Accurate prediction of the remaining useful life (RUL) of lithium batteries is essential for ensuring efficient equipment maintenance and energy management, particularly as these batteries serve as a core driver in the new energy technology revolution. While deep learning models such as Convolutional Neural Networks (CNNs), Long Short-Term Memory Networks (LSTMs), and their variants have demonstrated significant success in RUL prediction, they often face challenges related to inadequate modelling of long-term dependencies in complex degradation data. To overcome these limitations, this paper proposes a novel hybrid architecture that integrates the Kolmogorov-Arnold Network (KAN) with an Extended Long Short-Term Memory Network (xLSTM). The KAN component enhances high-dimensional function approximation and improves parameter efficiency by substituting traditional linear weights with B-spline-parameterized univariate functions. Meanwhile, the xLSTM introduces exponential gating mechanisms and covariance update rules to more effectively capture high-order long-term dependencies. Experimental results on the NASA lithium battery aging dataset demonstrate that the proposed KAN-xLSTM model significantly outperforms CNN, LSTM, and xLSTM models in prediction accuracy, particularly for battery groups with large capacity fluctuations.

References

• Ba, J., Hinton, G. E., Mnih, V., Leibo, J. Z., & Ionescu, C. (2016). Using fast weights to attend to the recent past. In Advances in Neural Information Processing Systems (Vol. 29, pp. 4338–4346). Neural Information Processing Systems Foundation, Inc.
• Beck, M., Pöppel, K., Spanring, M., Auer, A., Prudnikova, O., Kopp, M., Klambauer, G., Brandstetter, J., & Hochreiter, S. (2024). xLSTM: Extended long short-term memory. In Advances in Neural Information Processing Systems (Vol. 37, pp. 107547–107603). Neural Information Processing Systems Foundation, Inc.
• Bu, Z., Long, B., Liu, Z., Wu, K., Geng, H., & Cheng, Y. (2025). Multivariate adaptive Brownian motion-particle filter framework for remaining useful life prediction of nonlinear and state-noise coupled degradation process. Reliability Engineering & System Safety, 264,111356. https://doi.org/10.1016/j.ress.2025.111356
• Cai, X., Li, N., & Xie, M. (2024, April). RUL prediction for two-phase degrading systems considering physical damage observations. Reliability Engineering & System Safety, 244,109926. https://doi.org/10.1016/j.ress.2024.109926
• Cui, X., Chen, Z., Lan, J., & Dong, M. (2021). An online state of health estimation method for lithium-ion battery based on ICA and TPA-LSTM. In 2021 IEEE Industrial Electronics and Applications Conference (IEACON) (pp. 130–135). IEEE.
• Cybenko, G. (1989). Approximation by superpositions of a sigmoidal function. Mathematics of Control, Signals and Systems, 2(4), 303–314. https://doi.org/10.1007/BF02551274
• Dayan, P., & Willshaw, D. J. (1991). Optimising synaptic learning rules in linear associative memories. Biological Cybernetics, 65(4), 253–265. https://doi.org/10.1007/BF00206223
• Gao, K., Wu, D., Zhang, S., Peng, R., & Wu, S. (2025). The state-of-the-art development and new challenges: Operations management of metro systems. ICCK Transactions on Systems Safety and Reliability, 1(1), 4–20. https://doi.org/10.62762/TSSR.2025.246708
• Genet, R., & Inzirillo, H. (2025). Tkan: Temporal Kolmogorov-Arnold networks. https://arxiv.org/abs/2405.07344(open in a new window)
• Greff, K., Srivastava, R. K., Koutník, J., Steunebrink, B. R., & Schmidhuber, J. (2016). LSTM: A search space odyssey. IEEE Transactions on Neural Networks and Learning Systems, 28(10), 2222–2232. https://doi.org/10.1109/TNNLS.2016.2582924
• He, W., Williard, N., Osterman, M., & Pecht, M. (2011). Lithium-ion Battery Aging Dataset. Center for Advanced Life Cycle Engineering, University of Maryland.
• Hess, A., Calvello, G., & Frith, P (2005, March). Challenges, issues, and lessons learned chasing the ‘Big P’. Real predictive prognostics. Part 1. In 2005 IEEE Aerospace Conference (pp. 3610–3619). IEEE.
• Hochreiter, S., & Schmidhuber, J. (1997, November). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
• Hornik, K., Stinchcombe, M., & White, H. (1989). Multilayer feedforward networks are universal approximators. Neural Networks, 2(5), 359–366. https://doi.org/10.1016/0893-6080(89)90020-8
• Kolmogorov, A. N (1961). On the representation of continuous functions of several variables by superpositions of continuous functions of a smaller number of variables. In Collected works (pp. 25–46). American Mathematical Society.
