Assessment of Power Battery Health in New Energy Vehicles Through Driving Data Analytics

With the rapid development of modern society’s economy and the increase in the number of automobiles, issues of environmental pollution and energy shortages have become increasingly severe. Each country has started to vigorously promote the new energy vehicle industry. As the power source of the vehicle, the power battery is also the core component of electric vehicles. Accurate estimation of the state of the vehicle’s power battery can effectively enhance the operational safety of the entire vehicle and adjust the battery’s operation strategy, which is significant for delaying the performance degradation of the power battery. Estimating the state of health (SOH) of the power battery is particularly crucial. This paper proposes an estimation method for new energy vehicle batteries based on real vehicle operation data, including battery label capacity extraction, feature engineering selection, and neural network prediction based on TCN-LSTM. The predictive accuracy of this method has been enhanced in comparison to the single-method approach.