comparativestudiesofdifferentcontrolstrategiesof...

Hindawi Publishing CorporationAdvances in Power ElectronicsVolume 2012, Article ID 327186, 9 pagesdoi:10.1155/2012/327186

Research Article

Comparative Studies of Different Control Strategies ofa Dynamic Voltage Restorer Based on Matrix Converter

Amin Shabanpour and Ali Reza Seifi

School of Electrical & Computer Engineering, University of Shiraz, Shiraz 7134851151, Iran

Correspondence should be addressed to Ali Reza Seifi, [email protected]

Received 17 May 2012; Accepted 10 September 2012

Academic Editor: Pavol Bauer

Copyright © 2012 A. Shabanpour and A. R. Seifi. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

A dynamic voltage restorer (DVR) with no energy storage is studied. By using a matrix converter instead of the conventionalAC/DC/AC converters, elimination of the DC-link capacitor is possible. The switching algorithm of matrix converter is the spacevector modulation. There are different compensation algorithms to control the conventional DVR. These methods have beenanalyzed in this paper for the proposed matrix-converter-based DVR. A deep analysis through different diagrams would showthe advantages or disadvantages of each compensation method. Equations for all methods are derived, and the characteristics ofalgorithms are compared with each other.

1. Introduction

Advanced industrials usually use electronic devices such asprogrammable logic controllers or power electronic drives intheir devices. Power quality problems may affect the functionof these devices. They could malfunction or trip throughvoltage disturbances. One of the most important power qual-ity problems is voltage sag which is considered to be the mostoccurring issue in the power systems [1]. This phenomenonis specified as a sudden decrease in the voltage from 90% to10% of rated voltage which lasts for half cycle to one minute.The main source of voltage sags are system faults, switchingof heavy loads, or starting of large motors [2].

One of the best solutions to improve power quality is thedynamic voltage restorer (DVR). DVR is a kind of custompower devices that can inject active/reactive power to thegrids. This can protect critical loads from disturbances suchas sags or swells.

Four DVR topologies are compared in [3]. In this paper,the DVR with no energy storage devices and load-side con-nected shunt converter is selected. By using a matrix con-verter instead of the AC/DC/AC converters in that topology,the DC-link and its bulky capacitor would be eliminated.

Generally a matrix converter is a direct AC/AC con-verter. A three-phase matrix converter has nine bidirectional

switches that are able to connect any output phase of thesystem to any input phase at any time [4]. Some advantagesof the matrix converter are bidirectional power flow, con-trollable sinusoidal output voltage, ability to control theinput power factor, and a compact structure. However thecomplexity of its control is its main drawback [5].

Different DVR topologies based on matrix converters arepresented before. A flywheel-supported DVR controlled bythe matrix converter is proposed in [6]. Another topologyis studied in [7] that the behavior of a DVR with matrixconverter utilizing Alesina-Venturini modulation method isshown. In [8], a matrix converter-based DVR is analyzedwhich utilizes the space-vector modulation.

There are three compensation strategies for conventionalDVR in literatures [9–11]. Comparisons between them aredone in [12] for a general type of DVR but there is less studyon compensation strategies for a matrix-converter-basedDVR that is the aim of this paper. All compensation methodsincludes prefault, in-phase, and energy-optimized compen-sation algorithm are analyzed and compared in this paperfor the matrix-converter-based DVR. Equations for eachmethod are derived. Furthermore some parameters to char-acterize the behavior of each method for this special kind ofDVR are defined. Finally, comparisons are done in various

mailto:[email protected]

2 Advances in Power Electronics

Source

Series Shunt

Load

converter converter

Figure 1: A DVR with no energy storage system and load-side-connected converter.

Source Load

Matrixconverter

Figure 2: A DVR based on matrix converter.

aspects to show the characteristics of each compensatingalgorithm. Moreover, there are some studies on the DVR per-formance limitation in this topology.

In this paper, the structure of a DVR with matrix con-verter is introduced in Section 2. Different Compensationstrategies in DVR are studied in Section 3. Comparisonbetween the compensation methods with different aspectsare done in Section 4, followed by the limitation of thisDVR topology in Section 5. Finally, Section 6 concludes thediscussion.

