混合变异算子的自适应粒子群优化算法  

Adaptive particle swarm algorithm with hybrid mutation operator

作　　者：安晓会[1] 高岳林[2] 

机构地区：[1]宁夏大学数学计算机学院,银川750021 [2]北方民族大学信息与系统科学研究所,银川750021

出　　处：《计算机应用》2008年第B06期28-30,共3页journal of Computer Applications

基　　金：中国博士后基金资助项目(20060601001);国家教育部社科规划项目(06JA630056)

摘　　要：针对惯性权重线性递减粒子群算法(LDW)不能适应复杂的非线性优化搜索过程的问题,提出了一种非线性递减的惯性权重策略,使算法很快地进入局部搜索,并在算法中引入混合变异算子,克服算法易早熟收敛的缺陷。对几种典型函数的测试结果表明,本文算法的收敛速度和收敛精度都明显优于LDW算法。A new inertia weight with nonlinearly descending strategy was presented to solve the problem that the Linearly Decreasing Weight （LDW） of particle swarm algorithm could not adapt to the complex and nonlinear optimization process, which could quickly put the algortthm into local searching. Further more, mutation operator was introduced to overcome the problem of the premature and low precision of the standard PSO. The algorithm of LDW-PSO and our method were tested with five well-known benchmark functions. The experiments show that the convergence speed and accuracy of our method are significantly superior to that of LDW-PSO.

关 键 词：粒子群算法 自适应惯性权重 变异算子 全局优化 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程]