Fuzzy Logic Based SVC for Reactive Power ... - IEEE Xplore

Fuzzy Logic Based SVC for Reactive Power Compensation and Power Factor Correction S. Khanmohammadi, M. Tarafdar Hagh, M. Abapour

Abstract-- This paper presents an application of fuzzy control to determine the control signal of static var compensator (SVC) for reactive power compensation and power factor correction. Input signals for the FLC are chosen as load reactive power and initial firing angle of thyristors. The control signal is calculated using fuzzy membership functions. The effectiveness and feasibility of the TSK (Takagi-Sugeno-Kang) and mamdani type fuzzy controllers for thyristors firing control in SVC is compared. Effectiveness of the proposed technique is demonstrated by simulation studies on a single machine infinite bus system. Results obtained show improvement in the overall system characteristics using the proposed adaptive fuzzy logic SVC controller. Index Terms—SVC, Fuzzy controller, Power Factor, Reactive power.

I. INTRODUCTION

V

ar compensation is defined as the management of reactive power to improve the performance of ac power systems. The concept of Var compensation embraces a wide field of both system and customer problems, especially related with power quality issues, since most power quality problems can be solved with an proper control of reactive power [1-2]. In general, the problem of reactive power compensation is viewed from two parts: load compensation and voltage support. In load compensation the purposes are to increase the value of the system power factor, to balance the real power drawn from the ac supply, to compensate voltage regulation, and to eliminate current harmonic components produced by large and fluctuating nonlinear industrial loads [3-4]. The Static Var Compensator (SVC) has been designed to compensate reactive power, increase voltage stability

Sohrab Khanmohammadi is with University of Tabriz, Faculty of Electrical and Computer Engineering (e-mail: [email protected]) Mehrdad Tarafdar Hagh is with University of Tabriz, Faculty of Electrical and Computer Engineering, (e-mail: [email protected]) Mehdi Abapour is with University of Tabriz, Faculty of Electrical and Computer Engineering, (e-mail: [email protected])

and to reduce voltage oscillation [l]. The SVC consists of shunt reactors which current can be continuously controlled by thyristor valves thus the inductive power of reactors can be controlled. These reactors are called Thyristor Controlled reactors (TCR).ln the parallel of the TCR’s there are number of the filter capacitor banks .The Filter Circuits (FC) are providing needed amount of capacitive power and absorbing harmonic current generated by load and TCR. The difference between inductive power of TCR and capacitive power of filter circuits is the output power of SVC [5]. However, in recent years, static Var compensators (SVCs) employing thyristor-switched capacitors (TSCs) and thyristorcontrolled reactors (TCRs) to provide or absorb the required reactive power have been developed [1]. The application of control algorithms based on fuzzy sets theory, proposed by Zadeh [6], has grown in recent years [7-8]. This control method can be regarded as an adaptive control based on a linguistic process which is in turn based on the prior experience and heuristic rules used by human operators. The implementation of such control consists of translating the input variables to a language, like: positive big, zero, negative medium, etc. and to establish control rules so that the decision process can produce the appropriate outputs. If necessary, these linguistic outputs are transformed to numeric values [6]. Fuzzy logic control is one of the best and most successful techniques among expert control strategies, and is well known as an important tool to control non-linear, complex, vague, and ill-defined systems. The use of fuzzy set theory in providing effectiveness control based on the knowledge and technical experience of operators and the establishment of intelligent control have found favour in industry. The SVC system model based on the intelligent controller was given in the paper. This paper proposed a two input one-output fuzzy logic controller (FLC) for SVC. Also the effectiveness and feasibility of the TSK (Takagi-Sugeno-Kang) and mamdani methods for TCR thyristors firing control is compared.

