Optimization of AVR Parameters of a Multi-machine Power System Using Particle Swarm Optimization

Abstract

In this paper, a method for optimizing the parameters of Automatic Voltage Regulation (AVR) system installed on the generators of a multi-machine power system using Artificial Intelligence (AI) techniques is presented. Each AVR system is equipped with a PID (Proportional, Integral and Derivative) controller and a Power System Stabilizer (PSS). Two methods are presented, which are the Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). The robustness of the AI algorithms is examined by studying the time-domain behavior of the system following different disturbances. The AI techniques provide a much simpler way to solve this non-linear system compared to classical techniques.

Keywords: multi-machine power system stability, AVR system, power system stabilizer, PID controller, particle swarm optimization, genetic algorithm.