基于ELM-ARX组合模型的预测控制算法研究及应用  

Research and Application of Predictive Control Algorithm Based on ELM-ARX Combined Model

作　　者：陈颖 赵朋程 贾红霞 杨丽[1] 马健 Chen Ying;Zhao Pengcheng;Jia Hongxia;Yang Li;Ma Jian(College of Mechanical and Electrical Engineering,Shijiazhuang University Hebei,Shijiazhuang 050035)

机构地区：[1]石家庄学院机电学院,河北石家庄050035

出　　处：《科技风》2022年第21期4-10,共7页

基　　金：石家庄市科学技术研究与发展计划资助项目(191130701);石家庄学院博士科研启动基金资助项目(20BS021);河北省高等学校科学技术研究项目资助,项目编号:BJK2022070;河北省应用技术大学研究会项目,项目编号:JY2022111;河北省“工+匠”机电专业群课程思政示范中心课程思政示范课程建设项目,项目编号:GJSZ202204。

摘　　要：针对非线性控制系统模型精度及抗干扰能力的问题,首先,采用极限学习机(ELM)建立控制系统的稳态非线性模型,采用具有外界输入的自回归算法(ARX)建立控制系统的动态线性模型,通过调节增益的方式实现稳态模型和动态模型的有机结合,构建控制系统的ELM-ARX组合模型;其次,提出基于ELM-ARX组合模型的预测控制算法,推导预测步数内控制系统的被控变量预测输出值,采用滚动修正的方法给出被控变量输入增量表达式;最后,通过Matlab数据仿真实验以及水箱控制应用实验,验证提出的基于ELM-ARX组合模型的预测控制算法。实验结果表明,提出的ELM-ARX组合模型预测控制算法可获得较好控制效果。Aiming at the problem of model accuracy and anti-interference ability of nonlinear control system,firstly,the steady-state nonlinear model of control system is established by Extreme Learning Machine(ELM),and the dynamic linear model of control system is established by Auto Regressive with eXogenous inputs(ARX).The ELM-ARX combined model of control system is established.Then,the predictive output value of the controlled variable in the predictive step control system is derived,and the input increment expression of the controlled variable is given by using the rolling correction method.The predictive control algorithm based on ELM-ARX combined model is proposed.Finally,the predictive control algorithm based on ELM-ARX combined model is verified by Matlab data simulation experiment and water tank control application experiment.The experimental results show that the proposed ELM-ARX combined model predictive control algorithm can achieve good control effect.

关 键 词：极限学习机 自回归算法 组合模型 预测控制 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]