voltage regulation using pi control of statcom with ... · 2. statcom control model a. pi statcom...

http://www.iaeme.com/IJMET/index.asp 766 [email protected]

International Journal of Mechanical Engineering and Technology (IJMET) Volume 8, Issue 8, August 2017, pp. 766–776, Article ID: IJMET_08_08_084 Available online at http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=8&IType=8 ISSN Print: 0976-6340 and ISSN Online: 0976-6359 © IAEME Publication Scopus Indexed

VOLTAGE REGULATION USING PI CONTROL

OF STATCOM WITH FIREFLY ALGORITHM

S. Felix Stephen

Assistant Professor, Noorul Islam Centre for Higher Education, Kumaracoil, Kanyakumari District, Tamilnadu, India

I. Jacob Raglend

Professor, School of Electrical Engineering (SELECT), Vellore Institute of Technology, Vellore, Vellore District, Tamilnadu State, India

ABSTRACT

In power systems, voltage instability problems occur due to its continuous demand

in heavily loaded networks. FACTS devices are increasingly employed in stabilizing

the power systems. Among the FACTS devices, Static Synchronous Compensator

(STATCOM) injects the compensating current in phase quadrature with line voltage

and can imitate as inductive or capacitive reactance so as to produce capacitive

power for the AC grid or draws inductive power from the AC grid to control power

stream in the line. This paper proposes an algorithm named Firefly Algorithm, to get

the optimal gain of the PI control of STATCOM under disturbances and helps in

improving the performance and attaining a wanted response, irrespective of the

variation of working conditions. This work is implemented under

MATLAB/SIMULINK environment. And it shows better results over conventional PI

and Adaptive PI control.

Keywords: Voltage Stability, Reactive Power Compensation, Proportional Integral (PI) Control, STATCOM, Firefly Algorithm

Cite this Article: S. Felix Stephen and I. Jacob Raglend, Voltage Regulation Using PI Control of STATCOM with Firefly Algorithm, International Journal of Mechanical Engineering and Technology 8(8), 2017, pp. 766–776. http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=8&IType=8

1. INTRODUCTION

The stability of power system has become a significant problem for a secured system operation. Power system instability may occur due to large number of interconnections; more power transmissions through long transmission lines; new technologies; increased power consumption in heavy load areas; use of more number of induction machines and local uncoordinated system controls. The stability of power system is that for a given early working condition, it is the ability to redeem a state of operating steadiness when exposed to any

Voltage Regulation Using PI Control of STATCOM with Firefly Algorithm


physical disruption, with most of the system parameters restricted so that almost the whole system rests undamaged.

Voltage stability is a serious stability problem in enhancing the reliability and security of power schemes. Voltage stability is the ability in maintaining stable voltages on all its buses in the arrangement and also maintaining or restoring balance among demand and source of load from its specified early working circumstances under disturbances. Another problem, Voltage collapse which is highly complex than voltage instability is the course that the structure of voltage uncertainty leads to an unusual condition of small voltages blackout or blackout in important parts of a power system. Such voltage collapse has some symptoms like heavy reactive power flows; low voltage; heavily loaded systems and inadequate reactive support. Generally, sufficient reserves will be available those settle to a steady voltage level. Though, system instability may occur because of the combined effect of system conditions and events that the deficiency of added reactive power that leads to voltage downfall. Thus the system meets a partial or total collapse. In power systems, voltage steadiness is worried with load areas and load characteristics and basically it is load stability. Voltage stability is of four types as,

• Large disturbance voltage stability

• Small disturbance voltage stability

• Transient voltage stability

• Longer term voltage stability

A. Causes of Voltage Instability

• Increase in load demand

• Inability to meet reactive power demand

• Disorders such as system faults, circuit contingency or loss of generator, small perturbations

• Unfavourable fast acting load components (IM and DC converters)

• Higher loads in transmission lines

• Voltage sources are too distant from the load centres

• Very low generation

• Large distances between source and load

• ULTC action during low voltage conditions

• Poorly coordinated control and protective systems

• Insufficient load reactive compensation.

