具有执行器故障的随机鲁棒预测容错切换控制

Random Robust Predictive Fault-tolerant Switching Control with Actuator Fault

针对一类具有不确定性、未知扰动、执行器随机故障的工业过程,提出了一种随机鲁棒预测容错切换控制(SRPFTSC)方法.其主要贡献是融合了随机控制理论与鲁棒预测控制方法,解决了实际工业过程中执行器存在一定概率故障的问题,改进了传统的容错控制(FTC)方法.首先,建立含有系统状态增量和输出跟踪误差的新型多自由度状态空间扩展模型,以描述一类具有一定不确定性、未知扰动、执行器随机故障的工业过程;其次,基于该扩展模型设计含有故障随机特性的反馈控制律,实现了故障情况下的FTC控制和无故障情况下的正常控制.进一步给出了具有线性矩阵不等式(LMI)形式的相关定理和推论.同时利用指数稳定证明给出容错控制器切换的稳定条件,使系统在H∞意义下满足渐近稳定的要求.最后,以注塑成型保压过程为例,验证了该方法的有效性和可行性.

We propose a robust method for realizing stochastic predictive fault-tolerant switching control(SRPFTSC)for a class of industrial processes characterized by uncertainty,unknown disturbance,and random actuator failure.Our main contribution is the integration of the stochastic control theory with a robust predictive control method,which solves the problem of actuator failure in actual industrial processes and performs better than the traditional fault-tolerant control(FTC)method.First,we establish a new state-space extended model with multiple degrees of freedom,a system state increment,and output error tracking to describe industrial processes characterized by uncertainties,unknown disturbances,and random actuator failures.Secondly,based on the extended model,we design a feedback control law with fault stochastic characteristics to realize FTC with faults and normal control without faults.Correlation theorems and corollaries with linear matrix inequality are provided.To ensure that the system meets the requirement of H∞asymptotic stability,the stability condition of the switching of the fault-tolerant controller is obtained using an exponential stability proof.The effectiveness and feasibility of the method are verified by the results obtained from an injection molding process.