= best position of individual i until iteration. = best positionof the group until iteration k. The following weighting function is usually utilized.

Transcription

1 e t International Journal on Emerging Technologies 4(): 3-38(3) ISSN No. (Print): ISSN No. (Online): Particle Swarm Optimization based Load quency Control in Two Aa Power System Neha Modi, Amol Barve and Kanchan Chaturvedi Department of Electrical and Electronics Engineering, Laxmi Narain College of Technology Bhopal, (MP) (Received 5 March, 3, Accepted 9 May, 3) ABSTRACT: In this paper, determining the optimal proportional- integral-derivative (PID) controller gains for Two-aa load fquency control (LC) system using particle swarm optimization (PSO) is psented. The LC is notoriously difficult to control optimally using conventionally tuning a PID controller because the system parameters a constantly changing. It is for this ason the PSO as tuning strategy was applied. The PID control parameters a tuned based on PSO algorithm. Hence the sults establishes that tuning the PID controller using the PSO technique gives less over shoot, system is less sluggish and duces the integral time absolute error (ITAE). The simulated sult a obtained for diffent load configurations of the PSO based controller. The simulation has been conducted in MATLAB Simulink package for two aa power system Keywords: Load quency control, Two aa power system, Particle swarm optimization, PID Controller. I. INTRODUCTION In electric power generation, system disturbances caused by load fluctuations sult in changes to the desid fquency value. Load quency Control (LC) is a very important issue in power system operation and control for supplying sufficient and both good quality and liable power. Power networks consist of a number of utilities interconnected together and power is exchanged between the utilities over the tie-lines by which they a connected. The net power flow on tie-lines is scheduled on a priori contract basis. It is thefo important to have some dege of control over the net power flow on the tie-lines. Load quency Control (LC) allows individual utilities to interchange power to aid in overall security while allowing the power to be generated most economically. Generally, ordinary LC systems a designed with Proportional-Integral (PI) controllers. However, since the I control parameters a usually tuned, it is incapable of obtaining good dynamic performance for various load and system change scenarios. Many studies have been carried out in the past about the load fquency control. In literatu, some control strategies have been suggested based on the conventional linear control theory []. These controllers may be unsuitable in some operating conditions due to the complexity of the power systems such as nonlinear load characteristics and variable operating points. According to some authors, variable structu control [] maintains stability of system fquency. According to [3], conventional PID control schemes will not ach a high of control performances. Since the dynamic behavior even for a duced mathematical model of a power system is usually nonlinear, time-variant and governed by strong cross-couplings of the input variables, special ca has to be taken for the design of the controllers. or this ason, cently, a lot of artificial intelligence based robust controllers such as genetic algorithm, tabu search algorithm, fuzzy logic and neural networks a used for PID controller parameters tuning in LC by authors [4, 5, 6, 7]. Since, Particle Swarm Optimization algorithm is an optimization method that finds the best parameters for controller in the uncertainty aa of controller parameters and obtained controller is an optimal controller, it has been used in almost all sectors of industry and science. One of those aas is the load fquency control as shown in [8]. PSO is inspid by the ability of flock of birds or herd of animals to adapt to their environment. It was developed in 995 by James Kennedy and Russ Eberhart while attempting to simulate the choographed, graceful motion of the swarm of birds as a part of socio-cognitive study investigating the motion of collective intelligence in biological population. In PSO, a set of randomly generated solutions propagates in the designed space towards the optimal solution over a number of iteration based on large amount of information about the designed space [9]. The main objective of this study is to investigate the load fquency control and inter aa tie-power control problem for two-aa power system taking into consideration the uncertainties in the parameters of system. An optimal control scheme based particle swarm optimization (PSO) Algorithm method is used for determine the parameters of a PID controller.

