TWO STRATEGY COOPERATIVE PARTICLE SWARM OPTIMIZATION ALGORITHM WITH INDEPENDENT PARAMETER ADJUSTMENT AND ITS APPLICATION

摘要

Aiming at the premature convergence of particle swarm optimization (PSO) in solving complex multimodal problems, a two strategy cooperative particle swarm optimization algorithm (TSPSO) with independent parameter adjustment is proposed. In the proposed algorithm, the variance of population fitness, evolution ability and evolution rate of particles are defined firstly, and then the inertia weight and learning factor of each particle are adjusted adaptively, which effectively balances the exploitation and exploration ability of the algorithm. Secondly, according to the fitness value and evolution ability of particles in each generation, the population is divided dynamically. The inferior subgroup uses reconstruction strategy to generate new particles by learning from the particles in the superior subgroup, so as to speed up the convergence of the algorithm; the superior subgroup uses differential mutation to avoid the premature convergence of the algorithm, and maintain the diversity of the population. Finally, a large number of experiments are carried out on the CEC2013 standard test function set and flexible job shop scheduling problem, and the experimental results verify that TSPSO has high efficiency. The convergence analysis shows the effectiveness of the algorithm.