A multi-swarm cooperative hybrid particle swarm optimizer

Cooperative approaches have proved to be very useful in evolutionary computation. This paper a novel multi-swarm cooperative particle swarm optimization (PSO) is proposed. It involves a collection of two sub-swarms that interact by exchanging information to solve a problem. The two swarms execute IPSO (improved PSO) independently to maintain the diversity of populations, while introducing extremal optimization (EO) to IPSO after running fixed generations to enhance the exploitation. States of the particles are updated based on global best particle that has been searched by all the particle swarms. Synchronous learning strategy and random mutation scheme are both absorbed in our approach. Simulations on a suite of benchmark functions demonstrate that this method can improve the performance of the original PSO significantly.