An Orthogonal Local Search Genetic Algorithm for the Design and Optimization of Power Electronic Circuits

Abstract

In this paper, an orthogonal local search genetic algorithm (OLSGA) is proposed for the design and optimization of power electronic circuits. The genetic algorithm is accelerated with a fast local search operator that automatically adjusts the search direction and the step size. An experimental design method called orthogonal design is used to determine the most promising direction of the potential region in the local search. In each generation, the step size is adaptively expanded or shrunk according to whether there is a newly improvement in the given local region. As a result, with proper direction and step size, the local search operator is able to stride forward and provide better exploitation ability to speed up the convergence rate of the genetic algorithm. The proposed method is applied to design and optimize a buck regulator. The results in comparison with other published results indicate that our proposed algorithm is effective and efficient.