Figure 1 A Voltage Stability Phenomenon

S. Felix Stephen and I. Jacob Raglend


B. STATCOM

Power systems are complex, nonlinear and it is crucial to use the methods capable of handling any nonlinearity within the system. In power transmission and distribution system applications, Power electronic devices are becoming more popular. The increase in power system transmission capability and stability can be achieved by efficient and economic reactive power (VAR) compensation. The FACTS devices (Flexible AC Transmission Systems) have been introduced and more recent device STATCOM replaces the synchronous condenser by a converter i.e., a voltage source inverter VSI with a fixed dc link capacitor. The VAR control required is provided by a fixed set of capacitors that improves the quick control of power factor utility and bus voltage. The use of this device has more advantages like speed of response over conventional methods using thyristorised converters.

Figure 2 STATCOM Connected to Power System

In FACTS family, there is a shunt connected device named Static Synchronous Compensator (STATCOM) as shown in Figure 2. It is a 3Ø voltage system that lets both generation and intake of reactive power. This FACTS device comprises of a coupling transformer, measurement system, inverter/converter circuit, controller and a dc-link capacitor. Its steady-state capability is given by the V-I and V-Q characteristics as in Fig. 3 respectively. IQ, the reactive current can be fixed within its extreme inductive and capacitive bounds even during very low voltage circumstances.

Figure 3 STATCOM Characteristics

The reactive power yield is toughly dependent on the firing angle ‘a’ of thyristor. And ‘a’ is decided by the phase shift between STATCOM voltage E and bus voltage V. Based on this firing angle, the dc capacitor charging state changes and so the amplitude of bus voltage E of STATCOM varies. And the amplitude of reactive current injection into the system is determined by this variance in amplitude of network voltage and bus voltage of STATCOM in addition to leakage reactance XT of transformer.

� � �� (1)

Without a STATCOM, the voltage drops, when the load connected is highly inductive or if there is a surge in the active power which is drawn by the load. But in the presence of an



applied STATCOM, there is a flattened voltage profile, because of capacitive power provision when the voltage is inferior to the reference voltage and inductive power provision, when there is greater voltage than the reference voltage owing to a lower demand in load. Also at the same time, because of the device’s capacitive power support, higher transfer of power can be achieved to the load.

2. STATCOM CONTROL MODEL

A. PI STATCOM

Figure 4 shows the STATCOM’s equivalent circuit. In this system, Rs is the series resistance with the VSI. It is the totality of the conduction losses of inverter and the winding resistance losses of transformer. Ls is the transformer leakage inductance. Rc is the resistance in shunt with the capacitor. It signifies the totality of the power losses in the capacitor and the switching losses of the inverter. In Figure 2 Val, Vbl and Vcl are the 3Øbus voltages; Vas, Vbs and Vcs are the 3Øoutput voltages; and ias, ibs and ics are the 3Øoutput currents [8], [9].

Figure 4 STATCOM - Equivalent Circuit

The mathematical expressions of the STATCOM are given as [8], [9]:

� �� (2)

� �� (3)

� �� (4)

� ��

�� !" � �#�� $- �%�&��!'� (5)

Figure 5 Traditional STATCOM PI Control Block diagram



By using the abc/dq transformation, the equations can be rewritten as;

( )��*��

+ � ,---. � ' / 0 1

/ 2345�0 � ' / 1/ 4�65� 71

�8 2345 � 71�8 4�65 � �

'�8 9:::;) ��*��

+ � �/ )

��*�0 + (6)

Where,

�and�*- d and q currents corresponding to ��, ��>6?��;

K - factor relating the dc voltage to the highest value of phase to neutral voltage; �� - dc voltage; α – leading phase angle of the output voltage with respect to bus voltage; ω - angular rotational speed; ��>6?�*� - d and q axis voltage conforming to��,��>6?��. The active and reactive powers of the system can be determined by, @1 � 7� �� (7)

B1 = 7� ��* (8)

The old-style control approach can be determined based on the equation shown above and the STATCOM system diagram is shown in Figure 5 [3], [4]. As in Figure 5, the purpose of phase locked loop (PLL) is synchronizing on the positive order component of the 3Ø primary voltage. The PLL output is used to compute the voltage and current components in the direct axis and quadrature axis. And the measurement systems of STATCOM measure the d and q components. The measured bus line voltage Vm and the reference voltage are compared and the required value of reactive reference current is provided by the voltage regulator. Also the reactive current Iq of STATCOM and reference current Iqref, are compared and the current regulator provides the angle that the inverter voltage phase shifted with respective to the system voltage as its output. STATCOMs’ capability of maximum reactive power can be organised by the limiter which is the limit imposed on the control value.

