Cloud Estimation of Distribution Particle Swarm Optimizer

Cloud estimation of distribution particle swarm optimizer combining PSO and cloud model is introduced. In the algorithm´s offspring generation scheme, new particles are generated in the cloud estimation of distribution way or in the PSO way. The innovation of the algorithm is production of cloud particles according to the cloud model theory. The cognitive population obtained during optimization is used to estimate statistical characteristics of good solution regions by backward cloud generator. And then the estimated statistical characteristics are used to produce cloud particles by positive cloud generator. Both the global information from cloud particles and local information from PSO particles are used to guide the further search. The proposed algorithm is applied to some well-known benchmarks. The experimental results show that the algorithm has stronger global search ability than original version of PSO.