2. DVR Topology

DVR controls the load voltage by injecting an appropriatevoltage during disturbances. It restores the load voltage toits rated value so that sensitive loads could operate in theirnormal conditions during disturbance. Injecting voltage mayneed active or reactive power and DVR should be able to pro-vide that. Generally there are two types of DVR: the one withno energy storage system, and the one with energy storagesystem. Each type is divided to two categories: a supply-side-connected converter, and a load-side-connected converter.All of these topologies are compared in [3]. Results showthat the no energy storage DVR topology with load-side-connected converter has better advantages over other topolo-gies. This kind of structure for DVR is shown in Figure 1.

By replacing the two converters of this DVR with a matrixconverter, the DC-link is removed. This topology is shown inFigure 2. The structure of a three phase to three-phase matrixconverter is shown in Figure 3.

With appropriate switching algorithm, matrix converteris able to provide the required output voltage from its input

A

B

C

3-Phaseinput

voltage

iA

iB

iC

ia ib ic

a b c

3-Phaseoutputvoltage

S11

S12

S13

S21

S22

S23

S31

S32

S33

Figure 3: Matrix converter structure.

terminals. There are various modulation techniques to con-trol matrix converter such as Alesina-Venturini modulation,Optimum Alesian-Venturini modulation, and Space Vectormodulations (SVM). The last method is evaluated as the bestone which is studied in [5].

3. Compensation Strategies of DVR

Load compensation could be done through active or reactivepower injection by DVR. There are three compensationtechniques based on using active, reactive, or both powerby DVR. These methods are recognized as prefault compen-sation method, in-phase compensation method, and energy-optimized compensation method. These strategies are intro-duced in this section followed by derivation of their respec-tive equations. Some simulations would show the character-istics of each method individually.

3.1. Prefault Compensation Method. This method is mainlyused for loads that are sensitive to phase jump such asthyristor controlled drives. In other words, both magnitudeand phase of the load voltage are compensated in this strategysuch that the load voltage during disturbances is as sameas the one before disturbances. Figure 4 shows this kind ofcompensation in per-unit plane.

In Figure 4, IPrefaultload = 1�0 is the load current before

disturbance that is considered as the reference vector. Theload is selected as a general type so the load voltage could bedefined as VPrefault

load = 1�ϕ. The disturbance might change thesource voltage magnitude from 1 to k with δ degree phasejump. Therefore the source voltage after disturbance can bedefined as VPostfault

s = k�(ϕ+δ). DVR should inject a voltagewith appropriate magnitude and phase to compensate thisdisturbance. The DVR injected voltage is defined as VDVR =x�β. Also the load voltage and current after compensationare the same as their values before disturbance which meansVPostfault

load = VPrefaultload and IPostfault

load = IPrefaultload . Moreover the

resultant phase jump between load voltage before and afterdisturbance, γ, is zero in this compensating method. By these

Advances in Power Electronics 3

Vs

VDVR

β

δ

ϕγ = 0

Vload = Vload

Iload = Iload

Postfault

Postfault

PostfaultPrefault

Prefault

Figure 4: Prefault compensation strategy.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Source voltage magnitude (k)

DV

R v

olta

ge m

agn

itu

de (x)

δ = 40◦

δ = 30◦

δ = 20◦

δ = 10◦

δ = 0◦

Figure 5: Prefault compensation strategy: the magnitude of theinjected voltage versus the magnitude of the source voltage.

definitions, the DVR-injected voltage can be calculated asfollows:

VDVR = VPostfaultload −VPostfault

s =⇒VDVR = 1�ϕ− k�

(ϕ + δ

) =⇒VDVR =

[cos(ϕ)− k cos

(ϕ + δ

)]

+ j[sin(ϕ)− k sin

(ϕ + δ

)] =⇒

(1)

x =√

1 + k2 − 2k cos(δ), (2)

β = tan−1

(sin(ϕ)− k sin

(ϕ + δ

)

cos(ϕ)− k cos

(ϕ + δ

)

)

. (3)

Equation (2) shows that the magnitude of the injectedvoltage only depends on the degree of distortion in the inputvoltage waveform. Figure 5 shows this parameter versus themagnitude of the input voltage for different values of δ.

