A hybrid deep learning model for robust and data-efficient lithium-ion battery remaining useful life prediction

Abstract

Lithium-ion batteries are extensively utilized in applications like new energy vehicles and aerospace owing to their high energy density and safety, but their service life diminishes due to irreversible capacity degradation from repeated charge-discharge cycles, making accurate remaining useful life (RUL) prediction critical for reliability and operational safety. Current data-driven methods often struggle with long-range dependencies, noise from capacity regeneration, and efficient data utilization. To address these challenges, this study introduces a novel hybrid neural network architecture that integrates complementary ensemble empirical mode decomposition with adaptive noise (CEEMDAN) for denoising preprocessing with a Transformer-convolutional neural network (CNN)-bidirectional gated recurrent unit (BiGRU) model. The framework employs health indicators extracted from voltage and current profiles as inputs, where CEEMDAN mitigates interference effects, the Transformer captures global degradation trends via self-attention mechanisms, the CNN extracts localized short-term features, and the bidirectional GRU models temporal dependencies bidirectionally. Experimental validation on National Aeronautics and Space Administration (NASA) and our own test datasets demonstrates that the proposed approach significantly outperforms other models in key metrics such as mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE), achieving high accuracy even with minimal training data (e.g., only 40% of cycles). Furthermore, cross-dataset validation demonstrates robust generalization of our model, achieving MAPE below 3.5% when transferring between NASA and our battery data without retraining. This hybrid model offers a robust, data-efficient solution for enhancing RUL prediction in practical battery management systems, with strong generalization across diverse battery types.

Keywords

Lithium-ion battery / deep learning model / RUL prediction / CEEMDAN denoising / BiGRU / Transformer

References

[1] Goodenough JB.Challenges for rechargeable Li batteries.Chem Mater2010;22:587-603

[2] Huang J,Tarascon J.Sensing as the key to battery lifetime and sustainability.Nat Sustain2022;5:194-204

[3] Dembitskiy AD,Eremin RA,Fedotov SS.Benchmarking machine learning models for predicting lithium ion migration.npj Comput Mater2025;11:1571

[4] Attia PM,Jin N.Closed-loop optimization of fast-charging protocols for batteries with machine learning.Nature2020;578:397-402

[5] Downie LE,Wright ATB.Determination of the time dependent parasitic heat flow in lithium ion cells using isothermal microcalorimetry.J Phys Chem C2014;118:29533-41

[6] Severson KA,Jin N.Data-driven prediction of battery cycle life before capacity degradation.Nat Energy2019;4:383-91

[7] Tian H,Li K.A review of the state of health for lithium-ion batteries: research status and suggestions.J Clean Prod2020;261:120813

[8] Lin C,Yang F,Tsui K.Battery state of health modeling and remaining useful life prediction through time series model.Appl Energy2020;275:115338

[9] Shu X,Wei K.Research on capacity characteristics and prediction method of electric vehicle lithium-ion batteries under time-varying operating conditions.J Energy Storage2023;58:106334

[10] Wang Y.Application-oriented design of machine learning paradigms for battery science.npj Comput Mater2025;11:1575

[11] Yang L,Yang Y.Supervisory long-term prediction of state of available power for lithium-ion batteries in electric vehicles.Appl Energy2020;257:114006

[12] Ramadesigan V,De S,Braatz RD.Modeling and simulation of lithium-ion batteries from a systems engineering perspective.J Electrochem Soc2012;159:R31-45

[13] Deng Z,Deng H,Li D.Polynomial approximation pseudo-two-dimensional battery model for online application in embedded battery management system.Energy2018;142:838-50

[14] Karimi D,Van Mierlo J.Equivalent circuit model for high-power lithium-ion batteries under high current rates, wide temperature range, and various state of charges.Batteries2023;9:101

[15] Larsson C,Xu J,Asp LE.Electro-chemo-mechanical modelling of structural battery composite full cells.npj Comput Mater2025;11:1646

[16] Wang F,Zhao Z,Chen X.Physics-informed neural network for lithium-ion battery degradation stable modeling and prognosis.Nat Commun2024;15:4332 PMCID:PMC11109204

[17] E L,Yang R,Li H.A physics-informed neural network-based method for predicting degradation trajectories and remaining useful life of supercapacitors.Green Energy Intell Transp.2025;4:100291

[18] E L,Sun Y,Xiong R.Physics-informed hybrid neural architecture for coupled degradation modeling and remaining useful life prediction of LiFePO₄ batteries.J Electrochem Soc2025;172:060505

[19] Wen P,Li Y,Xie P.Physics-informed neural networks for prognostics and health management of lithium-ion batteries.IEEE Trans Intell Veh2024;9:2276-89

