A fault diagnosis method based on feature engineering and TranFD for battery packs

With the rapid development of electric vehicle industry, the safety of power battery pack has been paid more and more attention. High precision battery pack fault diagnosis can improve the safety of electric vehicles and reduce their operation and maintenance costs. In this research, a fault diagnosis method based on feature engineering and Transformer model is proposed for battery packs. We designed a feature engineering for voltage data to improve the learning ability of the model for fault features. We use the optimized rectangular convolution kernel to design the Transformer for fault diagnosis (TranFD). We use UDDS, FUDS and US06 working conditions to construct fault data sets including sensor faults to test models. The results show that our diagnostic method has excellent accuracy and high adaptability.