Vs

VDVR

ϕ

ϕ

Vload

Vload

Iload

Iload

γ = δ

β = ϕ + δ

1 P.U.

Postfault

Postfault

Postfault

Prefault

Prefault

Figure 6: In-phase compensation strategy.

3.2. In-Phase Compensation Method. In this method, DVRonly compensates the load voltage magnitude so the phasejump would be remained uncompensated. Figure 6 showsthis control strategy. It is obvious that there is a phase differ-ence between the load voltages before and after compensa-tion. By using the same definitions in the previous section,the DVR-injected voltage is as follows:

VDVR = 1�(ϕ + δ

)− k�(ϕ + δ

) = (1− k)�(ϕ + δ

) =⇒(4)

x = 1− k, (5)

β = ϕ + δ. (6)

Therefore, the load voltage after compensation can bewritten as VPostfault

load = 1�(ϕ + δ), so that the load currentwould be IPostfault

load = 1�δ. Moreover, there is a phase jumpequal to γ = δ in the load voltage.

The magnitude of the injected voltage for any distur-bance’s phase jump and any load power factor is shown inFigure 7 that satisfies (5).

3.3. Energy-Optimized Compensation Method. It should benoticed that both previous strategies need a storage system tooperate. However in energy-optimized method, the capacityof this storage system is minimized. Furthermore in shallowsags, the compensation could be done only with reactivepower, and the amount of active power injection is zero. Thisstrategy is based on injecting voltage in such a way that theinjected voltage vector would be at almost 90◦ to the resultantload current vector. This will minimize the injection of activepower and even it may be zero. Figure 8 shows the basics ofthis strategy.


0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1


DV

R v

olta

ge m

agn

itu

de (x)

Figure 7: In-phase compensation strategy: the magnitude of theinjected voltage versus the magnitude of the source voltage.

Vs

VDVR

Vload

Vload

Iload

Iload

γ

γ

γ

β

ϕ

ϕ

δ

1 P.U.

Prefault

Postfault

Postfault

Postfault

Prefault

Figure 8: Energy-optimized compensation strategy.

With the similar definitions like previous sections forvoltage and currents, the injected voltage can be calculatedas follows:

VDVR = VPostfaultload −VPostfault

s =⇒VDVR = 1�

(ϕ + γ

)− k�(ϕ + δ

) =⇒VDVR =

[cos(ϕ + γ

)− k cos(ϕ + δ

)]

+ j[sin(ϕ + γ

)− k sin(ϕ + δ

)] =⇒

(7)

x =√

1 + k2 − 2k cos(δ − γ

), (8)

β = tan−1

(sin(ϕ + γ

)− k sin(ϕ + δ

)

cos(ϕ + γ

)− k cos(ϕ + δ

)

)

. (9)

0

10

20

30

40

50

60

70

80

90

100

cos(ϕ) = 0.7

Ph

ase

jum

p (γ◦ )

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9


δ = 40◦

δ = 30◦

δ = 20◦

δ = 10◦

δ = 0◦

Figure 9: Energy-optimized compensation strategy: the phasejump of the load voltage versus the magnitude of the source voltagefor PF = 0.7.

According to Figure 8 the injection voltage phase angle isβ = π/2 + γ. By replacing this value in (9), the load phasejump can be calculated after some manipulations:

tan−1

(sin(ϕ + γ

)− k sin(ϕ + δ

)

cos(ϕ + γ

)− k cos(ϕ + δ

)

)

= π

2+ γ =⇒

sin(ϕ + γ

)− k sin(ϕ + δ

)

cos(ϕ + γ

)− k cos(ϕ + δ

) = −cotγ =⇒(10)

γ = ϕ + δ − cos−1(

1k

cos(ϕ)). (11)

To ensure that (11) is feasible, (1/k) cos(ϕ) ≤ 1 has to beestablished. So the source amplitude must be more than theload power factor or k ≥ cos(ϕ). If this situation does notyield, γ should be calculated through (12):

γ = ϕ + δ. (12)

Figures 9 and 10 show the load phase jump and theinjected voltage magnitude for different values of δ. The loadpower factor is chosen as 0.7.

