An improved cooperative particle swarm optimizer

Particle Swarm Optimization (PSO) is a population-based technique for optimization, which simulates the social behavior of the bird flocking, a novel Adaptive Cooperative PSO (ACPSO) with adaptive search is presented in this paper, the proposed approach combines both cooperative learning and PSO with adaptive inertia weight, cooperative learning is achieved by splitting a high-dimensional swarm into several smaller-dimensional subswarms to combat curse of dimensionality, the adaptive inertia weight is employed to control the balance of exploration and exploitation in all the smaller-dimensional subswarms, which cooperate with each other by exchanging information to determine composite fitness of the entire system. Finally, computer simulations over three benchmarks indicate that the proposed algorithm shows better convergence behavior, as compared to the Cooperative Genetic Algorithm (COGA), the PSO, and the CPSO, and then its adaptive search behavior is analyzed, demonstrating its superiority.