In July 2025, the TRANS Research Group published its latest study, “Scalable and Generalizable Deep Learning for Battery State of Health Estimation in On-Road Electric Vehicles,” in the Journal of Energy Chemistry (IF: 14.9), a Tier 1 journal in the field of energy and fuels classified by the Chinese Academy of Sciences. This work advances the transition of AI algorithms from laboratory research to real-world applications through three core technological breakthroughs:
- Scenario Expansion: A modeling and training framework based on real-world vehicle big data, overcoming the limitations of lab-level battery cell data.
- Algorithm Innovation: A streamlined feature system significantly improves computational efficiency, enabling large-scale deployment.
- Generalization Validation: Demonstrated root mean square error (RMSE) of <0.5% across comprehensive scenarios including passenger/commercial vehicles and LFP/NMC batteries, showcasing industry-leading practical applicability.
This research accelerates the adoption of intelligent and deployable battery health management systems for traction batteries.
The first author of the paper is Hao Jing, a 2024 PhD candidate in the TRANS Group, with Prof. Shiqi Ou serving as the first corresponding author. The second corresponding author is Dr. Jingyuan Zhao from the University of California, Davis. Collaborators include Dr. Jianyao Hu from the China Electronic Product Reliability and Environmental Testing Research Institute (CEPREI) and Zhilong Lv, a research assistant in the TRANS Group. The study received funding from the National Key R&D Program of China and the Department of Science and Technology of Guangdong Province. The successful publication of this work also benefited from the guidance and support of GAC Group.