In Figures 11 and 12, δ = 30◦ is constant and the loadphase jump and the injected voltage magnitude is plotted fordifferent load power factor. There are some notices in thesefigures. It should be noted that the injected voltage magni-tude is independent of δ. Moreover the break-points in thesefigures are relevant to the load power factor. In other words,the load can be compensated only through the injection ofreactive power for k ≥ cos(ϕ); otherwise there would be anactive power injection too.


0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1


DV

R v

olta

ge m

agn

itu

de (x)

cos(ϕ) = 0.7

δ = 0◦ = 10◦ = 20◦ = 30◦ = 40◦

Figure 10: Energy-optimized compensation strategy: the magni-tude of the injected voltage versus the magnitude of the sourcevoltage for PF = 0.7.

cos(ϕ) = 0.7

cos(ϕ) = 0.6

cos(ϕ) = 0.8

cos(ϕ) = 0.9

cos(ϕ) = 1

0

10

20

30

40

50

60

70

80

90

100

Ph

ase

jum

p (γ◦ )

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9


Figure 11: Energy-optimized compensation strategy: the phasejump of the load voltage versus the magnitude of the source voltagefor δ = 30◦.

4. Comparison betweenCompensation Strategies

In this section, all compensation strategies are comparedwith each other. There are different criteria for comparison.Suppose that a disturbance with δ = 45◦ occurs and the loadpower factor is cos(ϕ) = 0.6.

4.1. Injection Voltage Magnitude. In Figure 13 the magnitudeof the injected voltage versus the magnitude of the sourcevoltage is shown for all compensation methods. It could beobserved that the in-phase method has the least injectionvoltage magnitude. Furthermore for shallow disturbances,

cos(ϕ) = 0.7

cos(ϕ) = 0.6

cos(ϕ) = 0.8

cos(ϕ) = 0.9cos(ϕ) = 1

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1


DV

R v

olta

ge m

agn

itu

de (x)

Figure 12: Energy-optimized compensation strategy: the magni-tude of the injected voltage versus the magnitude of the sourcevoltage for δ = 30◦.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1


DV

R v

olta

ge m

agn

itu

de (x)

Energy-optimized method

Prefault method

In-phase method

Figure 13: Comparison between different control strategies: themagnitude of the injected voltage versus the magnitude of thesource voltage.

energy optimized method has less injection voltage magni-tude than the prefault method. For large disturbances, theprefault compensation strategy has the most magnitude ofthe DVR injected voltage. This means that in this compen-sation method, DVR components should have a higher volt-age rating compared with the other methods.

4.2. Load Voltage Phase Jump. Figure 14 shows the phasejump of the load voltage for the source voltage magnitudebetween 0.1 to 0.9 per unit. This figure shows that the energyoptimized method has the most load phase jump anglefollowed by the in-phase method. This parameter can be acriterion of the load voltage distortion especially at the startand the end of the disturbance occurring time. The prefaultmethod compensates the load voltage to its prefault value soit makes no phase jump in the load voltage.



Prefault method

In-phase method

0

10

20

30

40

50

60

70

80

90

100

Ph

ase

jum

p (γ◦ )

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9


Figure 14: Comparison between different control strategies: thephase jump of the load voltage versus the magnitude of the sourcevoltage.

Source

Matrixconverter

Load

−

−

+

+

Is IsVDVR

n1

VDVR

n

nIs

Vload Iload

IMC

Figure 15: Schematics of a DVR based on matrix converter.

4.3. Source Current Magnitude. According to Figure 15, thematrix-converter-based DVR has no energy storage and therequired power for compensation is taken from the griditself. So the source current can be a comparative characteris-tic of different compensation techniques for this kind ofDVR.

