Analysis and Simulation on a Class of Improved Swarm Optimizer

A class of improved swarm optimizer with controllable velocity factor was proposed, based on three strategies for velocity control of searching particles, i.e. the completely random one, the partial controllable one and the completely controllable one. Aiming at its algebraic and dynamic performance, contrastive researches were made on convergence precision and computation expense of these swarm optimizers under several tracks for optimization with different velocity-changing features. Simulations showed that the BPSO could be improved to some extent, by using the controllable modes for velocity-updating. In particular, the completely controllable strategy proved to be highly effective on reducing computation expense of swarm optimization.