2 The proposed controller is simulated for a two-aa power system. To show effectiveness of proposed method several changes in demand of first aa, demand of second aa and demand of two aas simultaneously a applied. Simulation sults indicate that PSO controllers guarantee the good performance under various load conditions. II. OVERVIEW O PARTICLE SWARM OPTIMIZATION The PSO method is a member of wide category of Swarm Intelligence methods for solving the optimization problems. It is a population based search algorithm whe each individual is ferd to as particle and psents a candidate solution. Each particle in PSO flies through the search space with an adaptable velocity that is dynamically modified according to its own flying experience and also the flying experience of the other particles. In PSO each particles strive to improve themselves by imitating traits from their successful peers. urther, each particle has a memory and hence it is capable of membering the best position in the search space ever visited by it. The position corsponding to the best fitness is known as pbest and the overall best out of all the particles in the population is called gbest [-]. PSO is basically developed through simulation of bird flocking in two-dimension space. The position of each agent is psented by XY axis position and also the velocity is expssed by Vx (the velocity of X axis) and Vy (the velocity of Y axis). Modification of the agent position is alized by the position and velocity information. Bird flocking optimizes a certain objective function. Each agent knows its best value so far (pbest) and its XY position. This information is analogy of personal experiences of each agent. Moover, each agent knows the best value so far in the group (gbest) among pbest. This information is analogy of knowledge of how the other agents around them have performed. Namely, each agent tries to modify its position using the following information: # The curnt positions (x, y), # The curnt velocities (vx, vy), Modi, Barve and Chaturvedi 33 # The distance between the curnt position and pbest # The distance between the curnt position and gbest This modification can be psented by the concept of velocity. Velocity of each agent can be modified by the following () Whe, = velocity of individual i at iteration k. = inertia weight parameter c,c = acceleration coefficients. r,r = random no between and = position of individual i at iteration k, k. = best position of individual i until iteration = best positionof the group until iteration k. The following weighting function is usually utilized. Whe = initial and final weight (3) = maximum iteration number, iter = curnt iteration number Each individual moves from the curnt position to the next one by the modified velocity in () using the following equation. = + (4) Whe = Curnt searching point = modified searching point = curnt velocity ig shows the concept of searching mechanism of PSO using the modified velocity and position of individual i based on () and (4) if the values of,, a. ig..

3 III. MODEL O TWO AREA POWER SYSTEM The two aa interconnected power system is shown in fig., whe and a the fquency deviations in aa and aa spectively in Hz. and a the load demand incments. A two aa system consists of two single aa systems, connected through a power line called tie-line, is shown in the igu: Modi, Barve and Chaturvedi 34 each aa feeds its user pool, and the tie line allows electric power to flow between the aas. Information about the local aa is found in the tie line power fluctuations. Thefo, the tie-line power is sensed, and the sulting tie-line power signal is fed back into both aas. It is conveniently assumed that each control aa can be psented by and equivalent turbine, generator and governor system. Symbol used with suffix fer to aa and those with suffix fer to aa. - ig.. Block diagram of Two aa power system. VI. MODEL WITH PROPOSED PSO-PID CONTROLLER The framework of PSO based self-tuning PID controller is depicted as igu 3. To find the optimum parameters (Kp, Ki, Kd) of PID controller, PSO program should search in 3- dimensional search space. In an ordinary load fquency control systems, Since a gulation constant R is used as Kp parameter in PID controller, especially I (integral) controller is used in LC systems. At the proposed system, R is also determine by PSO. So the gulation best value is used in LC system. Thus,for robustness, gulation constant is tuned according to load and system changes too. With the optimized parameter based on PSO algorithm, the proposed PID controller of the LC can achieve optimal properties. The block diagram of a two aa power system with this controller is shown in ig 3. V. SIMULATION RESULTS The Two aa power system parameter a shown in table and the value of gulation constant is also optimized by PSO. Simulation sult for Two aa power system a shown in table. During the simulation study, error signals, and tie line power which is quid for the controller is transferd to PSO softwa. All positions of particles on each dimension a clamped in limits which a specified by the user, and the velocities a clamped to the range [vmin., vmax.] given as []: a step load incase in demand of. p.u is applied to aa.

4 Modi, Barve and Chaturvedi 35 Step Sum Kh Th.s+ Governor Kt Tt.s+ Turbine Sum Ky Ty.s+ Inertia & Hold Out Out -K- -K- Scope Gain Scope3 PID PID Controller s Load Add Step Gain 3 Out3 Sum3 Kh Th.s+ Governor Kt Tt.s+ Turbine Sum Ky Ty.s+ Inertia & Hold Scope Gain PID PID Controller ig. 3. Block diagram of two aa power system with proposed PSO controller. The fquency deviation of the first aa f and the fquency deviation of the second aa f and inter aa tie-power signals of the closed-loop system a shown in ig. 4, 5 and 6. Similarly a step load incase in demand of. is applied to aa. quency deviation of the first aa f and the fquency deviation of the second aa f and inter aa tie-power signals of the closed-loop system a shown in ig. 7, 8 and 9. At the simulation, the population size is taken 5.As can be observed, the settling time, overshoots, undershoot, settling value and rise time with the proposed PSO-PID controller a much better than other controller. Thefo, the proposed PSO-PID controller provides better performance for the two aa power system In this simulation, the objective is to minimize the error and the maximum overshoot. or this ason the objective function is chosen as the Integral Time Absolute Error (ITAE). Table. Description Aa Aa Governor Gain GovernorTime Constant 8e-3 8e-3 Turbine Gain Turbine Tine Constant.3.3 Load Model Gain Load Time Constant Ptiemax Mw Mw Load change for quency change of..

