Cellular gas with active components of PSO: mutation and crossover

Abstract

Two powerful metaheuristics being used successfully since their creation for the resolution of optimization problems are Cellular Genetic Algorithm (cGA) and Particle Swam Optimization (PSO). Over the last years, interest in hybrid metaheuristics has risen considerably in the field of optimization. Combinations of operators and metaheuristics have provided very powerful search techniques. In this work we incorporate active components of PSO into the cGA. We replace the mutation and the crossover operators by concepts inherited by PSO internal techniques. We present four hybrid algorithms and analyze their performance using a set of different problems. The results obtained are quite satisfactory in efficacy and efficiency.