Sodoma: Self-organizing migrating algorithm in dyamic environment

Abstract

This paper is an experimental study investigating the capability of Self-Organizing Migrating Algorithm (SOMA) to solve dynamic problems. We have proposed an extension of SOMA algorithm based on the phenomenon of the limited lifetime of population individuals to keep the population diversity during the environment changes. We tested the performance of the algorithm on three representative dynamic benchmarks including moving peak, the 3rd de Jong function and two-extremes time-varying function. The experimental results confirmed the capability of the proposed dynamic SODOMA algorithms to effectively adapt the search process towards the nonstationary global optimum.