gain (KIi)

Constant coefficient (a)

Table 3: (continued )

System data Value

t′’qo : q axis subtransient opencircuit time constant

t′d : d axis transient open circuittime constant

ta : Armature time constant




review megha goyal, yuanyuan fan, arindam ghosh* and ...€¦ · megha goyal, yuanyuan fan, arindam...

Documents