The fault identification method of high energy-consuming transformer interturn short circuit considering feature coupling

In view of the extracted fault signal, it is necessary to select suitable features to describe the characteristics of inter-turn short circuit fault. However, relying only on artificial experience for feature selection is subjective and uncertain. Therefore, a fault identification method of high energy-consuming transformer inter-turn short circuit considering feature coupling is proposed. Based on the topology of high energy consuming transformer interturn short circuit, Bayesian inference algorithm and cluster search algorithm are introduced to realize multi-fault feature fusion diagnosis of interturn short circuit. Simulink software is used to build an interturn short circuit model of transformer, and the performance of this method is tested. The experimental results show that the response characteristic curve of the proposed method is very close to the fault curve when the short-circuit fault of turn 1 and turn 2, which indicates that the proposed method can accurately detect the short-circuit fault of the transformer between turns.