Multiswarms, exclusion, and anti-convergence in dynamic environments

Many real-world problems are dynamic, requiring an optimization algorithm which is able to continuously track a changing optimum over time. In this paper, we explore new variants of particle swarm optimization (PSO) specifically designed to work well in dynamic environments. The main idea is to split the population of particles into a set of interacting swarms. These swarms interact locally by an exclusion parameter and globally through a new anti-convergence operator. In addition, each swarm maintains diversity either by using charged or quantum particles. This paper derives guidelines for setting the involved parameters and evaluates the multiswarm algorithms on a variety of instances of the multimodal dynamic moving peaks benchmark. Results are also compared with other PSO and evolutionary algorithm approaches from the literature, showing that the new multiswarm optimizer significantly outperforms previous approaches.