• Krotov, D., & Hopfield, J. (2018). Dense associative memory is robust to adversarial inputs. Neural Computation, 30(12), 3151–3167. https://doi.org/10.1162/neco_a_01143
• Krotov, D., & Hopfield, J. J. (2016). Dense associative memory for pattern recognition. In Advances in Neural Information Processing Systems (Vol. 29). Neural Information Processing Systems Foundation, Inc.
• Lai, C., Baraldi, P., & Zio, E. (2025, July). Gradient-enhanced physics-informed long short-term memory networks for stable and accurate prediction of the RUL of electronic components. Reliability Engineering & System Safety, 265,111485. https://doi.org/10.1016/j.ress.2025.111485
• Li, N., Wang, M., Lei, Y., Yang, B., Li, X., & Si, X. (2025). Remaining useful life prediction of lithium-ion battery with nonparametric degradation modeling and incomplete data. Reliability Engineering & System Safety, 256,110721. https://doi.org/10.1016/j.ress.2024.110721
• Li, S., Fang, H., & Shi, B. (2021). Remaining useful life estimation of lithium-ion battery based on interacting multiple model particle filter and support vector regression. Reliability Engineering & System Safety, 210,107542. https://doi.org/10.1016/j.ress.2021.107542
• Li, Y., Xiong, B., Vilathgamuwa, D. M., Wei, Z., Xie, C., & Zou, C. (2021, January). Constrained ensemble Kalman filter for distributed electrochemical state estimation of lithium-ion batteries. IEEE Transactions on Industrial Informatics, 17(1), 240–250. https://doi.org/10.1109/TII.9424
• Liu, J., Hou, B., Lu, M., & Wang, D. (2024). Box-Cox transformation based state-space modeling as a unified prognostic framework for degradation linearization and RUL prediction enhancement. Reliability Engineering & System Safety, 244, 109952. https://doi.org/10.1016/j.ress.2024.109952
• Liu, K., Hu, X., Wei, Z., Li, Y., & Jiang, Y. (2019, December). Modified Gaussian process regression models for cyclic capacity prediction of lithium-ion batteries. IEEE Transactions on Transportation Electrification, 5(4), 1225–1236. https://doi.org/10.1109/TTE.6687316
• Liu, K., Shang, Y., Ouyang, Q., & Widanage, W. D. (2021, April). A data-driven approach with uncertainty quantification for predicting future capacities and remaining useful life of lithium-ion battery. IEEE Transactions on Industrial Electronics, 68(4), 3170–3180. https://doi.org/10.1109/TIE.41
• Liu, Z., Wang, Y., Vaidya, S., Ruehle, F., Halverson, J., Soljačić, M., Hou, T. Y., & Tegmark, M. (2024). Kan: Kolmogorov-Arnold networks. Preprint. arXiv:2404.19756(open in a new window)
• Milakov, M., & Gimelshein, N. (2018). Online normalizer calculation for softmax. Preprint. arXiv:1805.02867(open in a new window)
• Niu, H., Zeng, J., Shi, H., Zhang, X., Liang, J., & Shi, G. (2025). Remaining useful life prediction for multi-component systems with stochastic correlation based on auxiliary particle filter. Reliability Engineering & System Safety, 264,111357. https://doi.org/10.1016/j.ress.2025.111357
• Peiseler, L., Schenker, V., Schatzmann, K., Pfister, S., Wood, V., & Schmidt, T. (2024, May). Carbon footprint distributions of lithium-ion batteries and their materials. Preprint.
• Perumal, T., Mustapha, N., Mohamed, R., & Shiri, F. M. (2024). A comprehensive overview and comparative analysis on deep learning models. Journal on Artificial Intelligence, 6(1), 301–360. https://doi.org/10.32604/jai.2024.054314
• Ramsauer, H., Schäfl, B., Lehner, J., Seidl, P., Widrich, M., Adler, T., Gruber, L., Holzleitner, M., Pavlović, M., Sandve, G. K., Greiff, V., Kreil, D., Kopp, M., Klambauer, G., Brandstetter, J., & Hochreiter, S. (2021). Hopfield networks is all you need. Preprint. https://arxiv.org/abs/2008.02217(open in a new window).
• Ren, L., Zhao, L., Hong, S., Zhao, S., Wang, H., & Zhang, L. (2018). Remaining useful life prediction for lithium-ion battery: A deep learning approach. IEEE Access, 6, 50587–50598. https://doi.org/10.1109/ACCESS.2018.2858856
• Rosenblatt, F. (1958). The perceptron: A probabilistic model for information storage and organization in the brain. Psychological Review, 65(6), 386–408. https://doi.org/10.1037/h0042519
• Saha, B., & Goebel, K. (2007). Battery Data Set. NASA Ames Research Center, NASA Prognostics Data Repository.
