动态邻域混合粒子群优化算法

Dynamic Neighborhood Hybrid Particle Swarm Optimization Algorithm

粒子群优化(PSO)算法对于多峰搜索问题一直存在早熟收敛问题。为在增强PSO算法全局搜索能力的同时提高收敛速度,提出一种动态邻域混合粒子群优化算法DNH_PSO,采用PSO局部模型,将随机拓扑和冯诺依曼拓扑相结合形成动态邻域,提高算法的全局搜索能力,为增强算法的局部搜索能力并加快收敛速度,使用粒子邻域全面学习策略,将拟牛顿法引入算法中。与其他PSO实验对比分析表明,该算法对于多峰搜索问题具有较好的全局收敛性。

Particle Swarm Optimization（PSO） algorithm has existed premature convergence for multimodal search problems. In order to enhance the global search ability and increase the speed of convergence, this paper proposes a Dynamic Neighborhood Hybrid Particle Swarm Optimization（DNH PSO） algorithm using local particle swarm model, the random topology and the von Neumann topology are combined to form dynamic neighborhood topology, improving the algorithm＇s global search ability, meanwhile in order to enhance the local search ability and convergence speed, the use of particles neighborhood comprehensive learning strategy, and introduction of quasi-Newton method. Experimental comparative analysis with other variant PSO shows that the algorithm for the multimodal search problems has better global convergence.