Research on Fault Diagnosis Method of Electric Vehicle Dynamic Wireless Power Transfer

In recent years, with the gradual popularization and application of WPT technology, the demand for timely diagnosis of WPT system faults has gradually emerged. The WPT system is a complex high-order nonlinear system, and the fault diagnosis is difficult. At present, there is no mature fault diagnosis scheme for WPT system. Therefore, the WPT system urgently needs a reliable fault diagnosis scheme, which is of great significance to improve the reliability of the system. In this paper, based on the electric vehicle dynamic wireless power transfer (EV DWPT) system, aiming at the problem of soft fault identification of main circuit device, the energy entropy and average value of wavelet coefficients, the effective value of current and waveform index are obtained by wavelet transform as feature vectors, and the support vector machine (SVM) algorithm is used for fault diagnosis classification. Firstly, the system fault state table is established, and the system model is built in the simulink simulation platform, and the current waveform of the main circuit soft fault is simulated respectively. The results show that the method can better identify the soft fault of the main circuit device, and the diagnosis rate can reach up to 98%.