journal article·2026-11Open access

Electrochemically constrained transformer with steady-state probability for lithium-ion battery risk evaluation

Guanhao Li, Yong Gao, Duanfeng Chu, Ao Chi, Qingsong Jiang, Juntian Wang, Wen Fu

Abstract

Accurate health assessment and early risk warning are essential for ensuring the safety and reliability of lithium-ion batteries. However, the complexity and uncertainty of real-world operating conditions hinder anomaly identification, physically consistent state prediction, and dynamic risk evaluation. To address these challenges, this study proposes a mechanism–data fusion framework integrating steady-state probabilistic analysis, electrochemically constrained Transformer model, and Bayesian dynamic inference. The framework employs a steady-state probability model to identify abnormal intervals and critical cells under unsupervised conditions, while a Transformer model constrained by a single-particle electrochemical model is developed to improve prediction accuracy and electrochemical consistency. Building upon anomaly indicators and prediction residuals, a multi-scale dynamic risk evaluation framework is established to characterize risk evolution and quantify system-level risk. Validation using real-world vehicle operation data demonstrates that the proposed framework achieves 92.3% accuracy in abnormal-cell identification and significantly outperforms benchmark models in battery state prediction, with a maximum error reduction of 67.3%. Moreover, the framework provides an early warning approximately 100 s before the onset of abnormal voltage fluctuations. This work highlights the potential of integrating electrochemical mechanisms with data-driven intelligence for battery safety management and offers a practical solution for next-generation intelligent battery monitoring systems.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Advanced Battery Technologies Research · Advancements in Battery Materials · Advanced battery technologies research

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (46)

Show source-provided references
1. Environmental impacts and supply risks for LiFePO4 – LiCoxNiyMn1-x-yO2 hybrid battery pack in China10.1016/j.psep.2025.107115 ↗
2. Prospective environmental and resource sustainability assessment of advanced matching technologies for sodium-ion and lithium-ion batteries10.1016/j.energy.2026.140679 ↗
3. Battery lifetime prediction using surface temperature features from early cycle data10.1039/D4EE05179C ↗
4. Mapping, understanding and reducing belief in misinformation about electric vehicles10.1038/s41560-025-01790-0 ↗
5. Toward advanced estimation of state of health for integral lithium-ion battery pack
6. Increasing generalization capability of battery health estimation using continual learning
7. A review of battery energy storage systems and advanced battery management system for different applications: challenges and recommendations10.1016/j.est.2024.111179 ↗
8. Progress and challenges in ultrasonic technology for state estimation and defect detection of lithium-ion batteries
9. A new SOH estimation method for lithium-ion batteries based on model-data-fusion10.1016/j.energy.2023.129597 ↗
10. Hybrid deep neural network with dimension attention for state-of-health estimation of lithium-ion batteries10.1016/j.energy.2023.127734 ↗
11. Deep learning to estimate lithium-ion battery state of health without additional degradation experiments
12. Health prognostics for lithium-ion batteries: mechanisms, methods, and prospects10.1039/D2EE03019E ↗
13. Data-driven prediction of battery cycle life before capacity degradation10.1038/s41560-019-0356-8 ↗
14. Machine learning pipeline for battery state-of-health estimation10.1038/s42256-021-00312-3 ↗
15. Physics-informed machine learning10.1038/s42254-021-00314-5 ↗
16. Data-driven state-of-charge estimation of a lithium-ion battery pack in electric vehicles based on real-world driving data10.1016/j.est.2024.113986 ↗
17. Deep learning framework for lithium-ion battery state of charge estimation: recent advances and future perspectives
18. Deep learning for time series Forecasting: tutorial and literature survey10.1145/3533382 ↗
19. Multi- forword-step state of charge prediction for real-world electric vehicles battery systems using a novel LSTM-GRU hybrid neural network10.1016/j.etran.2024.100322 ↗
20. A novel state of health prediction method for battery system in real-world vehicles based on gated recurrent unit neural networks10.1016/j.energy.2023.129918 ↗
21. Mitigating “remaining fire”-“Re-burn”: Multi-dimensional dynamic thermal runaway evolution mechanism and suppression for commercial lithium-ion batteries from gas perspective
22. Wavelet packet energy proportion-based early warning for the failure of lithium-ion batteries10.1109/TTE.2024.3417949 ↗
23. Battery abnormality detection based on important sample selection for large-scale vehicle monitoring10.1016/j.apenergy.2025.125667 ↗
24. A comprehensive review of lithium-ion battery modelling research and prospects: in-depth analysis of current research and future directions10.1016/j.apenergy.2025.126688 ↗
25. Real-world cross-battery state of charge prediction in electric vehicles with machine learning: data quality analysis, data repair and training data reconstruction10.1016/j.energy.2025.138322 ↗
26. MR-Transformer: multiresolution transformer for multivariate time series prediction10.1109/TNNLS.2023.3327416 ↗
27. PMformer: a novel informer-based model for accurate long-term time series prediction10.1016/j.ins.2024.121586 ↗
28. Enhancing time series forecasting: a hierarchical transformer with probabilistic decomposition representation10.1016/j.ins.2023.119410 ↗
29. Battery fault diagnosis and failure prognosis for electric vehicles using spatio-temporal transformer networks10.1016/j.apenergy.2023.121949 ↗
30. Battery SOC estimation with physics-constrained BiLSTM under different external pressures and temperatures10.1016/j.est.2025.116205 ↗
31. Deep learning approach towards accurate state of charge estimation for lithium-ion batteries using self-supervised transformer model10.1038/s41598-021-98915-8 ↗
32. Specialized convolutional transformer networks for estimating battery health via transfer learning
33. Multivariate prognosis of battery advanced state of health via transformers
34. Attention towards chemistry agnostic and explainable battery lifetime prediction10.1038/s41524-024-01286-7 ↗
35. Robust diagnosis of capacity and SOC consistency in battery pack based on OCV reconstruction in real-time battery management System10.1109/TTE.2024.3489962 ↗
36. Leveraging multi-view imputation strategy for robust battery lifetime prediction under missing-data scenarios
37. Battery state-of-charge estimation amid dynamic usage with physics-informed deep learning10.1016/j.ensm.2022.06.007 ↗
38. Physics-informed neural network for lithium-ion battery degradation stable modeling and prognosis
39. Attention-based BILSTM for the degradation trend prediction of lithium battery10.1016/j.egyr.2023.03.056 ↗
40. Predicting abrupt depletion of dissolved oxygen in Chaohu lake using CNN-BiLSTM with improved attention mechanism10.1016/j.watres.2024.122027 ↗
41. SOC estimation for lithium batteries using a CNN-attention-LSTM model10.1016/j.est.2025.117479 ↗
42. Fault prognosis of battery system based on accurate voltage abnormity prognosis using long short-term memory neural networks10.1016/j.apenergy.2019.113381 ↗
43. Equivalent circuit modeling of sodium-ion batteries10.1016/j.est.2021.103233 ↗
44. State of charge estimation for lithium-ion battery using Transformer with immersion and invariance adaptive observer10.1016/j.est.2021.103768 ↗
45. Online State of health estimation of lithium-ion batteries based on charging process and long short-term memory recurrent neural network10.3390/batteries9020094 ↗
46. State of health estimation for lithium-ion batteries based on healthy features and long short-term memory10.1109/ACCESS.2020.2972344 ↗