带变异算子的粒子群优化算法

针对PSO算法存在易陷入局部最优点的缺点,该文提出了带变异算子的PSO算法。在算法搜索的后期引入变异算子,使算法摆脱后期易于陷入局部极优点的束缚,同时又保持前期搜索速度快的特性。通过对三个多峰的测试函数和一个问题空间为非凸集的实例所做的对比实验,表明改进的PSO算法增强了全局搜索能力,搜索成功率得到大大提高,克服了基本PSO易于收敛到局部最优点的缺点。

Particle Swarm Optimization with Mutation Operator

Aiming at the shortcoming of PSO algorithm,that is,easily plunging into the local minimum,we opposed an advanced PSO algorithm with mutation operator.By adding the mutation operator to the algorithm in the later phase of convergence,the advanced algorithm can not only escape from the local minimum's basin of attraction of the later phase,but also maintain the characteristic of fast speed in the early convergence phase.By the contrast experiments of three multimodal test functions and a example whose problem space is non-convex set,it has been proved that the advanced PSO algorithm can improve the global convergence ability,greatly enhance the rate of convergence and overcome the shortcoming of basic PSO algorithm,that is,easily plunging into the local minimum.