A hybrid optimization method of Particle Swarm Optimization and Cultural Algorithm

A novel hybrid optimization algorithm, PSOCCA, based on the fusion of the Particle Swarm Optimization (PSO) and Cultural Algorithm (CA), is proposed in this paper to enhance the convergence characteristics of the original PSO. In the PSOCCA, the particle swarm is considered to be a population component of the CA, and three knowledge sources are stored in the belief space to update the population space and establish the relationship between the two CA spaces. A unique mutation operator inspired by the Differential Evolution (DE) is used to mutate those particles with unsatisfactory performances in the population space. A new kind of knowledge, average value knowledge, that is stored in the belief space, is also introduced to regulate the variation ratio in the swarm population. The optimization performance of our PSOCCA is investigated using ten high-dimension and multi-peak functions. Simulation results demonstrate that it can be superior to the regular PSO.