B. Control Equations

Both the inner and outer loop controls are similar and the mathematical model is determined for PI controller gain adjustments in the outer loop. Similarly inner loop gains can also be adjusted. Vdl(t) and Vql(t) can be computed with the d-q transformation.

)��( )�*�( )0 + = �7 ,---. 1 − �� − ��0 √7� − √7��√� �√� �√� 9:

::; D��( )��( )��( )E (9)

�F( ) = G��( ) + �*��( ) (10)

�HIJ( ) = � − K� − �F( )LM�NO (11)

PQ_�( ) = 1S×∆�(�)(∆�(�)VFS×W X�)YZ� Y (12)

P�_�( ) = [� × PQ_�( ) (13)

PQ_\( ) = 1]×∆\^(�)(∆\^(�)VF]×W _�)YZ� Y (14)

P�_\( ) = [\ × PQ_\( ) (15)



C. Flowchart of PI STATCOM

Figure 6 Flowchart of PI STATCOM

The system bus voltage which is measured over time Vm(t) is sampled to a favourite sampling rate and is then related with Vss. There is no need to adjust any of the parameters, Kp_V(t), Ki_V(t), Ki_I(t) and Kp_I(t) if, Vm(t) = Vss. And it is considered as the smooth run of the power system. But the PI control will begin if, Vm(t) ≠ Vss. The measured bus voltage Vm(t) is compared with Vref(t). Then, as a result of firefly algorithm the optimal gain values of Kp_V and Ki_V are given to the voltage regulator and thereby an updated Iqref is obtained through the current limiter as shown in Fig. 4. Then, this Iqref and measured q-current Iq are compared. Then the control gains Ki_I(t) and Kp_I(t) can be given to the current regulator block. At last the phase angle α is obtained and given into a limiter for output, that chooses the required amount of reactive power from the STATCOM. Following, a small value of tolerance threshold such as 0.0001 p.u. is chosen. If |∆�� !|�s greater than the tolerancethreshold, the current regulator and voltage regulator blocks have to be repeated until|∆�� !| becomes less than the given tolerance threshold. Hence, the values for Kp_V(t), Ki_V(t), Ki_I(t) and Kp_I(t) are maintained as an optimal value using the firefly algorithm.

3. FIREFLY ALGORITHM

A. Fundamentals of Firefly Algorithm

Firefly Algorithm is a nature stimulated algorithms that is founded on the flashing light of fireflies. This algorithm consumes three perfect rules that are particularly centred on some major blinking features of real fireflies. Some features are:



• All the fireflies are naturally unisex, and normally they will move towards the more brighter and attractive ones irrespective of their gender.

• The firefly’s degree of attraction is relative to its intensity, brightness those decrease with the increase in distances from other fireflies.

• The firefly will move casually aimlessly in absence of more attractive or brighter fireflies.

The fitness function required for any optimization problem is related with the flashing light of fireflies effective best solutions. The important stages of Firefly algorithm are explained the flowchart. And while searching for an optimal solution the fireflies use two important procedures namely movement and attractiveness.

• Attractiveness

The following decreasing function gives the firefly’s attractiveness function, a�b! �β0exp(-γrm), with m ≥ 1 (16)

Where,

r – space between two fireflies

β0- initial attractiveness at r= 0

γ- absorption parameter that controls reduction in light intensity.

Consider i and j as any two fireflies and their position be xi and xj, correspondingly. Thisis defined as,

rij� cd Ke�,f– eh,fLfi�

� (17)

Where,

xi,k- kth component of the spatial coordinate xiof the ith firefly

d- dimension number

• Movement

Consider two fireflies i and j. The firefly i is attracted by the brighter firefly jand thereby i moves towards j and the equation showing this movement is given by,

xi= xi + β0 * exp(-γ rij2) * (xj-xi) + α * (rand–½) (18)

Here the first term of this equation shows the present location of a firefly, the second term shows the attractiveness of firefly to light intensity realized by nearby fireflies and the third term shows the firefly’s random movement if there are no brighter fireflies.

