摘要： 针对引力搜索算法局部搜索能力较弱,搜索过程容易出现早熟的现象,提出一种基于多样性和局部优化能力协同优化的引力搜索算法。将粒子群算法中局部最优解和细菌趋化中排斥操作的概念引入到引力搜索算法中,通过帮助粒子接近最优位置和逃离最差位置,改进了搜索算法中粒子的局部优化能力及种群多样性,并使用标准函数进行测试。结果表明,该算法能够实现全局搜索与局部搜索的平衡,最大程度地保持种群多样性,提高算法搜索能力。

关键词: 引力搜索算法, 细菌趋化, 排斥操作, 多样性, 局部搜索, 最优位置

Abstract: Aiming at the problem that the Gravitational Search Algorithm(GSA) plays bad performance in the local search ability and is easy to get into the premature convergence in the searching process,a new GSA based on Diversity and Local Optimization strategy(DLOGSA) is proposed.The ideas about the local optimal solution of particle swarm optimization and the chemo-repellents of bacterial chemotaxis are introduced to the gravitational search algorithm,which improves the local optimization ability of particles and the diversity of population in GSA through helping the particles approach the optimal position and flee the worst position.The validity of the improved method is confirmed by testing the benchmark functions.Results show that the GSA based on the improvement of diversity and local optimization ability can balance the global search ability and the local search ability,keep the diversity of population at utmost,and improve the search capability substantially.

Key words: Gravitational Search Algorithm(GSA), bacterial chemotaxis, exclusive operation, diversity, local search, optimal position

中图分类号: 

• TP18