As it has been mentioned, matrix converter is able tocontrol its input power factor. The matrix converter inputvoltage is the load voltage which is VPostfault

load = 1�(ϕ+γ) aftercompensation. By considering a unity power factor for theinput of matrix converter, the current of matrix converterwould be IPostfault

MC = y�(ϕ + γ). Also the source currentis defined as IPostfault

s = z�(ζ). Because there is no energystorage in the structure, the input and output active power ofthe matrix converter should be equal. Therefore

PPostfaultMC1

= PPostfaultMC2

=⇒

Re{[

VPostfaultload

][IPostfault

MC

]∗}

= Re{[

1nVDVR

][nIPostfault

s

]∗} =⇒

Re{[

1�(ϕ + γ

)][y�(ϕ + γ

)]∗}

= Re{[x�β

][z�ζ]∗

}=⇒

(13)

y = xz cos(ζ − β

). (14)

Now by using KCL we have

IPostfaultload = IPostfault

s − IPostfaultMC = z�ζ − y�

(ϕ + γ

) =⇒1�γ = [z cos(ζ)− y cos

(ϕ + γ

)]

+ j[z sin(ζ)− y sin

(ϕ + γ

)] =⇒(15)

√z2 + y2 − 2zy cos

(ζ − ϕ− γ

) = 1, (16)

tan−1

(z sin(ζ)− y sin

(ϕ + γ

)

z cos(ζ)− y cos(ϕ + γ

)

)

= γ. (17)

After some manipulation, (17) yields to

sin(ζ − γ

) = y sin(ϕ + γ

)− y tan(γ)

cos(ϕ + γ

)

z√

1 + tan2(γ) . (18)

By inserting (14) in to (18) we have

ζ = tan−1

⎛

⎝sin(γ)√

1 + tan2(γ)

+ x cos(β)A

cos(γ)√

1 + tan2(γ)

+ x sin(β)A

⎞

⎠, (19)

where A denotes [sin(ϕ + γ) − tan(γ)csos(ϕ + γ)] and byinserting (14) in to (16):

z = 1√

1 + x2 co s2(ζ − β

)− 2x cos(ζ − β

)cos(ζ − ϕ− γ

) .

(20)

Equations (19) and (20) can be applied to all of the com-pensation strategies. It should be noted that for the prefaultmethod γ = 0, for the in-phase method γ = δ, and for theenergy optimized method γ can be calculated by (11) or (12).

Figure 16 shows the magnitude of the input source cur-rent versus the magnitude of the source voltage for differentcontrol strategies. It shows that the energy optimized methodand the prefault method has the most and the least magni-tude of the input current, respectively. The higher value ofthe input current makes a more severe disturbance for theupstream loads because it makes more voltage drop. More-over, the DVR components should have a higher currentrating.

4.4. Matrix Converter Current Magnitude and Capacity.According to Figure 15, the postfault matrix converter cur-rent is IPostfault

MC = y�(ϕ + γ). The capacity of the matrixconverter can be calculated as follows:

|SMC| =∣∣∣VPostfault

load

∣∣∣∣∣∣IPostfault

MC

∣∣∣ = y. (21)

Figure 17 shows the magnitude of this current (or the matrixconverter capacity) versus the magnitude of the source volt-age. It can be seen that the energy optimized method needsleast matrix converter rating to perform.

4.5. Voltage Transfer Ratio of Injection Transformer. Accord-ing to Figure 15, the input and output voltage of matrix con-verter are related to each other:

∣∣∣∣VDVR

n

∣∣∣∣ = q

∣∣∣VPostfault

s

∣∣∣ =⇒ n = x

qk, (22)


0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90

1

2

3

4

5

6

7

8

9

10


Prefault methodIn-phase method


Sou

rce

curr

ent

mag

nit

ude

(z)

Figure 16: Comparison between different control strategies: themagnitude of the input source current versus the magnitude of thesource voltage.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

0

1

2

3

4

5

6

7

8

9



Mat

rix

conv

erte

r in

put

curr

ent

(y)

Mat

rix

conv

erte

r ca

paci

ty (S M

C)


Figure 17: Comparison between different control strategies: Matrixconverter input current (or capacity) versus the magnitude of thesource voltage.

where q is the matrix converter transfer ratio. The maximumvalue of q is 0.866 [5]. Therefore the minimum value ofthe transformer voltage ratio can be calculated for everycompensation algorithm. The result is plotted in Figure 18.For example the minimum value of n to compensate a dis-turbance with k = 0.55 are 1.52, 0.94, and 1.68 for prefault,in-phase, and energy-optimized method, respectively.

