Solid oxide fuel cell systems fault diagnosis: Critical summarization, classification, and perspectives

Global interests in solid oxide fuel cells (SOFCs) has been dramatically increased over the last few decades thanks to their high energy conversion efficiency, prominent stability, and powerful sustainability. In particular, fault diagnosis of SOFC systems plays a crucial role in their overall optimal control and reliable operation. Thus far, a large number of fault diagnosis methods have been developed to improve the durability and promote the commercialization of SOFC systems. Until now, several reviews have been proposed to summarize these methods but all exist distinct drawbacks, such as incomplete classification and discussion, lack of practical recommendations, etc. Hence, this paper attempts to offer a systematic summarization and classification of various SOFC systems fault diagnosis approaches. Note that eighteen specific faults are addressed detailedly, along with fifteen fault detection and isolation (FDI) methods which are categorized into four groups, i.e., electrochemical impedance spectroscopy (EIS)-based, model-based, data-driven, and hybrid FDI. Besides, practical applications, diagnosis accuracy, diagnostic diversity, complexity, advantages, and disadvantages of all methods are thoroughly compared and evaluated. Lastly, eleven perspectives are proposed for future in-depth researches.