Two sub-swarms evolutionary particle swarm optimization based on team progress learning

The sociological background of particle swarm optimization is analyzed, and for preventing from premature convergence, a modified approach to enhance organizational management mechanism in group is proposed. Two sub-swarms evolutionary particle swarm optimization based on team progress learning is proposed by borrowing ideas from social division and progress learning ideas in management team. The team members are divided into the elite and plain groups in the algorithm, the role epitomes of both groups are determined, the manipulations of learning and exploration are properly defined, and the member renewal rules are reasonably established. The abilities of global, local refined search are possessed in the algorithm during the search procedure. Simulation experiments are performed with four benchmark functions, and the significant performance in quality of the optimal solutions, the global search ability and convergence speed of the algorithm are validated in this paper.