Fuel cell system fault diagnosis method based on fault severity

摘要

According to the present invention, a method for diagnosing malfunction in a fuel cell system based on the severity of malfunction can preferentially diagnose the malfunction with higher severity by designing and learning a plurality of artificial neural networks to diagnose a plurality of malfunction groups with different severity respectively, and thus the prompt follow-up is possible. In addition, by using the plurality of artificial neural networks each having different sensitivity to diagnose the plurality of malfunction groups with different severity, the malfunction can be diagnosed through an artificial neural network optimized for each malfunction group, thereby enabling more accurate diagnosis.