APP下载

基于改进粒子群算法的PID参数优化方法研究

2025-08-06孔明张琳张龙龙

计算技术与自动化 2025年2期
关键词:参数优化

摘 要:为解决粒子群优化(PSO)算法探索和利用不平衡导致寻优能力下降问题,提出了一种改进的PSO(IPSO)算法。首先,引入莱维飞行建立种群初始搜索模型提升算法探索能力;其次,建立种群探索行为模型、种群开发行为模型和探索与开发的平衡模型,自适应地平衡寻优过程中算法的探索和利用性能;最后,设计基于IPSO算法的直流电机PID控制器参数优化流程。实验结果表明,相较于对比算法,IPSO算法在CEC 2017函数问题上是最优算法,基于IPSO算法的PID控制器具有更优的控制性能,提出的方法为工程中PID控制器的参数优化提供了参考。

关键词:PID;PSO算法;探索和利用;莱维飞行;参数优化

中图分类号:TP391文献标识码:A

Study on PID Parameter Optimization Method Based on Improved

Particle Swarm Optimization Algorithm

KONG Ming1,ZHANG Lin2,ZHANG Longlong1

(1.Shinva Medical Instrument Co., Ltd., Zibo, Shandong 255000, China;

2.Shandong Xinhua United Orthopedic Equipment Co., Ltd., Zibo, Shandong 255000, China)

Abstract:In order to solve the problem of decreasing the optimisation search ability due to the imbalance between exploration and exploitation of particle swarm optimisation (PSO) algorithms, an improved PSO (IPSO) algorithm is proposed. Firstly,Lévy flight is introduced to establish a population initial search model to enhance the algorithm’s exploration ability; secondly, a population exploration behaviour model, a population exploitation behaviour model and an exploration exploitation equilibrium model are established to adaptively balance the algorithm's exploration and exploitation performances in the optimization process; and lastly, a parameter optimization process of the PID controller of a DC motor based on the IPSO algorithm is designed. The experimental results show that the IPSO algorithm is the best algorithm in the CEC 2017 function problem compared to the comparison algorithm, and the PID controller based on the IPSO algorithm has a better control performance, and the proposed method provides a reference for the parameter optimisation of PID controllers in engineering.

Key words:PID; PSO algorithm; exploration and exploitation; Lévy flight; parameter optimization

比例积分微分(Proportional Integral Differential,PID)控制器是一种应用广泛的机器人控制器,具有结构简洁、鲁棒性强和布置方便等优势[1-4]。PID控制器包括三个重要参数,分别是比例参数、积分参数和微分参数,参数与受控系统的匹配程度直接决定了PID控制器的控制效果。目前,PID参数的整定方法可主要分为两类,分别是经验整定方法和基于智能算法的整定方法[5]。相较于经验整定方法,基于智能算法的整定方法具有时间成本低和调节精度高的优势。

基于群体的元启发式算法是一类经典的智能算法,因其简洁、易实现且寻优性能优异被广泛地用于PID参数优化任务[6-9],改进算法性能有益于提升PID参数整定效果,进而提升PID控制器的控制效果。为了获取更优质的PID参数,该领域的研究人员围绕算法改进方法展开了大量研究工作。……

登录APP查看全文

猜你喜欢

参数优化
防眩网结构参数对夜间眩光控制效果分析
基于数值模拟的软弱围岩隧道初期支护参数优化研究