GA-EAM based hybrid algorithm

Abstract

The methods of searching optimal solutions are distinct in different evolutionary algorithms. Some of them do search by exploiting whereas others do by exploring the whole search space. For example Genetic Algorithm (GA) is good in exploitation whereas the Environmental Adaption Method (EAM) performs well in exploring the whole search space. Individually these algorithms have some limitations. In this paper a new hybrid algorithm has been proposed, which is created by combining the techniques of GA and EAM. The proposed algorithm attempts to remove the limitations of both GA and EAM and it is compared with some state-of-the-art algorithms like Particle Swarm Optimization-Time Variant Acceleration Coefficient (PSO-TVAC), Self-Adaptive Differential Evolution (SADE) and EAM on six benchmark functions with experimental results. It is found that the proposed hybrid algorithm gives better results than the existing algorithms. © 2014 Springer International Publishing Switzerland.