自适应惯性权重的粒子群优化算法  

Particle Swarm Optimization Algorithm with Adaptive Inertial Weight

作　　者：张渊博 邹德旋 张春韵 杜星瀚 ZHANG Yuan-bo;ZOU De-xuan;ZHANG Chun-yun;DU Xing-han(School of Electrical Engineering and Automation,Jiangsu Normal University,Xuzhou Jiangsu 221116,China)

机构地区：[1]江苏师范大学电气工程及自动化学院,江苏徐州221116

出　　处：《计算机仿真》2023年第4期350-357,共8页Computer Simulation

基　　金：国家自然科学基金(62003152);江苏省研究生科研创新计划项目(2020XKT081)。

摘　　要：针对粒子群优化算法容易陷入局部极值、进化后期的收敛速度慢和精度低等问题,提出自适应惯性权重的粒子群优化算法。算法采用自适应更新惯性权重,添加影响算子,并通过惯性权重自适应调整学习因子,然后加入随机局部搜索策略;最后使用测试函数,通过和3种优化算法进行30次重复实验。结果表明,提出的算法具有更好的全局收敛能力,且收敛精度、和稳定性都有明显的提升。In order to solve the problems of particle swarm optimization(PSO),such as easy to fall into local extremum,slow convergence speed and low accuracy in the later stage of evolution,an adaptive inertia weight PSO algorithm is proposed.The algorithm adaptively updated the inertia weight,added the influence operator,adaptively adjusted the learning factor through the inertia weight,and then added the random local search strategy.Finally,the test functions and three optimization algorithms were used for 30 repeated experiments.The results show that the proposed algorithm has better global convergence ability,and the convergence accuracy and stability are significantly improved.

关 键 词：粒子群优化算法 自适应惯性权重 影响算子 学习因子 测试函数 收敛精度 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]