作　　者： ; ;

机构地区： 福建师范大学软件学院

出　　处： 《小型微型计算机系统》 2012年第1期130-134,共5页

摘　　要： 在求解高维空间中复杂多峰函数的实时优化问题时,传统的粒子群算法在收敛速度和局部搜索能力等方面表现出严重不足,针对这些问题,启发于鲦鱼效应的生物现象,引入团队领导机制,提出基于多leader交叉的PSO算法MLCPSO,该算法集成了两种新的粒子飞行策略.实验表明,从实验结果的平均情形上看,与SGA算法与SPSO算法相比较,MLCPSO算法具有更优的收敛性与扩展性. For realtime optimization of complex multi-peaks function with high dimensions, the traditional PSO has some serious disadvantages such as slow convergence and weak ability of local search. Inspired by minnow effect in biology, a novel PSO algorithm, known as MLCPSO, is proposed. The new MLCPSO algorithm is based on the team leadership mechanism, and characterized with a multi-leader crossover and two new particle swarm flying strategies. The results from our extensive experiment have indicated that in general, MLCPSO has much better convergence and extensibility than SGA and SPSO.

关 键 词： 群智能 粒子群算法 交叉 团队领导

领　　域： [自动化与计算机技术] [自动化与计算机技术]