An Improved SegRNN-Based Hierarchical Warning Method for Power Battery Thermal Runaway
- 1State Key Laboratory of Light Superalloys, Wuhan University of Technology, Wuhan 430070, China
- 2Hubei Key Laboratory of Advanced Technology for Automotive Components, Wuhan University of Technology, Wuhan 430070, China
- 3SAIC GM Wuling Automobile Co., Ltd., Liuzhou 545007, China
Abstract
Early detection of thermal runaway in electric vehicle batteries remains challenging, and warning lead times during real-world operation are often limited. This paper proposes a hierarchical thermal runaway warning method based on an improved SegRNN model. A cross-variable self-attention module and a one-dimensional convolutional channel fusion layer are integrated into the SegRNN encoder to improve voltage prediction accuracy. One year of operational data from 23 electric vehicles of the same model, including 17 normal vehicles and 6 vehicles with confirmed thermal runaway events, is used for validation. Pearson correlation analysis is applied to select five input variables for charging and discharging conditions separately. A three-level warning strategy is developed based on voltage prediction residuals, with thresholds of 0.05, 0.12, and 0.24 V. On the test set, the proposed model achieves MAE and RMSE values of 0.48624 × 10−3 and 0.64961 × 10−3 during charging, and 3.76869 × 10−3 and 8.2487 × 10−3 during discharging, respectively. Retrospective validation on the six thermal runaway vehicles shows average warning lead times of 87.3 days for Level 1 and 2.3 days for Level 3. These results demonstrate the potential of the proposed method for early thermal runaway risk identification under real-world operating conditions.
Related
- Acoustic Emission to Provide Early Warning Detection of Lithium-Ion Battery Thermal Runaway
- Control Volume Free Element Method for Simulating Thermal Runaway of Lithium-Ion Battery
- Cross-Family Transferability of Multimodal Sensor Information for Short-Horizon Forecasting of Mechanically Induced Lithium-Ion Battery Thermal Runaway
- Cell-Wise Fault Warning for Lithium-Ion Battery Packs Using Local Outlier Factor and Validity-Gated Extreme-Value Thresholds
- Thermal Stability and Gas Evolution of Li-Rich Mn-Based Cathodes and Their Links to Cell-Level Thermal Runaway: A Review
BibTeX
@article{liu2026improved,
title = {{An Improved SegRNN-Based Hierarchical Warning Method for Power Battery Thermal Runaway}},
author = {Jing Liu and Yixuan Zhou and Dunye Liao and Shijie Cai and Yugang Liu and Bowen Yang and Xiao Zhang and Ranjun Huang and Pei Zhang and Zhihong Wang and Min Wei and Jie Hu},
journal = {Batteries},
year = {2026},
month = nov,
doi = {10.3390/batteries12110466},
url = {https://www.mdpi.com/2313-0105/12/11/466},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - An Improved SegRNN-Based Hierarchical Warning Method for Power Battery Thermal Runaway
AU - Jing Liu
AU - Yixuan Zhou
AU - Dunye Liao
AU - Shijie Cai
AU - Yugang Liu
AU - Bowen Yang
AU - Xiao Zhang
AU - Ranjun Huang
AU - Pei Zhang
AU - Zhihong Wang
AU - Min Wei
AU - Jie Hu
PY - 2026
DA - 2026/11/04
JO - Batteries
DO - 10.3390/batteries12110466
UR - https://www.mdpi.com/2313-0105/12/11/466
AB - Early detection of thermal runaway in electric vehicle batteries remains challenging, and warning lead times during real-world operation are often limited. This paper proposes a hierarchical thermal runaway warning method based on an improved SegRNN model. A cross-variable self-attention module and a one-dimensional convolutional channel fusion layer are integrated into the SegRNN encoder to improve voltage prediction accuracy. One year of operational data from 23 electric vehicles of the same model, including 17 normal vehicles and 6 vehicles with confirmed thermal runaway events, is used for validation. Pearson correlation analysis is applied to select five input variables for charging and discharging conditions separately. A three-level warning strategy is developed based on voltage prediction residuals, with thresholds of 0.05, 0.12, and 0.24 V. On the test set, the proposed model achieves MAE and RMSE values of 0.48624 × 10−3 and 0.64961 × 10−3 during charging, and 3.76869 × 10−3 and 8.2487 × 10−3 during discharging, respectively. Retrospective validation on the six thermal runaway vehicles shows average warning lead times of 87.3 days for Level 1 and 2.3 days for Level 3. These results demonstrate the potential of the proposed method for early thermal runaway risk identification under real-world operating conditions.
SN - 2313-0105
ER -