II. SHUNT COMPENSATION Fig. 1 shows the principles and theoretical effects of shunt reactive power compensation in a basic ac system, which comprises a source, a power line, and a typical

1241 c 2007 RPS 978-981-05-9423-7

inductive load. Fig. 1(a) shows the system without compensation and its associated phasor diagram. In the phasor diagram, the phase angle of the current has been related to the load side, which means that the active current is in phase with the load voltage. Since the load is assumed inductive, it requires reactive power for proper operation and hence, the source must supply it, increasing the current from the generator and through power lines. If reactive power is supplied near the load, the line current can be reduced or minimized, reducing power losses and improving voltage regulation at the load terminals [1]. This can be done in three ways: 1) with a capacitor; 2) with a voltage source; or 3) with a current source. In Fig. 1(b), a capacitor device is being used to compensate the reactive component of the load current. As a result, the system voltage regulation is improved and the reactive current component from the source is reduced or almost eliminated. V2

V1

capacitors) is that the reactive power generated is independent of the voltage at the point of connection.

III. TCR (THYRISTOR CONTROLLED REACTOR) BRANCH The TCR branch makes use of the TSC overcompensation to fine-tune the SVC VARs supplied to the load. This fine- tuning is completed by varying the firing delay angle of the thyristors. The firing delay angle is the time delay from the start of each half-cycle that the TCR is turned on. Firing angles of TCR will be in the range of (S DSfor each half-cycle. Eq. (1) derived from the conduction angle equation in [4], illustrates the relationship between the firing angle and the TCR reactance (XTCR). (1) X TCR [2(S D ) sin( 2(S D )) /(S u X L )]1 Fig. 2 shows the relationship between the firing angle and the TCR reactance. 500

Line impedance AC

I M

source

IP

V2 I

IQ

400 TCR impedance (ohm)

V1

M

450

Load 3

X L*I RL*I

(a)

350 300 250 200 150 100 50

V2

V1

0 90

Line impedance AC

source

IM

IP c omponseto r

IQ

Load 3

100

110

120 130 140 Thyristor firing angle (degree)

XL *IP

V2 R L*IP I

(b)

Fig. 1. Principles of shunt compensation in a radial ac system,(a) without reactive compensation. (b) With shunt compensation

If the load needs leading compensation, then an inductor would be required. Also, a current source or a voltage source can be used for inductive shunt compensation. The main advantage of using voltage- or current-source Var generators (instead of inductors or

1242

170

Fig. 2 TCR impedance vs thyristor firing angle

V1

IQ

160

Fig. 3 shows the relationship between the firing angle and the SVC impedance. This figure illustrate that the SVC impedance is so sensitive in small firing angles.

IP

M

150

The 8th International Power Engineering Conference (IPEC 2007)

1

1500

0.9 0.8

Power factor

SVC impedance (ohm)

0.7 1000

500

0.6 0.5 0.4 0.3 0.2 0.1

0 90

100

110

120 130 140 150 Thyristor firing angle (degree)

160

170

0

180

0

0.2

0.4

0.6

0.8 Time (s)

1

1.2

1.4

Fig. 5. Power factor (PF) of the load without the branches compensating

Fig. 3 SVC impedance vs thyristor firing angle

IV. SIMULATION RESULTS

180

140 120 100 80 60 40 20 0 -20

Line impedance

Active power Reactive powert

160

Power (W)

Several simulink simulations were created to verify the feasibility of the SVC design and the TCR fuzzy controller. For the particular simulink model is shown in Fig. 4. Three reactive loads are switched in different times. Load 2 is switched on at t1=0.4s and switched off at t3=1.2s and load 3 is switched on at t2=0.8s. The TSC branch and a TCR branch were modeled with their respective controllers. The TSC branch consists of a single switched capacitor while the TCR branch consists of a reactor that is fed a firing angle determined by the fuzzy controller. The fuzzy controller accepts the phase angle difference of the load and firing angle of thyristors as an input and outputs the optimum firing angle of TCR.

0

0.2

0.4

0.6

0.8 Time (s)

1

1.2

1.4

Fig. 6. Active and reactive power at load side without compensation BR1 AC

BR2

source

Load 1

TCR

Load 2

Load 3

A. Mamdani type fuzzy controller In this section a Mamdani type double input single output (DISO) Fuzzy Linguistic Controller has been designed which has the following memberships and rules. NVB NB 1

TSC

NM

NS

NVS

ZERO

VS

S

M

B

-150

-100

-50

0 input1

50

100

150

200

VB

Fig. 4. Analyzed power circuit topology Degree of membership

0.8

Fig. 5 shows the displacement power factor (PF) of the load without the branches compensating. Fig. 6 shows the active and reactive power at load side without compensation. As shown in this figure at t1=0.4s and t2=0.8s.reactive power increases and power factor (PF) decreases by inductive load increasing. Also at t3=1.2s one of inductive loads retreat from power system and therefore power factor increases and reactive load decreases.

