Adaptive Unknown Input Observer for Actuator Fault Detection and Isolation of PEM Fuel Cell Systems

Abstract

AbstractFor detection and isolation of stuck actuator fault, an adaptive unknown input observer (AUIO) for the PEMFC systems is developed. A group of observers is built in a multiple-model scheme. These rely on the system model in the case of a specific fault in the actuator. Based on the evolved AUIO, these are designed to produce fault-dependent residuals. The residual is zero if the model and system match. Alternatively, this is certainly nonzero, and the defective signal is regulated uniquely. Numerical results show the success of the proposed AUIO and the identification and isolation algorithm for faults. This adaptive UIO-based technique can be further used and extended to multiple actuator fault cases in the PEMFC system and is easier to implement than other existing methods. One of its advantages is that it satisfies the existence conditions of normal UIO. This is also very useful for the fault diagnosis and fault-tolerant control (FTC) of the PEMFC systems.KeywordsPEMFCAUIOFDILMI