基于免疫粒子群算法的PID参数优化研究  

Optimization of PID control parameters based on immune particle swarm optimization

作　　者：吕红芳[1] 王涛 嵇月强 高慧 徐斌 王秋婷 LYU Hongfang;WANG Tao;JI Yueqiang;GAO Hui;XU Bin;WANG Qiuting(School of Electrical Engineering,Shanghai Dianji University,Shanghai 201306,China;Technology Center,Hangzhou UWNTEK Automatic System Co.,Ltd.,Hangzhou 311100,Zhejiang,China)

机构地区：[1]上海电机学院电气学院,上海201306 [2]杭州优稳自动化系统有限公司技术中心,浙江杭州311100

出　　处：《中国工程机械学报》2022年第3期194-198,共5页Chinese Journal of Construction Machinery

基　　金：基于时间敏感网络技术的PLC资助项目(TC190A449);浙江省重点研发计划资助项目(2021C01032)。

摘　　要：针对传统PID控制系统中存在的调节时间长、控制精度低、响应速度慢等问题,提出使用免疫粒子群算法对PID参数进行优化。首先,对粒子群优化算法(PSO)中的惯性权重、学习因子、粒子学习模式进行改进,使得微粒更新适应各个阶段;其次,引入人工免疫思想形成免疫粒子群算法,保证了迭代过程中粒子的多样性,提高了算法精度;最后,在仿真环境下经过对ZN公式法、参数改进后的粒子群算法、免疫粒子群算法在PID控制参数优化效果进行对比可知,免疫粒子群算法的优化控制效果更佳。Aiming at the problems of long adjustment time,low control accuracy,and slow response speed existing in traditional PID control systems,an immune particle swarm algorithm is proposed to optimize PID parameters.First,the inertia weight,learning factor,and particle learning mode in the particle swarm optimization algorithm are improved to make the particle update adapt to each stage.Secondly,artificial immune thought is introduced to form an immune particle swarm algorithm,which ensures the diversity of particles in the iterative process and improves the accuracy of the algorithm.Finally,in the simulation environment,after the ZN formula method and the improved particle swarm algorithm with parameters,the comparison of the immune particle swarm algorithm in the PID control parameter optimization effect shows that the immune particle swarm algorithm has better optimization control effect.

关 键 词：惯性权重 学习因子 学习模式 粒子群优化算法 PID控制参数 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]