Transformer and Long Short-Term Memory-Based Framework with Nonlinear Feature Selection for Simultaneous Prediction of Lithium-Ion Battery State of Health and Remaining Useful Life
Source abstract
Abstract
Precise prediction of a lithium-ion battery’s state of health (SOH) and remaining useful life (RUL) is essential for battery-management systems to schedule operations safely and keep performance optimized throughout its service life. However, existing methods often sacrifice either accuracy or generalizability due to reliance on feature engineering, difficulty capturing long-range degradation, weak SOH-RUL coupling, and manual hyperparameter tuning. To address these issues, we develop a neural architecture that fuses a transformer encoder with a long short-term memory (LSTM) network. First, hierarchical correlation analysis selects nonlinear degradation features from cyclic data. Second, a shared-backbone network with dual output heads simultaneously regresses SOH and RUL, explicitly capturing their coupling. Finally, Bayesian optimization identifies optimal hyperparameters within a constrained computational budget. Experiments on public NASA and Oxford datasets demonstrate that, compared to multilayer perceptron, LSTM networks, and transformers, our method achieves MAE, RMSE, and MAPE reductions of up to 47.1%, 45.9%, and 47.0%, respectively, while consistently maintaining superior performance in cross-dataset validation. These findings offer a promising path toward real-time, high-precision, and transferable intelligent prognostic models for battery lifetime prediction.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Battery Technologies Research · Machine Fault Diagnosis Techniques · Big Data and Digital Economy
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (46)
Show source-provided references
1. A review of data-driven whole-life state of health prediction for lithium-ion batteries: Data preprocessing, aging characteristics, algorithms, and future challenges10.1016/j.jechem.2024.06.017 ↗
2. Hybrid data-driven method with TADBO-SVMD for lithium-ion battery remaining useful life prediction10.1088/1361-6501/addc00 ↗
3. Lithium battery SOH prediction based on frequency-enhanced cross variable for short-term dependency recognition framework10.1088/1361-6501/add48a ↗
4. Prognostics of battery capacity based on charging data and data-driven methods for on-road vehicles10.1016/j.apenergy.2023.120954 ↗
5. State of health prediction of lithium-ion batteries based on machine learning: Advances and perspectives10.1016/j.isci.2021.103265 ↗
6. A review of state of health and remaining useful life estimation methods for lithium-ion battery in electric vehicles: Challenges and recommendations10.1016/j.jclepro.2018.09.065 ↗
7. A comparative study of different features extracted from electrochemical impedance spectroscopy in state of health estimation for lithium-ion batteries10.1016/j.apenergy.2022.119502 ↗
8. A procedure for evaluating the SOH of Li-ion batteries from data during the constant voltage charge phase and the use of an ECM with internal resistance10.1016/j.est.2024.115074 ↗
9. Interpretable machine learning prediction for Li-ion battery’s state of health based on electrochemical impedance spectroscopy and temporal features10.1016/j.electacta.2024.144449 ↗
10. Adaptive piecewise equivalent circuit model With SOC/SOH estimation based on extended Kalman filter10.1109/TEC.2022.3218613 ↗
11. A comprehensive review of battery modeling and state estimation approaches for advanced battery management systems10.1016/j.rser.2020.110015 ↗
12. An encoder-decoder fusion battery life prediction method based on Gaussian process regression and improvement10.1016/j.est.2022.106469 ↗
13. Gaussian process regression for forecasting battery state of health10.1016/j.jpowsour.2017.05.004 ↗
14. Online state-of-health estimation of lithium-ion battery based on relevance vector machine with dynamic integration10.1016/j.asoc.2022.109615 ↗
15. Application of the LSTM-GRU compressed model for battery state of health estimation on smart mobile devices10.1016/j.est.2025.116641 ↗
16. Improved LSTM based state of health estimation using random segments of the charging curves for lithium-ion batteries10.1016/j.est.2023.109370 ↗
17. Jellyfish optimized recurrent neural network for state of health estimation of lithium-ion batteries10.1016/j.eswa.2023.121904 ↗
18. Multi-modal data information alignment based SOH estimation for lithium-ion batteries using a local–global parallel CNN-transformer network10.1016/j.est.2025.117178 ↗
19. A convolutional neural network model for SOH estimation of Li-ion batteries with physical interpretability10.1016/j.ymssp.2022.110004 ↗
20. SOH prediction for lithium-ion batteries by using historical state and future load information with an AM-seq2seq model10.1016/j.apenergy.2023.120793 ↗
21. SOH prediction of lithium-ion batteries using a hybrid model approach integrating single particle model and neural networks10.1016/j.est.2024.114579 ↗
22. SOH and RUL prediction of lithium batteries based on fusions of RLOESS filtered electrochemical and thermal features by bidirectional gated recurrent unit network10.1016/j.est.2024.114134 ↗
23. Rapid health estimation of in-service battery packs based on limited labels and domain adaptation10.1016/j.jechem.2023.10.056 ↗
24. Battery Data Set
25. Oxford battery degradation dataset 110.5287/bodleian ↗
26. Research on state of health prediction model for lithium batteries based on actual diverse data10.1016/j.energy.2021.120851 ↗
27. State of Health (SoH) estimation methods for second life lithium-ion battery—review and challenges10.1016/j.apenergy.2024.123542 ↗
28. Critical summary and perspectives on state-of-health of lithium-ion battery10.1016/j.rser.2023.114077 ↗
29. Battery monitoring and prognostics optimization techniques: challenges and opportunities10.1016/j.energy.2022.124538 ↗
30. Voltage and temperature information ensembled probabilistic battery health evaluation via deep Gaussian mixture density network10.1016/j.est.2023.108587 ↗
31. Electric Vehicles Batteries: Requirements and Challenges10.1016/j.joule.2020.01.013 ↗
32. State-of-health monitoring of lithium-ion battery modules and packs via incremental capacity peak tracking10.1016/j.apenergy.2016.07.126 ↗
33. State-of-health (SOH) evaluation on lithium-ion battery by simulating the voltage relaxation curves10.1016/j.electacta.2019.02.055 ↗
34. State-of-health estimation based on differential temperature for lithium ion batteries10.1109/TPEL.2020.2978493 ↗
35. A novel SOH estimation method with attentional feature fusion considering differential temperature features for lithium-ion batteries10.1109/TIM.2024.3476593 ↗
36. A weighted Pearson correlation coefficient based multi-fault comprehensive diagnosis for battery circuits10.1016/j.est.2022.106584 ↗
37. Linear correlation between state-of-health and incremental state-of-charge in Li-ion batteries and its application to SoH evaluation10.1016/j.electacta.2022.141300 ↗
38. Online state of health estimation of Li-ion polymer batteries using real time impedance measurements10.1016/j.apenergy.2021.118210 ↗
39. State of health estimation for lithium-ion batteries based on incremental capacity analysis and Transformer modeling10.1016/j.asoc.2024.112072 ↗
40. Neural machine translation by jointly learning to align and translate
41. Layer normalization
42. An end-to-end neural network framework for state-of-health estimation and remaining useful life prediction of electric vehicle lithium batteries10.1016/j.rser.2021.111843 ↗
43. Long short-term memory10.1162/neco.1997.9.8.1735 ↗
44. Reliability enhancement of state of health assessment model of lithium-ion battery considering the uncertainty with quantile distribution of deep features10.1016/j.ress.2024.110002 ↗
45. Hyperparameter optimization for machine learning models based on bayesian optimizationb10.11989/JEST.1674-862X.80904120 ↗
46. Neural ordinary differential equations and recurrent neural networks for predicting the state of health of batteries10.1016/j.est.2022.104209 ↗