Also here,

α- randomization parameter that can be found by the problem of interest

rand - a random number generator uniformly dispersed in the space [0, 1].

B. Firefly Algorithm Flowchart

Firefly algorithm is one of the metaheuristics optimization approaches. First it creates initial population of possible candidate solutions randomly. In the search space the fireflies in the population are treated towards finding the best location by checking their fitness i.e. flashing light and randomness.

In the three dimensional search space each firefly moves with an attractiveness which is dynamically restructured by understanding the firefly and its neighbours. This type of search scheme is applied on several iterations and as a result of reducing the fitness function, FA



finds the optimized values for PI control parameters. The PI controllers’ optimal gain can be obtained by following the proposed algorithm and its flowchart is shown in Figure 7.

Figure 7 Firefly Algorithm Flowchart

The process of optimization initiates with the creation of population. Here consider three populations P1, P2 and P3. Next on comparing the populations P1, P2 and P1, P3 for fitness, the fitted ones from P1are moved towards P2 and a new population Pnew is created. Then the Pnew is compared with P1 to get the updated P1. Similarly a Pnew is created from P1 and P3. This new Pnew is compared with the updated P1 to create the Best P1. Similarly this process can be repeated for n number of populations. And the best P1 will give the optimal gain parameters for the voltage regulation in PI control of STATCOM.

Figure 8 PI Tuning Scheme with FA

4. RESULTS AND DISCUSSIONS

The simulations of PI control for STATCOM are done in MATLAB/SIMULINK and the test system is shown in Figure 9. In Matlab/Simulink library a standard STATCOM system sample is chosen. ASTATCOM of 100-MVARisappliedwitha48-pulseVSC and associated to a 500-kV bus. And the machines taken in the simulation work are all dynamical models [1], [3], [4].Also here, the control performance of STATCOM is clearly focused in the bus voltage regulation mode. In traditional method, the current and voltage regulator control gains largely affect the regulation speed and the reactive power compensation. This traditional control gains



are now optimised by the Firefly Algorithm and used for better regulation of voltage. The optimal gain values from Firefly algorithm are given in Figure 10. Table I shows the comparison of output voltage values with the reference voltages at various time. And it is observed that the error difference is very minimum.

Table 1 VoutvsVref

Time Vref Vout

0 1 0.9896

2.5x10-4 1 0.9907

2.75x10-4 0.9585 0.9510

3.75x10-4 0.9666 0.9582

5.25x10-4 0.9727 0.9626

6.25x10-4 0.9752 0.9640

0.0010 0.9817 0.9661

0.0014 0.9854 0.9666

0.2 1 0.9730

0.2 1 0.4736

0.2004 0.5 0.4766

0.5 0.4948 0.4919

0.5 0.5 0.1922

0.5002 0.2 0.1934

0.5007 0.2 0.1954

0.7004 0.2 0.2013

0.9999 0.2 0.2010

1 0.2 0.2010

Figure 11 & Figure12 show the output voltage characteristics with respect to steady state and reference voltages respectively.

Figure 9 PI STATCOM in MATLAB



Figure 10 Results of Firefly Algorithm

Figure 11 VoutvsVss Figure 12 VoutvsVref

5. CONCLUSION AND FUTURE WORK

The PI controller’s parameters tuning problem has been power fully done in this paper by having the new Metaheuristic based firefly algorithm. Previously the voltage regulation stability problems have been discussed in many literatures with different STATCOM control methods using PI controllers. However, the PI gains of the regulator are obtained as extensive studies of controller performance and applicability or trial and error approach. Hence, at any given operating point, the control parameters for the optimal performance may not be effective for all the times at a different working point. Also in some papers, Adaptive PI control is proposed for STATCOM for voltage regulation. But the optimal gain values for the voltage and current regulator blocks can be achieved only by this proposed Metaheuristic based firefly algorithm. In this simulation study, the suggested scheme of PI control is related with the Firefly Algorithm for getting the optimal gains. And it is proved in Figure 11 and Figure 12 that the proposed algorithm gives outstanding performance even under different system conditions. The result shows that the proposed algorithm for the PI control performs more efficient and also improves the system response speed consistently. In future this work can be extended in systems with multiple STATCOMs, and also various optimization intelligent techniques can be implemented to improve the performance further.

