Big Data-Driven Thermal Runaway Early Warning Method for Power Battery for Electric Vehicles

Abstract

This study presents a big data-driven thermal runaway early warning method for electric vehicle batteries. The proposed system
integrates advanced machine learning algorithms, including Deep Neural Networks (DNN) and Extreme Gradient Boosting (XGBoost), with
multi-level data fusion to predict potential thermal runaway events. Extensive simulations and real-vehicle tests conducted under diverse operating conditions demonstrate that the method outperforms existing approaches in reducing false alarms while maintaining high sensitivity.
This research contributes to improving electric vehicle reliability and supports the transition towards sustainable transportation. Future work
will focus on long-term real-world studies to further validate the system’s effectiveness and reliability.