0.6

0.4

0.2

0 -200

Fig. 7 The fuzzy membership function of reactive power at load side (input 1 of fuzzy controller)


1243

TABLE. I FUZZY CONTROLLER RULES (MAMDANI TYPE FUZZY CONTROLLER) Reactive power D

VS

S

VS S M B VB VVB

M

B

NVB

NB

NM

NS

NVS

Z

VS

S

M

B

VB

NVVB NVVB NVVB NVVB NVVB NVVB

NVB NVVB NVVB NVVB NVVB NVVB

NM NB NVVB NVVB NVVB NVVB

NS NS NM NVVB NVVB NVVB

NVS NS NM NB NVB NVVB

Z Z Z Z Z Z

VS S M B VB VVB

S S M VB VVB VVB

M B VB VVB VVB VVB

VB VVB VVB VVB VVB VVB

VVB VVB VVB VVB VVB VVB

0.6

0.8 Time (s)

VB

VVB

1

180

140 0.6

120 100

0.4

power (W)

Degree of membership

Active Power Reactive Power

160

0.8

0.2

80 60 40

0 4

5

6

7 input2

8

9

-3

x 10

0

Fig. 8 The fuzzy membership function of initial thyristor firing angle (input 2 of fuzzy controller) NVB

NVVB 1

20

10

NB NM NS NVS ZEROVS S

M

B

VB

-20

VVB

0

0.2

0.4

1

1.2

1.4

Fig. 11. Active and reactive power at load side with compensation (Mamdani type fuzzy controller)

Degree of membership

0.8

As shown in figures 10 and 11 with using fuzzy controller we have suitable power factor and reactive power compensation.

0.6

-3

0.4

1

x 10

0.2

-5

-4

-3

-2

-1

0 output1

1

2

3

4

5 -3

x 10

Fig. 9 The fuzzy membership function of variation of thyristor firing angle (output of fuzzy controller) 1

Variation of TCR firing angle

0.8 0

0.6

0.4

0.2

0.9 0.8

Power factor

0.7

0

0.6

0

0.2

0.4

0.6

0.8 Time (s)

1

1.2

1.4

Fig. 12. Variation of TCR firing angle (Mamdani type fuzzy controller)

0.5 0.4

B. Sugeno type fuzzy controller

0.3 0.2 0.1 0

0

0.2

0.4

0.6

0.8 Time (s)

1

1.2

1.4

Fig. 10. Power factor (PF) of the load with the branches compensating (Mamdani type fuzzy controller)

1244

In this section a sugeno type double input single output (DISO) Fuzzy Linguistic Controller has been designed which has same memberships at input. This controller has following rules


TABLE. II. FUZZY CONTROLLER RULES (SUGENO TYPE FUZZY CONTROLLER) Reactive power D

NVB

NB

NM

NS

NVS

Z

VS

S

M

B

VB

NH NH NH NH NH NH

NVB NH NH NH NH NH

NB NVB NVVVB NH NH NH

NS NS NM NVVB NH NH

NVVVS NVVS NVVS NVS NVB NH

Z Z Z Z Z Z

VVVS VVS VVS VS VB H

S S M VVB H H

B VB VVVB H H H

VB H H H H H

H H H H H H

VS S M B VB VVB

4

As shown in fig. 14 and 15 using sugeno type controller results power factor correction and reactive power compensation similar to section (a). But as shown in fig. 17 using mamdani type fuzzy controller has results better characteristics in reactive power and power factor.