REFERENCES

[1] Matlab & Simulink, GTO-Based STATCOM December 2013. [Online]. Available: http://www.mathworks.com/help/physmod/sps/powersys/ug/gto-based-statcom.html, Febuary 2012.

[2] M. Merkle, and A. M. Miri, Modelling of Industrial Loads for Voltage Stability Studies in Power Systems, Proceedings of IEEE Canadian Conference of Electrical and Computer Engineering, 2001.



[3] A. H. Norouzi and A. M. Sharaf, Two Control Schemes to Enhance the Dynamic Performance of the STATCOM and SSSC, IEEE Transactions on Power Delivery, Vol. 20, No. 1, pp. 435–442, January 2005.

[4] M. S. E. Moursi and A. M. Sharaf, Novel Controllers for the 48-Pulse VSC STATCOM and SSSC for Voltage regulation and Reactive Power Compensation, IEEE Transactions

on Power Delivery, Vol. 20, No. 4, pp. 1985–1997, November 2005.

[5] P.W.D.S. Lehn, Modeling Analysis and Control of Current Source Inverter Based STATCOM, IEEE Transactions on Power Delivery, No.17, 2002, pp. 248-253.

[6] M. A. Golkar and M., Zarringhalami, Coordinated Design of PSS and STATCOM Parameters for Power System Stability Improvement Using Genetic Algorithm, Iranian Journal of Electrical and Computer Engineering, Vol. 8, pp. 80-88, 2009.

[7] P.W.D.S. Lehn, Modeling Analysis and Control of Current Source Inverter Based STATCOM, IEEE Transactions on Power Delivery, No.17, 2002, pp. 248-253.

[8] P. Rao, M. L. Crow, and Z. Yang, STATCOM Control for Power System Voltage Control Applications, IEEE Transactions on Power Delivery, Vol. 15, No. 4, pp. 1311–1317, October 2000.

[9] C. Schauder and H. Mehta, Vector Analysis and Control of the Advanced Static VAR Compensators, Proceedings on Institution of Electrical Engineering, Generation,

Transmission, Distribution, Vol. 140, No. 4, pp. 299–306, July 1993.

[10] S. K. Sethy and J. K. Moharana, Modeling, Design and Simulation of Current and Voltage Linear Controller of a STATCOM for Reactive Power Compensation, NSPEES-12, September 29-30, GIET, BBSR, pp. 37-44, 2012.

[11] Panda, S., Padhy, N.P., Optimal Location and Controller design of STATCOM for Power System Stability Improvement Using PSO,” J Franklin Institute, 2008, Vol. 345, No. 2, pp. 166-181.

[12] X. S. Yang, Nature-Inspired Metaheuristic Algorithm, Luniver Press (2008).

[13] Bendjeghaba O, Isha kBoushaki S, Zemmour N, Firefly Algorithm for Optimal Tuning of PID Controller Parameters, International Conference on Power Engineering, Energy and Electrical Drives Istanbul, Turkey, 13-17 May 2013.

[14] : Anoop Kiran. Evolution of Voltage Regulator to System on Chip Applications. International Journal of Advanced Research in Engineering and Technology, 6(8), 2015, pp. 42-56.

[15] M. Dilip Kumar, Dr. S. F. Kodad and Dr. B. Sarvesh, Fault Analysis For Voltage Source Inverter Driven Induction Motor Drive. International Journal of Electrical Engineering & Technology, 8(1), 2017, pp. 01–08.

[16] E.G.Chinju, S.Reshmila, Fault Protection of A Loop Type Low Voltage Dc Bus Based Microgrids, International Journal Of Electrical Engineering & Technology (IJEET), Volume 5, Issue 12, December (2014), pp. 240-247

voltage regulation using pi control of statcom with ... · 2. statcom control model a. pi statcom...

Documents