Novel Battery State of Health Estimation and Lifetime Prediction Method Based on the Catboost Model
Chi Zhang, Jichao Hong, Fengwei Liang, Huaqin Zhang, Facheng Wang, Xinyang Zhang, Jingsong Yang, Zhongguo Huang, Kerui Li
Abstract
Battery health and safety estimation is important in electric vehicle (EV) battery system research. In this article, a battery state of health (SOH) estimation method based on the Catboost model is proposed using real vehicle data. A capacity calibration method is proposed by collecting and analyzing the data of one brand of EV for nearly one year. First, sufficient and appropriate charging interval data are selected, and then the selected charging segments are separated to construct a data set, and then the battery SOH is calculated by incremental capacity analysis and the ampere-hour integration method. The battery SOH is evaluated based on the Catboost model, and the error between the actual SOH value and the evaluated value is controlled to be within 0.3 and compared and analyzed with the LSTM model. Finally, a polynomial fitting method is proposed to preliminarily predict the remaining battery life, and a battery aging estimation strategy is proposed. The method plays a role in promoting the SOH research of electric vehicles, which helps electric vehicles develop in a more sustainable and intelligent direction.