[20] Lv K,Bao C.Indirect prediction of lithium-ion battery RUL based on CEEMDAN and CNN-BiGRU.Energies2024;17:1704

[21] Xu Q,Khoo E,Li X.A hybrid ensemble deep learning approach for early prediction of battery remaining useful life.IEEE/CAA J Autom Sinica2023;10:177-87

[22] Zheng D,Guo X,Wei R.Research on the remaining useful life prediction method for lithium-ion batteries based on feature engineering and CNN-BiGRU-AM model.Ionics2025;31:5717-36

[23] Wang C,Yang X.RUL prediction of rolling bearings based on a DCAE and CNN.Appl Sci2021;11:11516

[24] Guo L,Jia F,Lin J.A recurrent neural network based health indicator for remaining useful life prediction of bearings.Neurocomputing2017;240:98-109

[25] Zhang H,Shao S,Yang X.Attention-based LSTM network for rotatory machine remaining useful life prediction.IEEE Access2020;8:132188-99

[26] Zeng F,Jiang Y.A deep attention residual neural network-based remaining useful life prediction of machinery.Measurement2021;181:109642

[27] Tan R,Cheng M,Huang J.Forecasting battery degradation trajectory under domain shift with domain generalization.Energy Storage Mater2024;72:103725

[28] Wei M,Wang Q,Twajamahoro JP.Remaining useful life prediction of lithium-ion batteries based on stacked autoencoder and gaussian mixture regression.J Energy Storage2022;47:103558

[29] Xing J,Zhang J.Remaining useful life prediction of - lithium batteries based on principal component analysis and improved Gaussian process regression.Int J Electrochem Sci.2023;18:100048

[30] Li H,Zhang Y.Remaining useful life prediction using multi-scale deep convolutional neural network.Appl Soft Comput2020;89:106113

[31] Wu L,Suzana AF.Three-dimensional coherent X-ray diffraction imaging via deep convolutional neural networks.npj Comput Mater2021;7:644

[32] Zhou J,Chen D,Qian Q.Remaining useful life prediction of bearings by a new reinforced memory GRU network.Adv Eng Inform2022;53:101682

[33] Gao K,Lyu C.Multi-scale prediction of remaining useful life of lithium-ion batteries based on variational mode decomposition and integrated machine learning.J Energy Storage2024;99:113372

[34] Ardeshiri R, Ma C. Multivariate gated recurrent unit for battery remaining useful life prediction: a deep learning approach.Int J Energy Res2021;45:16633-48

[35] Wang Z,Chen C.Adaptive self-attention LSTM for RUL prediction of lithium-ion batteries.Inf Sci2023;635:398-413

[36] Wang F,Chou J.Online remaining useful life prediction of lithium-ion batteries using bidirectional long short-term memory with attention mechanism.Energy2022;254:124344

[37] Tian Y,Yang Y,Zeng J.State-of-health prediction of lithium-ion batteries based on CNN-BiLSTM-AM.Batteries2022;8:155

[38] Wang, T.; Wang, K.; Pan, K. A Transformer-CNN-BiLSTM architecture for SOH prediction of battery storage systems. In 2024 IEEE 8th Conference on Energy Internet and Energy System Integration (EI2), Shenyang, China, November 29-December 2, 2024; IEEE: New York, USA, 2024; pp 4591-6.

[39] Jin Y,Chen Y.A Time Series Transformer based method for the rotating machinery fault diagnosis.Neurocomputing2022;494:379-95

[40] Zhang Q,Zhang Y,Zhang C.Transformer-based attention network for stock movement prediction.Expert Syst Appl2022;202:117239

[41] Hu W.Remaining useful life prediction of lithium-ion batteries based on wavelet denoising and transformer neural network.Front Energy Res2022;10:969168

[42] Huang X,Shen Y.Long time series of ocean wave prediction based on PatchTST model.Ocean Eng2024;301:117572

[43] Tang P,Yao Z.Lithium-ion battery RUL prediction based on optimized VMD-SSA-PatchTST algorithm.Sci Rep2025;15:26824 PMCID:PMC12287278

[44] Li X,Wang Z.Remaining useful life prediction for lithium-ion batteries based on a hybrid model combining the long short-term memory and Elman neural networks.J Energy Storage2019;21:510-8

[45] Jiao R,Dong J.Remaining useful life prediction of lithium-ion batteries based on conditional variational autoencoders-particle filter.IEEE Trans Instrum Meas2020;69:8831-43

[46] Ahwiadi M.An enhanced mutated particle filter technique for system state estimation and battery life prediction.IEEE Trans Instrum Meas2019;68:923-35

[47] Rahimifard S,Habibi S.Interacting multiple model strategy for electric vehicle batteries state of charge/health/power estimation.IEEE Access2021;9:109875-88