4.6. Injection Transformer Capacity. The injection trans-former capacity can be calculated as follows:

|ST | = |VDVR|∣∣∣IPostfault

s

∣∣∣ = xz (23)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90

2

4

6

8

10

12


Prefault method

In-phasemethod


Inje

ctio

n t

ran

sfor

mer

vol

tage

rat

io (n

)

Figure 18: Comparison between different control strategies: theinjection transformer voltage transfer ratio versus the magnitudeof the source voltage.

0

1

2

3

4

5

6

7

8

9




0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

Inje

ctio

n t

ran

sfor

mer

cap

acit

y (S

T)

Figure 19: Comparison between different control strategies: theinjection transformer capacity versus the magnitude of the sourcevoltage.

and Figure 19 shows this parameter characteristic for allcompensation strategies.

5. DVR Performance Limitation

In the matrix-converter-based DVR, the compensation limitdepends on the thevenin impedance of the network. Forinstance, in a deep Sag condition DVR consumes a largesource current to compensate this disturbance because in thistopology there is no energy storage device. This large currentproduces a large voltage drop on the network impedances.As it was mentioned before, the maximum voltage transferratio of a matrix converter is 0.866, which is less than unity.


Zs∠α

Vs∠A

z∠ζ −

−

−+

+

+

k∠(ϕ + δ)

x∠β

n1

1∠(ϕ + γ)

1∠γ

y∠(ϕ + γ)

LoadSource

Matrixconverter

Figure 20: Effect of network thevenin impedance on DVR compen-sation limits.

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

n = 1

n = 3

n = 2

Source impedance (P.U.)

Dis

turb

ance

mag

nit

ude

(P.

U.)

Figure 21: Maximum compensation capability versus the networkthevenin impedance for different voltage ratio of injection trans-former.

Therefore the maximum compensation capability of theDVR can be calculated. Equations (8) and (22) yield to:

k =cos(δ − γ

)−√

cos2(δ − γ

)+ n2q2 − 1

1− n2q2. (24)

According to Figure 20, Vs and δ can be calculated asfollows:

Vs∠ν− (Zs∠α)(z∠ζ) = k∠(ϕ + δ

) =⇒ (25)

δ = −ϕ + tan−1

(Vs sin(ν)− Zsz sin(α + ζ)Vs cos(ν)− Zsz cos(α + ζ)

)

, (26)

Vs =√K2 + Z2

s z2 + 2kZsz cos(ϕ + δ − α− ζ

). (27)

Consider that γ = 0 for the Prefault compensationmethod. Equations (9), (19), (20), (22), (24), (26), and (27)could be solved by numerical techniques. Figure 21 shows themaximum disturbance magnitude that can be compensatedversus the amount of the network thevenin impedance forthree transformer voltage ratio (cos(ϕ) = 0.86, α = 85◦, ν =30◦). It shows that for small-grid impedances a larger trans-former voltage ratio is needed to compensate voltage distur-bances. However, this situation is different when the networkimpedance is large such as weak distribution networks.

6. Conclusion

In this paper, different compensation strategies to providethe injection voltage for a matrix-converter-based DVR areanalyzed. Different comparison criteria are studied. Resultsshow that if the load is sensitive to voltage phase jump, pre-fault method should be used. However, this method drawsmore current from the source which needs DVR componentswith higher current rating. If the load is insensitive to phasejump, either in-phase or energy optimized methods can beused to compensate it. The in-phase method has the leastinjection voltage magnitude that makes the componentscheaper. Moreover for similar disturbance compensation,the in-phase strategy needs less transformer voltage ratiocompared to other methods. The energy-optimized methodneeds the least input current magnitude to compensatedisturbances. This will create less voltage drop for upstreamloads and also the possibility to design a DVR with less cur-rent rating components. At the end the DVR limits are calcu-lated for this topology. Results show that for a strong networka higher transfer ratio of the injection transformer brings theability to compensate deeper sags. However for a weak net-work, that has larger short circuit impedance, a lower transferratio makes DVR compensating more severe disturbances.

