Diagnosis of Gas Leakage in SOFC Stack Based on Gas Sensing Data

Monitoring and fault diagnosis playa major role in improving the performance and reliability of Solid Oxide Fuel Cell (SOFC) systems. However, current SOFC-related studies do not take into account the variation of gas concentration, leaving space for SOFC performance enhancement. In this paper, a model-based fault diagnosis method for gas leakage of SOFC stack is proposed based on gas sensing data. Compared to previous models, the fault model based on gas concentration is simpler, which facilitates for real-time online operation. The state values of the stack are defined in fault model to reflect the state of the stack and the adaptive thresholds are designed to eliminate the effects of data noise. Then, the gas leakage of SOFC stack can be diagnosed based on the relationship between the stack state values and the adaptive thresholds. The simulation results show that the method can diagnose the gas leakage fault of SOFC stack very well and the design of adaptive threshold can improve the robustness of the system and reduce the occurrence of misclassification.