Dynamic Mutation Control in Continuous Genetic Algorithms

Abstract

In this paper the adaptability of the mutation operation in continuous genetic algorithms (CGAs) is taken into consideration from an analytical perspective. For this purpose, based on the notation that has previously been used to analyze the classical, binary genetic algorithm, a dynamic system model of CGA has been created. In order to adapt the mutation probability in successive generations, a linear controller has been applied. It allows one to accelerate the evolution process. As a result, faster convergence is obtained, as required in computationally intensive optimization problems.