5 Modi, Barve and Chaturvedi 36 Table. System Performance for proposed PSO-PID controller for % load at aa. Parameter Change in Change in Change in fquency in fquency in Tie line Aa Aa power Overshoot e-5-6.8e e-6 Undershoot e Settling Value e e e-6 Settling Time Rise Time e-6 7.8e-6 Table 3. System Performance for proposed PSO-PID controller for % load at aa. Parameters Change in fquency in Aa Change in fquency in Aa Change in Tie line power Overshoot 4.894e Under Shoot.79e e-5 Settling Value 4.893e Settling Time Rise Time e e-8.5 x -3 Plot for Load -.5 cy en qu ig. 4. Change in fquency in aa for a step load of % in aa with proposed PSO-PID controller. 7 x -5 Plot for Power System y nc ue q ig. 5. Change in fquency in aa for a step load of %In aa with proposed PSO-PID controller.

6 Modi, Barve and Chaturvedi 37.5 x -3 Plot for Power System.5 cy n ue q ig. 6. Change in tie line power for a step load of % at aa. 7 x -4 Plot for Load cy en u q ig. 7. Change in fquency in aa for a step load of % in aa with proposed PSO-PID controller..5 x -5 Plot for Power System -.5 ncy que ig. 8. Change in fquency in aa for a step load of % in aa with proposed PSO-PID controller.. x -3 Plot for Power System.8 y e n c qu ig. 9. Change in tie line power for a step load of % at aa.

7 Modi, Barve and Chaturvedi 38 VI. CONCLUSIONS AND UTURE WORK In this study, a new particle swarm optimized LC has been investigated for load fquency control of a Two aa power systems. or this purpose, first, to obtain mo adaptive tuning mechanism for the PID controller parameters and sensitivity of the system is incased. It has been shown that the proposed control algorithm is effective and provides significant improvement in system performance. Thefo, the proposed PSO-PID controller is commended to generate good quality and liable electric energy. In addition, the proposed controller is very simple and easy to implement since it does not qui many information about system parameters. It is found very clearly that the PSO based controller drastically duces the overshoot by a large value. Settling time, Rise Time and Peak Time have also improved. As a further study, the proposed method can be applied to multi aa power system load fquency control and also optimum values can be obtained by Particle Swarm optimization and compa with Genetic Algorithm. RERENCES [] A. Kumar, O.P. Malik, G.S. Hope, Variable structu- system control applied to AGC of an interconnected power system, IEE Proceedings, Vol. 3, Pt. C, No., pp. 3-9, January 985. []. Z.M. Ai-Hamouz, Y.L. Abdel-Magid, Variable structu load fquency controllers for multi aa power systems, Int. J. Electr. Power Energy Syst., 5 (5), pp. 3-9, 993. [3] Unbehauen, H., Keuchel, U., Kocaarslan, I., Real- Time Adaptive Control of Electrical Power and Enthalpy for a 75 MW Once- Through Boiler, Proceedings of IEE International Control Confence 9, Edinburg, Scotland, Vol., pp. 4-47, 5-8 March 99. [4] Shayeghi, H., Jalili, A., Shayanfar, H.A., Robust Modified GA Based Multi-Stage uzzy LC, Elsevier Energy Conversion and Management 48, , 7. [5] Al-Hamouz, Z., Al-Musabi, N., Al-Duwaish, H., A Tabu Search Approach or The Design of Variable Structu Load quency Controller Incorporating Model Nonlinearities, Journal of Electrıcal Engıneerıng, Vol. 58, No. 5, 64 7, 7. [6] Taher, S.A., Hemati, R., Abdolalipour, A., Tabie, S.H., Optimal Decentralized Load quency Control Using HPSO Algorithms in Degulated Power Systems, American Journal of Applied Sciences 5 (9): 67-74, 8. [7] J. Talaq and. Al-Basri, Adaptive fuzzy gain scheduling for loadfquency control, IEEE Trans. Power Syst., vol. 4, no., pp , eb [8] C.S. Chang and W.H. u, Aa load fquency control using fuzzygain scheduling of PI controllers, Elect. Power Syst. Res., vol. 4, no., pp , 997 [9] Eberhart,R., Kennedy,J., Particle swarm optimization, in Proc. IEEE Int. Conf. Neural Networks, vol. IV, Perth, Australia, 995, pp []. Y. L. Abdel-Magid, M. A. Abido, Robust Coordinated Design of Excitation and TCSC-based Stabilizers using Genetic Algorithms, Int. J. of Electrical Power & Energy Systems, Vol. 69, No. -3, pp. 9-4, 4. []. H.. Wang,. J. Swift, A Unified Model of ACTS Devices in Damping Power System Oscillations Part-: Single-machine Infinite-bus Power Systems, IEEE Trans. Power Delivery, Vol., No., pp , osha C.E., Elgerd O.I., The megawatt fquency control theory, IEEE Trans. Power Appl. Syst. (97) Vol.89, pp