基于动态矩阵控制算法的参数整定研究  

Research on parameter tuning based on dynamic matrix control algorithm

作　　者：臧春华 王培彧 ZANG Chunhua;WANG Peiyu(School of Shenyang Institute of Chemical Technology and Information Engineering,Shenyang 110142,China)

机构地区：[1]沈阳化工大学信息工程学院,沈阳110142

出　　处：《自动化与仪器仪表》2023年第5期22-26,共5页Automation & Instrumentation

摘　　要：针对动态矩阵控制算法中预测时域,控制时域,控制权重难以整定的问题,提出了一种基于遗传算法的改进型最优性能指标算法。方法通过利用遗传算法来优化各个参数的范围,随后结合控制系统响应曲线和最优性能指标来得出各个参数的最优值。最后以单变量模型和多变量模型为研究对象进行仿真实验,将文中方法和传统的DMC参数整定算法进行对比分析,MATLAB仿真结果表明:相比于传统的整定算法,文中提出的方法所整定的DMC控制系统具有更快的响应速度和更好的鲁棒性。An improved genetic algorithm-based optimal performance indicator algorithm is proposed for the dynamic matrix control algorithm in which the prediction time domain,control time domain,and control weights are difficult to be rectified.The method uses the genetic algorithm to optimize the range of each parameter,and then combines the control system response curve and the optimal performance index to derive the optimal value of each parameter.Finally,the univariate model and multivariate model are used for simulation experiments,and the comparison between the proposed method and the traditional DMC parameter tuning algorithm is carried out.The MATLAB simulation results show that the DMC control system tuned by the proposed method has faster response and better robustness than the traditional tuning algorithm.

关 键 词：动态矩阵控制 参数整定 最优性能指标 遗传算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]