基于观测器的离散互联系统的故障诊断

Observer-based fault diagnosis of discrete interconnected systems

研究了一类不确定离散互联系统的故障诊断问题。将改进的快速自适应估计（FAFE）算法应用到了离散互联系统的故障估计中,设计了分布式故障观测器。由于采用了改进的FAFE算法,在设计过程中可以避免计算原来FAFE算法需要的约束条件,从而解决了系统中带有不确定参数时这种约束条件难以计算的问题。同时改进的FAFE算法可以保持原来算法的精度。基于H∞鲁棒稳定性理论,证明了观测器的故障估计误差和状态估计误差是鲁棒稳定的且满足H∞性能指标。最后,在仿真中考虑了多个故障情形,并通过实例证明了本理论和方法的正确性。

This work studies the fault diagnosis method for discrete interconnected systems with uncertain parameters.The improved fast adaptive fault estimation（FAFE） algorithm is applied to the fault estimation for the discrete interconnected systems,and distributed fault observer is designed.The constraints of the original FAFE algorithm can be avoided in the design process in the improved FAFE algorithm.As a result,the system with uncertain parameters can also be calculated.Besides,the improved FAFE algorithm can guanrentee the accuracy of the original algorithm.Finally,various fault situations are considered in the simulation.