Fault detection and assessment for solid oxide fuel cell system gas supply unit based on novel principal component analysis

Abstract After long-term running of solid oxide fuel cell system, the performance degradation of solid oxide fuel cell stack is caused by the fault of gas supply equipment. The more serious problem is that the stack performance does not match external load requirement. In order to guarantee the normal discharge of the stack, the fault detection of the system balance of plant components is a prerequisite for maintaining stack normal state, and is an important guarantee for increasing its operating cycle and remain useful lifetime. This paper proposes a novel statistical method based on principal component analysis and exponentially weighted moving average control chart for solid oxide fuel cell system gas supply fault detection. The gas supply fault is derived from the reformer and heat exchangers of the real system. The fault detection is carried out with collected data of real system and novel detection method. The results show that the novel method has higher efficiency and better effect than the traditional method under the system gas supply fault. Multi-times experimental data of real system is used to verify the robustness of the novel methods on-site fault detection.