• Schlag, I., Irie, K., & Schmidhuber, J. (2021). Linear transformers are secretly fast weight programmers. In International Conference on Machine Learning (pp. 9355–9366). Proceedings of Machine Learning Research.
• Schmidhuber, J. (1992). Learning to control fast-weight memories: An alternative to dynamic recurrent networks. Neural Computation, 4(1), 131–139. https://doi.org/10.1162/neco.1992.4.1.131
• Sejnowski, T. J. (1977). Storing covariance with nonlinearly interacting neurons. Journal of Mathematical Biology, 4(4), 303–321. https://doi.org/10.1007/BF00275079
• Shu, Q., Zhang, F., Shen, L., & Ng, H. K. T. (2024, January). RUL prediction with cross-domain adaptation based on reproducing kernel Hilbert space. IEEE Transactions on Reliability, 74(3), 3871–3883. https://doi.org/10.1109/TR.2024.3488792
• Sun, Y., Dong, L., Huang, S., Ma, S., Xia, Y., Xue, J., Wang, J., & Wei, F. (2023). Retentive network: A successor to transformer for large language models. Preprint. arXiv:2307.08621(open in a new window)
• Vaca-Rubio, C. J., Blanco, L., Pereira, R., & Caus, M. (2024). Kolmogorov-Arnold networks (KANs) for time series analysis. Preprint. arXiv:2405.08790(open in a new window)
• Wang, F. K., Amogne, Z. E., Chou, J. H., & Tseng, C. (2022). Online remaining useful life prediction of lithium-ion batteries using bidirectional long short-term memory with attention mechanism. Energy, 254,124344. https://doi.org/10.1016/j.energy.2022.124344
• Wang, H., Ma, X., & Zhao, Y. (2019). An improved Wiener process model with adaptive drift and diffusion for online remaining useful life prediction. Mechanical Systems and Signal Processing, 127, 370–387. https://doi.org/10.1016/j.ymssp.2019.03.019
• Wang, J., Zhang, F., Ng, H. K. T., & Shi, Y. (2024). Remaining useful life prediction via information enhanced domain adversarial generalization. IEEE Transactions on Reliability, 74(2), 2837–2850. https://doi.org/10.1109/TR.2024.3441592
• Xiang, S., Qin, Y., Luo, J., Pu, H., & Tang, B. (2021). Multicellular LSTM-based deep learning model for aero-engine remaining useful life prediction. Reliability Engineering & System Safety, 216,107927. https://doi.org/10.1016/j.ress.2021.107927
• Xiang, S., Qin, Y., Zhu, C., Wang, Y., & Chen, H. (2020). Long short-term memory neural network with weight amplification and its application into gear remaining useful life prediction. Engineering Applications of Artificial Intelligence, 91,103587. https://doi.org/10.1016/j.engappai.2020.103587
• Xu, A., Fang, G., Zhuang, L., & Gu, C. (2025, September). A multivariate student-t process model for dependent tail-weighted degradation data. IISE Transactions, 57(9), 1071–1087. https://doi.org/10.1080/24725854.2024.2389538
• Zhang, F., Ng, H. K. T., & Shen, L. (2023, January). Robust estimation and selection for degradation modeling with inhomogeneous increments. IEEE Transactions on Reliability, 73, 560–575.
• Zhang, S., Zhai, Q., & Li, Y. (2023). Degradation modeling and RUL prediction with wiener process considering measurable and unobservable external impacts. Reliability Engineering & System Safety, 231,109021. https://doi.org/10.1016/j.ress.2022.109021
• Zhu, R., Hu, J., & Peng, W. (2025). Bayesian calibrated physics-informed neural networks for second-life battery SOH estimation. Reliability Engineering & System Safety, 264,111432. https://doi.org/10.1016/j.ress.2025.111432
• Zhu, R., Peng, W., Ye, Z. S., & Xie, M. (2025, January). Collaborative prognostics of lithium-ion batteries using federated learning with dynamic weighting and attention mechanism. IEEE Transactions on Industrial Electronics, 72(1), 980–991. https://doi.org/10.1109/TIE.2024.3387115
• Zhu, Y., Zhang, F., Chen, W., Cheng, Z., & Shen, L. (2026). Order-preserving kernel contrastive learning with applications to cross-domain RUL prediction. IEEE Transactions on Reliability, 75, 97–111. https://doi.org/10.1109/TR.2025.3635402

To cite this article: Yunhao Hu, Qiuyi Ge, Ziqing Tian, Zuwei Zhang & Fode Zhang (03 Mar 2026): Hybrid Kolmogorov-Arnold and xLSTM network for enhanced RUL prediction of lithium batteries, Statistical Theory and Related Fields, DOI: 10.1080/24754269.2026.2635735

To link to this article: https://doi.org/10.1080/24754269.2026.2635735