【摘要】 针对粒子群算法应用于复杂函数优化时可能出现过早收敛于局部最优解的情况,提出了一种改进的算法结构.通过构造单个粒子的多个进化方向和类似于蚂蚁群算法信息素表的选择机制,保留了粒子的多种可能进化方向,并对全局最优解进行变异.提高了粒子间的多样性差异,从而改善算法能力.改进后的粒子群算法的性能优于带线性递减权重的粒子群算法.更多还原

【Abstract】 To improve the PSO algorithm which is a new population based optimization algorithm against trapping into local minima,a new PSO algorithm combing ant colony strategy and mutation strategy with PSO(AMPSO) is presented.In AMPSO some potential evolution directions are constructed for each particle in PSO,at the same time a strategy is presented to choose which one may be the local best for PSO evolution process just like pheromone table in ant colony algorithm,and the global optimization is mutated,Thus algorithm is improved.It is shown by tested with well-known benchmark functions that AMPSO algorithm is better than PSO algorithms with linearly decreasing weight.更多还原