Genetic algorithm aided controller design with application to SVC

Abstract

Usually, it is difficult to use current methods to design a controller with combined continuous and discrete actions. The paper presents a novel approach to solving this problem, i.e. transforming the controller design problem into an optimisation problem and then searching the optimal (or suboptimal) controller parameters by a genetic algorithm. This new approach has been applied to the controller design of a static VAR compensator (SVC) with satisfactory results, which show that the new approach is simple yet powerful.