Fault Diagnosis of Smart Substation Secondary System Based on ANN

This paper presents an artificial neural network (ANN)-based fault diagnosis method for the secondary system in smart substations. The aim of this study is to enhance the efficiency of the operation and maintenance of the substation. The proposed method involves sorting out the reasoning knowledge base for fault diagnosis from plug-in faults of protection devices. The method also proposes a representation technique for fault feature information based on the publish/subscribe relationship of packets and the flow status of switches. An artificial neural network-based fault diagnosis model is then built, and the real-time diagnosis steps are provided. The proposed fault diagnosis method is verified using a typical 110kV smart substation protection system, and the accuracy of the method in the face of various fault types is demonstrated. The study also shows the superiority of this method in feature selection and recognition accuracy compared to existing methods.