An improved particle swarm optimization algorithm

A hierarchical structure poly-particle swarm optimization (HSPPSO) approach using the hierarchical structure concept of control theory is presented. In the bottom layer, parallel optimization calculation is performed on poly-particle swarms, which enlarges the particle searching domain. In the top layer, each particle swam in the bottom layer is treated as a particle of single particle swarm. The best position found by each particle swarm in the bottom layer is regard as the best position of single particle of the top layer particle swarm. The result of optimization on the top layer particle swarm is fed back to the bottom layer. If some particles trend to local extremum in particle swarm optimization (PSO) algorithm implementation, the particle velocity is updated and re-initialized. The test of proposed method on four typical functions shows that HSPPSO performance is better than PSO both on convergence rate and accuracy.