Research on Knowledge Graph-based Fault Diagnosis Method for Substation Equipment Operation

Abstract The current image processing-based fault diagnosis method for substation equipment operation realizes fault diagnosis by monitoring the thermal status of the equipment, which leads to low diagnostic accuracy due to the lack of analysis of alarm signals. In this regard, the research of substation equipment operation fault diagnosis method based on knowledge mapping is proposed. A fault diagnosis model is established by building a behavior map of substation accident-like alarm signals, analyzing the alarm behavior as well as the causes of faults, and predicting the posterior probability of faults. In the experiment, the diagnosis accuracy of the proposed method is verified. The analysis of the experimental results shows that the fault diagnosis model constructed by the proposed method has good diagnostic performance and high diagnostic accuracy.