An Abnormal Electrical Phenomena Identification Method for Vehicle-grid Electrical Coupling System

The timely and accurate identification of abnormal electrical phenomena (AEP) for vehicle-grid coupling system(VGECS) is critical to guarantee the safe and stable operation of vehicles. The topology of VGECS changes dynamically, hence the types of AEP are complex and varied and the VGECS contains not only the single type AEP, but also the type of compound AEP. At present, there are few identification methods for AEP of VGECS, especially for the compound type AEP. In view of this, a method based on wavelet packet transform and extreme gradient boosting(XGBoost) to identify the AEP of VGECS is proposed in this paper. This method not only identifies the single type AEP, but also enables the identification of compound type AEP. The results show that the proposed method has a high recognition accuracy for AEP. At the same time, the method has good real-time performance, which can meet the practical engineering requirements.