Chaotic Optimization Versus Genetic Algorithm for Optimal Tuning of Static Synchronous Series Compensator Stabilizing Controller

Abstract

Two robust meta-heuristic optimization algorithms: Chaotic Optimization (CO) and Genetic Algorithm (GA) were competing in this article for optimal tuning of the Static Synchronous Series Compensator (SSSC) stabilizing controller. The functionality of these two approaches was thoroughly investigated regarding the computation requirement and the solution quality. The tuning of the SSSC secondary controller was treated here as nonlinear constrained optimization problem. The objective functions considered the local and intra-area signals, to enhance the controller applicability and robustness. A traditional tuning for this auxiliary controller via eigenvalues was used as a reference for the competitive candidates: CO and GA. The dynamic performances of Single Machine Infinite Bus (SMIB) and Multi Machine Power (MMP) systems equipped with SSSC were investigated under several transient disturbances and fault scenarios, while the SSSC damping controller was tuned via CO, GA and eigenvalues approaches. Matlab and its dynamic platform, Simulink, were used to simulate the systems under concern. The simulation results demonstrated the effectiveness and the ability of the two optimization techniques in restoring system stability following large disturbance and/or fault while operating at different loading conditions either in SMIB or MMP systems. They successfully damped out electromechanical oscillations provoked by disturbances/faults. Moreover, they maintained system stability even under hypothetical operating conditions, where the eigenvalues technique was producing inadequate response. However, CO was found to produce better performance and require smaller computation time and storage than GA.