基于反馈控制系统实时寿命预测的延寿策略研究  

Study on life extension strategy based on real-time life prediction of feedback control system

作　　者：李炜[1,2,3] 李宗仁 毛海杰[1,2,3] LI Wei;LI Zong-ren;MAO Hai-jie(College of Electrical and Information Engineering,Lanzhou Univ.of Tech.,Lanzhou 730050,China;Key Laboratory of Gansu Advanced Control for Industrial Processes, Lanzhou Univ. of Tech., Lanzhou 730050, China;National Demonstration Center for Experimental Electrical and Control Engineering Education,Lanzhou Univ.of Tech.,Lanzhou 730050,China)

机构地区：[1]兰州理工大学电气工程与信息工程学院,甘肃兰州730050 [2]兰州理工大学甘肃省工业过程先进控制重点实验室,甘肃兰州730050 [3]兰州理工大学电气与控制工程国家级实验教学示范中心,甘肃兰州730050

出　　处：《兰州理工大学学报》2021年第6期74-83,共10页Journal of Lanzhou University of Technology

基　　金：国家自然科学基金(62063017,61364011,61763027)。

摘　　要：针对反馈控制系统性能因执行器隐含退化而无法满足控制任务时限要求的问题,提出了一种基于反馈系统实时寿命预测的动态矩阵控制(DMC)延寿方法.该方法首先基于Wiener和Poisson过程,建立了更加贴合执行器实际退化过程的复合模型;其次,通过分析执行器隐含退化对控制系统性能的影响给出了系统剩余寿命(RUL)的解析解,实现了系统RUL的实时快速预测;接着依据RUL的预测信息并结合DMC方法,通过在线针对性地对约束矩阵Q和R元素值的共同调整来延长系统的RUL,确保了控制任务时限性要求的目标达成;最后以双容水箱液位系统验证了本文方法的有效性,且较已有的LQR方法延寿效果和鲁棒性都更优.Aiming at the problem that the performance of the feedback control system cannot meet the time limit of the control task due to the implicit degradation of the actuator,a dynamic matrix control(DMC)life extension method based on the real-time life prediction of the feedback system is proposed.Firstly,based on the Wiener and Poisson process,a composite model more suitable for the actual degradation process of actuator is established.Secondly,by analyzing the impact of the implicit degradation of the actuator on the performance of the control system,the analytical solution of the remaining life(RUL)of the system is given,and the real-time and rapid prediction of system RUL is realized.Then,based on the RUL prediction information and combined with the DMC method,the RUL of the system is extended by the online targeted adjustment of the element values of the constraint matrix Q and R,so as to ensure the achievement of the time limit requirements of control tasks.Finally,the effectiveness of the proposed method is verified by a double tank liquid level system,and the life extension effect and robustness are better than the existing LQR method.

关 键 词：执行器 复合退化 系统寿命预测 鲁棒性 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]