基于随机差分进化算法的最优PID参数整定方法  

Optimal PID Parameter Tuning Method Based on Stochastic Differential Evolution Algorithm

作　　者：杨茂华 谭飞 尹宋麟 YANG Maohua;TAN Fei;YIN Songlin(School of Automation and Information Engineering,Sichuan University of Science&Engineering,Yibin 644000,China;Sichuan Provincial Key Laboratory of Artificial Intelligence,Yibin 644000,China)

机构地区：[1]四川轻化工大学自动化与信息工程学院,四川宜宾644000 [2]人工智能四川省重点实验室,四川宜宾644000

出　　处：《四川轻化工大学学报（自然科学版）》2022年第6期50-56,共7页Journal of Sichuan University of Science & Engineering(Natural Science Edition)

基　　金：国家自然科学基金项目(61902268);四川省科技计划项目(2018JY0197,19ZDZX0037,2019YFSY0045,2016SZ0074)。

摘　　要：针对PID(Proportion Integration Differentiation)控制参数整定方法的经验性,在差分进化算法寻优的基础上,提出一种基于随机差分进化(Stochastic Differential Evolution,SDE)算法的最优PID参数整定方法。将差分进化算法进行改进,引入随机变异和随机学习因子,同时选择精英个体进行交叉,既保证了种群的多样性,又提高了算法的优化速度和精度。以ITAE为误差指标进行差分算法的PID参数整定,通过大量的参数整定结果结合内模PID控制理论,提取出PID参数整定公式。最终结合基于随机差分进化算法的高阶系统用一阶时滞系统近似,解决高阶系统的参数整定。通过Matlab验证,该公式得到了更好的PID整定结果,较普通PID参数整定公式和智能算法PID参数整定效率更高。Aiming at the experience of PID(Proportion Integration Differentiation) control parameter tuning method, an optimal PID control parameter tuning method based on Stochastic Differential Evolution(SDE)algorithm is proposed based on the optimization of differential evolution algorithm. The differential evolution algorithm is improved, random mutation and random learning factors are introduced, and elite individuals are selected for crossover, which not only ensures the diversity of the population, but also improves the optimization speed and accuracy of the algorithm. Taking ITAE as the error index, the PID control parameter tuning of the differential evolution algorithm is carried out, and the PID control parameter tuning formula is extracted through a large number of parameter tuning results combined with the internal model PID control theory. Finally,combined with the high-order system based on the stochastic differential evolution algorithm, the first-order timedelay system approximation is used to solve the parameter tuning of the high-order system. Through Matlab verification, the formula obtains better PID control parameter tuning results, which is better than the common PID control parameter tuning formula and intelligent algorithm PID control parameter tuning efficiency is higher.

关 键 词：PID参数整定 随机差分进化算法 一阶时滞系统 最优PID参数整定方法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]