基于线性矩阵不等式的离散系统集成故障估计与容错控制

LMI-based Integrated Fault Estimation and Fault-tolerant Control for Discrete System

离散系统在进行故障估计与容错控制时,由于两者相互耦合,会产生较大的故障估计误差和系统输出误差。针对这一问题,提出了一种基于线性矩阵不等式的集成故障估计与容错控制方法。首先为系统设计了故障观测器和容错控制器,然后基于两者的耦合关系,建立了系统误差和状态的增广系统模型;其次,基于D-稳定引理和H_∞控制理论为增广系统设计控制器,进而利用线性矩阵不等式求得故障观测器和容错控制器的增益矩阵,实现多性能约束条件下的集成故障估计与容错控制。最后以飞行控制系统验证了所提方法的有效性,与传统分离设计方法相比,故障估计误差减小了66.7%,估计的快速性提升了70%。

The coupling between fault estimation and fault-tolerant control will induce larger errors of fault estimations and system outputs of discrete systems. In order to solve the problem, an integrated fault estimation and fault-tolerant control method based on linear matrix inequality(LMI) is proposed. Firstly, the fault observer and fault-tolerant controller are designed. The augmented system model of the errors and states is established based on the coupling relationship between them. Then, the controller is designed for the augmented system by using D-stability lemma and H_∞ control theory. The integrated design of fault estimation and fault-tolerant control under multiple performance constraints is realized and the gain matrices of the fault observer and fault-tolerant controller can be obtained by linear matrix inequality. Finally, the effectiveness of the proposed method is verified by the flight control system. Compared with the traditional separation design method, the estimation error of the fault is reduced by 66.7%, and the speed of estimation is improved by 70%.