A Hybrid Particle Swarm Optimizer with Adaptive Pattern Search

摘要：

In this paper, we present a novel hybrid algorithm based on the adaptive pattern search method and particle swarm optimization. In the proposed PSO, a new adaptive pattern search procedure is applied to the global best particle to extend more latitude of search space to anchor the global optimum. Moreover, we introduce a velocity disturbance strategy to add population diversity since a particle tends to be stagnant when its velocity is near to zero. The main purpose of this paper is to demonstrate how the standard particle swarm optimizer can be improved by incorporating a hybridization strategy. The experimental results on a series of well-known benchmarks show that the hybrid PSO optimizer outperforms the standard PSO and some up-to-date PSO procedures appearing in the literature in terms of solution quality and convergence rate.

作者: Ping Yan Lixin Tang

作者单位: Logistics Institute of Northeastern University in Shenyang of China

会议类型: 国际会议

会议名称: Third International Symposium on Intelligence Computation and Applications(ISICA 2008)(第三届智能自动化、计算与制造国际研讨会)

会议地点: 武汉

会议语种:英文

页码: 325-329

在线出版日期: 2008-12-19（万方平台首次上网日期，不代表论文的发表时间）