Optimization of PID Controller based on Taguchi Combined Particle Swarm Optimization for AVR System of Synchronous Generator

This paper presents the optimal design for PID controller parameters by using Taguchi combined particle swarm optimization (TPSO) method for automatic voltage regulator (AVR) system in the synchronous generator. Taguchi method consists of two sub-methods which each method is to consider the analysis of means (ANOM) and the analysis of variance (ANOVA) by using the orthogonal array to design effectively the minimum number of experiments for finding the minimum of the maximum percent overshoot, the rise time, the settling time and the steady-state error. These values can show the quality of terminal voltage response of AVR system. ANOM examines the means which are significantly different from the overall response's combined means to find approximate values of PID controller parameters while ANOVA determines the two most influential parameters with the response of AVR system. Taguchi method can reduce the number of experiments and find approximate values of PID controller parameters; however, Taguchi method cannot be used alone for obtaining the sufficient accurate parameters. Taguchi method should be combined with other methods in order to get the best solution. In this paper, Taguchi method will be combined with PSO method since PSO approach will improve the result from Taguchi method in convergence time and can generate the high-quality solution. In order to show the effectiveness result of TPSO technique, the step response of AVR system is used in the simulation software to compare the result with particle swarm optimization (PSO) and the Taguchi combined genetic algorithm (TCGA) method. The result shows that TPSO technique is provided the better result when compared with other approaches.