3 2

TCR current (A)

1 0

-4

7

x 10

-1

6

-3 -4

0

0.5

1

1.5

Time (s)

Fig. 13. TCR current at different time 1 0.9

Variation of TCR firing angle

-2

5 4 3 2 1

0.8

0 -1

0.6

0

0.2

0.4

0.6

0.5

0.3

1

0.2

0.9

0.1

0.8

0

0.8 Time (s)

1

1.2

1.4

Fig. 16. Variation of TCR firing angle (Sugeno type fuzzy controller)

0.4

0

0.2

0.4

0.6

0.8 Time (s)

1

1.2

0.7

1.4

Fig. 14. Power factor (PF) of the load with the branches compensating (Sugeno type fuzzy controller)

Power factor

Power Factor

0.7

0.6 0.5 0.4

180

0.3

Active power Reactive power

160

0.2 140

0.1

With Mamdani type fuzzy controller With Sugeno type fuzzy controller

120

Power (W)

0 100 80

0

0.2

0.4

0.6

0.8 Time (s)

1

1.2

1.4

Fig. 17. Comparison of “Mamdani” and “Sugeno” type fuzzy controller for power factor correction

60 40

V. CONCLUSION

20 0 -20

0

0.2

0.4

0.6

0.8 Time (s)

1

1.2

1.4

Fig. 15. Active and reactive power at load side with compensation (Sugeno type fuzzy controller)

The fuzzy logic control strategy of SVC is researched in the paper. The effectiveness and feasibility of the TSK (Takagi-Sugeno-Kang) and Mamdani type fuzzy controllers for thyristors firing control in SVC is shown and compared clearly. Simulation results show good performance of fuzzy controller in power factor correction and reactive power


1245

compensation. Also in this paper demonstrated that using Mamdani type fuzzy controller has results better characteristics in reactive power and power factor.

VI. REFERENC [1] Dixon, J., Moran, L., Rodriguez, E., Domke, R.; "Reactive Power Compensation Technologies: State-of-the-Art Review", Proceedings of the IEEE, Volume 93, Issue 12, Dec. 2005 pp: 2144 – 2164. [2] T. J. Miller, Reactive Power Control in Electric Systems. New York: Wiley, 1982. [3] E. Wanner, R. Mathys, and M. Hausler, “Compensation systems for industry,” Brown Boveri Rev., vol. 70, pp. 330–340, Sep./Oct. 1983. [4] G. Bonnard, “The problems posed by electrical power supply to industrial installations,” Proc. IEE Part B, vol. 132, pp. 335–340, Nov. 1985. [5] Hedayati, M.; "Technical Specification and Requirements of Static VAR Compensation (SVC) Protection Consist of TCR , TSC and Combined TCR/TSC" Universities Power Engineering Conference. Vol 1, 6-8 Sept. 2004 pp : 261 - 264. [6] Ronan Marcel0 Martins “Fuzzy Logic Based Control for Electric Power System” [7] Suiton, L. C. R. & Towill, D. R., “An Introduction to the use of Fuzzy Sets in the Implementation of Control Algorithms”, JIERE, vol. 55, N. 10, pp. 357-367, Oct. 1965. [8] Graham, B. P., Newell, R. B., “Fuzzy Adaptive Control of a FirstOrder Process”, Fuzzy sets and Systems, N. 31, pp. 47-65, North Holland, 1989.

Sohrab Khanmohammadi was born in Khoy, Iran, on February 15, 1953. He graduated from the Sharif University of Technology in Iran, He has got his D.E.A on automatic from university of Paul Sabatier, D.S.A on advanced Automatic and System and Engineering Doctoral From Ecol National D’Aeronautique et de l’Espace , Toulouse France. He is currently the Professor of Control and System at the Control department, Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran. His interested research fields are Fuzzy systems, Decision making, simulations and artificial intelligence. Mehrdad Tarafdar Hagh received his M. Sc. with first honor and PhD. both in power engineering from University of Tabriz, Iran in 1992 and 2000, respectively. He joined faculty of electrical and computer engineering of University of Tabriz in 2000. He has published more than 70 papers in power system and power electronics related topics. His interest topics include power system operation, FACTS and power quality.

Mehdi Abapour received his B.S. degree in power engineering from University of Tabriz in 2005. He is M.Sc student in power electronic engineering. His interest topics include fault current limiters, power quality and power system operation.

VII. BIOGRAPHIES

1246
