An Improved Particle Swarm Algorithms for Global Optimization

Particle swarm optimization (PSO) algorithm is a robust and efficient approach for solving complex real-world problems. In this paper, a modified particle swarm algorithm (IMPSO) is introduced for unconstrained global optimization. The whole swarm is partitioned to three different sub-populations according to their fitness, and different velocity updating strategies are used to different sub-populations. Besides, we take advantage of crossover to maintain the diversity of the swarm and avoid getting into local optimum. IMPSO are extensively compared with other two modified PSO algorithms on three well-known benchmark functions with different dimensions. Experimental results show that IMPSO achieves not only better solutions but also faster convergence.