【摘要】 为提高粒子群优化的求解性能,在分析了粒子群优化原理的基础上,给出了两种混沌映射的映射规则.构建了基于Logistic映射的混沌粒子群优化方法以及基于Lozi s映射的混沌粒子群优化方法,并给出了两类约束条件的处理方法.采用基于Logistic映射的混沌粒子群优化方法和基于Lozi s映射的混沌粒子群优化方法以及标准粒子群优化方法分别对benchmark有约束优化实例进行求解.对各种方法获得的最优解、成功率指标、平均有效迭代数、迭代占用时间等方面作对比,结果表明:采用基于Lozi s射映的混沌粒子群优化方法具有求解精度高、优化效率高等优点.更多还原

【Abstract】 In order to improve the solving performance of particle swarm optimization(PSO),first the basic principle of PSO is analyzed and mapping rules for two types of chaos map are given.Next,PSO based on the Logistic map(LGM-PSO) and Lozi’s map(LZM-PSO) are constructed,and the methods for treating two types of constraints are given.To compare the performances of LGM-PSO,LZM-PSO and the standard particle swarm optimization,the three methods are used to solve the benchmark constrained optimal problem.Their performances are compared in terms of optimal solution,success ratio,average valid evaluation number,iterative occupancy hours and so on.Comparison results indicate that the LZM-PSO has many advantages,such as higher solution accuracy and higher computational efficiency.更多还原