摘要
通过分析粒子群算法和人工鱼群算法的优缺点,利用粒子群算法收敛速度快及人工鱼群算法能较好地收敛到全局最优解的特点,提出了一种新的混合算法.算法以粒子群为基础进行设计,根据人工鱼群的公告板、群聚和随行策略的模式对粒子群进行速度与位置变更,使原有的粒子群变成具有一定智能的粒子,从而达到提高搜索精度及效率的目的.通过Generalize-Schwefel等3个经典函数进行优化仿真后发现,该混合算法具有搜索精度更高及收敛速度更快的特点,同时该算法在求解高维问题时具有明显优势.
A new hybrid algorithm with fast convergence speed and capability of searching optimal solution within defined space was proposed by uniting the advantages of particle swarm optimization and artificial fish swam algorithm.In the new algorithm,the velocity and position of the particle swarm were modified in optimization according to the bulletin boards,cluster of artificial fish and accompanying strategy model.Then the original particle swarm was turned to intelligent particles and the search precision and efficiency were improved.The simulations prove that the new hybrid algorithm possesses the characteristic of higher accuracy search and faster convergence by using three classic functions in optimization,like Generalize-Schwefel function etc.
出处
《上海理工大学学报》
CAS
北大核心
2014年第3期223-226,238,共5页
Journal of University of Shanghai For Science and Technology
基金
上海市教委科研创新(重点)资助项目(14ZZ131)
上海市研究生创新基金资助项目(JWCXSL1302)
关键词
粒子群算法
人工鱼群算法
混合算法
函数
particle swarm optimization （ PSO ）
artificial fish school algorithm （AFSA ）hybrid algorithm
