基于时滞分解的线性时滞系统主动容错H_∞控制

Delay Decomposition Approach to Active Fault Tolerant H_∞ Control for Linear Time-Delay Systems

采用一种具有广义内模控制结构的反馈控制器,研究一类线性时滞系统的主动容错H∞控制问题。针对可由外系统描述其动态行为的故障,设计基于观测器的故障估计器,并将故障估计用于反馈控制的故障补偿。采用一种新的时滞分解方法选取Lyapunov-Krasovskii泛函,推导并证明问题可解的充分条件,最后给出主动容错控制系统的设计算法。最后通过算例验证了提出方法的有效性,并且利用时滞分解方法,保守性减小。

Active fault tolerant H∞ control problem for a class of linear time-delay systems is studied by employing a generalized internal model control architecture. For the possible fault whose dynamic behavior can be described by an exosystem, an observer-based fault estimator is used to estimate the fault. Based on the fault estimation, the active fault tolerant control system can implement online fault compensation. Using a new delay decomposition approach to choose the Lyapunov-Krasovskii functional, delay dependent sufficient conditions on the existence of such an active fault tolerant control system are derived, and a brief design procedure is also provided. Finally, a numerical example is given to demonstrate the reduced conservatism and effectiveness of the proposed techniques