Toward safe carbon–neutral transportation: Battery internal short circuit diagnosis based on cloud data for electric vehicles

Abstract

To promote electrified transportation and achieve carbon neutrality, Li-ion batteries with excellent energy storage performance are widely adopted in electric vehicles (EVs). However, internal short circuit (ISC) of batteries is a serious safety hazard of EVs. ISC diagnosis using the powerful storage and computing capabilities of the cloud platform is a promising method to enhance the safety performance of battery packs in EVs, but the recording period of cloud data is larger than that of battery management systems in EVs, which brings great challenges to the existing ISC diagnosis methods. In this paper, the new mean-normalization is employed to amplify the voltage characteristics of the ISC cell in the battery pack. The adaptive Kalman filter algorithm is deployed to filter the mean-normalization values to obtain the vivid ISC features. Furthermore, the density-based spatial clustering of applications with noise algorithm is exploited to automatically detect and locate the ISC cell. Even with the 40 s sparse data in the real vehicle, the ISC cell also can be timely diagnosed within 2.51 h.