基于MMST分组的一类MIMO非线性系统执行器故障自适应补偿控制

摘要

在多输入多输出(Multiple-input multiple-output, MIMO)非线性系统的执行器故障容错控制问题中,控制器能够处理的执行器故障集合的大小与执行器分组方法有很大关系。为扩大系统可处理的执行器故障集合,本文针对一类具有执行器故障的MIMO非线性最小相位系统,提出基于多模型切换(Multiple model switching and tuning, MMST)执行器分组的自适应补偿控制方法。考虑系统的执行器卡死、部分失效和完全失效故障,在微分几何反馈线性化的基础上,研究基于多模型切换的执行器分组切换指标和切换策略,设计了基于反演控制的自适应补偿跟踪控制律,所设计的控制律能保证系统在执行器故障时闭环稳定,渐近跟踪给定的参考信号,且提出的分组方法扩大了可补偿的执行器故障集合。仿真结果表明了本文设计方法的有效性。%For multiple-input multiple-output (MIMO) nonlinear systems fault tolerant control problems, compensable actuator failure sets are highly interrelated with the actuator grouping method. In order to enlarge the set of compensable actuator failures, an adaptive compensation control scheme with multiple model switching and tuning (MMST) actuator grouping is proposed for a class of nonlinear MIMO minimum phase systems with multiple actuator failures. Based on differential geometry feedback linearization, a backstepping adaptive compensation control law is designed for the systems with actuators locked in space or losing of effectiveness, and a new actuator grouping scheme based on MMST which enlarges the set of compensable actuator failures is proposed. The proposed control law can guarantee that the closed-loop systems with actuator failures are stable and asymptotically track of the given reference signals. Simulation results demonstrate the effectiveness of the proposed method.