Research on Fault Diagnosis of Surge Arresters Based on Support Vector Recurrent Neural Network

Surge arresters are crucial protective components within electrical power systems, and the proper functioning is vital for the safety and reliability of the entire system. However, due to factors such as prolonged operation and environmental influences, surge arresters can gradually deteriorate, leading to a decline in their performance and reliability. This deterioration can subsequently pose potential risks to the power system. Therefore, accurate diagnosis of the fault condition of surge arresters is of paramount importance for the operation and maintenance of the power system. In this paper, a fault diagnosis model for surge arresters based on Support Vector Recurrent Neural Networks (SVRNN) is proposed and analyzed. This model leverages the strengths of Support Vector Machines (SVM) and Recurrent Neural Networks (RNN) to effectively capture the nonlinear characteristics and temporal dependencies of surge arresters, thereby enabling more precise identification of their fault conditions.