Acknowledgment

It is a pleasure to thank those who made this work possi-ble, especially Shiraz Electrical Distribution Company thatsupported this paper.

References

[1] D. D. Sabin, “An assessment of distribution system powerquality,” Tech. Rep. TR-106294-V2, Electric Power ResearchInstitute, Palo Alto, Calif, USA, 1996.

[2] IEEE STD 1159–1995, “IEEE Recommended Practice forMonitoring Electric Power Quality”.

[3] J. G. Nielsen and F. Blaabjerg, “A detailed comparison of sys-tem topologies for dynamic voltage restorers,” IEEE Transac-tions on Industry Applications, vol. 41, no. 5, pp. 1272–1280,2005.

[4] M. Venturini and A. Alesina, “Generalised transformer: a newbidirectional, sinusoidal waveform frequency converter withcontinuously adjustable input power factor,” in Proceedinds ofthe IEEE Power Electronics Specialists Conference (PESC ’80),pp. 242–252, Atlanta, Ga, USA, June 1980.

[5] D. Casadei, G. Serra, A. Tani, and L. Zarri, “Matrix convertermodulation strategies: a new general approach based on space-vector representation of the switch state,” IEEE Transactions onIndustrial Electronics, vol. 49, no. 2, pp. 370–381, 2002.

[6] B. Wang and G. Venkataramanan, “Dynamic voltage restorerutilizing a matrix converter and flywheel energy storage,” inProceedings of the IEEE Industry Applications Conference 42ndIAS Annual Meeting, pp. 208–215, September 2007.

[7] E. Babaei and M. F. Kangarlu, “A new topology for dynamicvoltage restorers without dc link,” in Proceedinds of the IEEESymposium on Industrial Electronics & Applications (ISIEA’09), pp. 1016–1021, October 2009.

[8] J. M. Lozano, J. M. Ramirez, and R. E. Correa, “A novel dynam-ic voltage restorer based on matrix converters,” in Proceedings


of the IEEE Proceedings Modern Electric Power Systems (MEPS’10), Wroclaw, Poland, 2010, paper 15.2.

[9] M. Jazayeri and H. Abdollahzadeh, “A novel DVR control sys-tem design for compensating all types of voltage sags based onpre-fault method,” European Journal of Scientific Research, vol.33, no. 1, pp. 70–85, 2009.

[10] S. Zhe, G. ChunLin, X. YongHai, X. XiangNing, L. Yingying,and T. Shun, “A new analysis method for compensation strat-egy of DVR and minimum energy control,” in Proceedings ofthe 5th IEEE Conference on Industrial Electronics and Appli-cations (ICIEA ’10), pp. 1321–1325, June 2010.

[11] J. G. Nielsen, F. Blaabjerg, and N. Mohan, “Control strategiesfor dynamic voltage restorer compensating voltage sags withphase jump,” in Proceedings of the16th Annual IEEE AppliedPower Electronics Conference and Exposition (APEC ’01), vol.2, pp. 1267–1273, March 2001.

[12] R. Strzelecki and G. Benysek, “Control strategies and com-parison of the dynamic voltage restorer,” in Proceedings ofthe 6th International Conference on Power Quality and SupplyReliability, pp. 79–82, August 2008.

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttp://www.hindawi.com Volume 2010

RoboticsJournal of

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation http://www.hindawi.com

Journal ofEngineeringVolume 2014

Submit your manuscripts athttp://www.hindawi.com

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation http://www.hindawi.com

Volume 2014

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

SensorsJournal of


Modelling & Simulation in EngineeringHindawi Publishing Corporation http://www.hindawi.com Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks


comparativestudiesofdifferentcontrolstrategiesof...

Documents