ISSN 1008-5548

CN 37-1316/TU

最新出版

人工智能在锂离子电池研究中的应用

Applications of artificial intelligence in lithium‑ion batteries research

陈亚楠1, 汪玉洁2

1.天津大学 材料科学与工程学院, 天津 300072; 2.保山学院 资源与环境学院, 云南 保山 678000


引用格式:

陈亚楠, 汪玉洁. 人工智能在锂离子电池研究中的应用[J]. 中国粉体技术, 2026, 32(3): 1-13.

Citation:CHEN Yanan, WANG Yujie. Applications of artificial intelligence in lithium‑ion batteries research[J]. China Powder Science and Technology, 2026, 32(3): 1-13.

DOI:10.13732/j.issn.1008-5548.2026.03.006

收稿日期: 2024-12-31, 修回日期: 2025-04-11,上线日期: 2025-08-24。

基金项目: 国家自然科学基金项目, 编号: 92372107; 国家重大基础研究项目, 编号: 52171219。

第一作者简介: 陈亚楠(1990—),男,研究员,博士,博士生导师,中国科协青年托举人才,研究方向为高通量筛选与数据获取。E-mail:yananchen@tju.edu.cn。

通信作者简介:汪玉洁(1986—),男,教授,博士,研究方向为功能材料合成。E-mail:wangyujieufo@163.com。

摘要: 【目的】 为了开发高性能锂离子电池,针对人工智能技术在锂离子电池研究中的应用成果开展综述分析,有助于推动锂离子电池技术的创新与发展。【研究现状】 综述人工智能技术在锂离子电池领域中的应用研究,概括机器学习算法使用模型及其对锂离子电池性能、寿命提升作用,总结人工智能技术在锂离子电池材料性质预测和开发、状态预测、制造工艺、监测和运行管理等方面的应用。【结论与展望】提出人工智能显著提升锂离子电池生产制造质量和效率,缩短锂离子电池研发周期,但是在数据、模型、计算资源和实际应用等方面存在不足;认为未来的研究方向应聚焦于数据高质量获取、共享和标准化,模型可解释性和通用性提升,与其他技术融合强化,新型电池体系中的推广应用,跨学科合作发展。

关键词: 人工智能; 锂离子电池; 电池性能; 材料开发; 电池寿命监测


Abstract

Significance To develop high-performance lithium-ion batteries, this paper reviews and analyzes the applications of artificial intelligence(AI) technologies in lithium-ion batteries,promoting the innovation and development of lithium-ion battery technolo-gies.

Progress This work provides a comprehensive overview of AI technologies used in lithium-ion batteries (LIBs),covering descriptive, predictive, and prescriptive applications, as well as their core machine learning (ML) components,including artificial neural networks (ANNs), support vector machines (SVMs), logistic regression, partial least squares (PLS) regression, and random forest (RF) algorithms. First, by exploring hidden statistical patterns within high-dimensional data, machine learning can efficiently and accurately obtain the composition-structure-performance relationships in LIB materials.This capability guarantees the batteries’ operational safety and facilitates efficient battery management. The crystal structure of the electrode material determines the intrinsic physical and chemical properties of batteries.Based on this structure-property relationship, feature engineering in ML can identify key conditional attributes, which are then used to train models that establish relationships between these conditional factors and decision attributes.The trained models can subsequently predict battery properties such as voltage, capacity, and ionic conductivity. The selection of appropriate descriptors and models isalso a key step in achieving the accurate prediction of material properties.However, the relationships between selected descriptors and target properties are typically complex and nonlinear. Therefore, substantial datasets are required to effectively train the ML models. In the reverse design approach, where material properties serve as the inputs and structure and composition as the outputs, success depends on three factors: 1) identifying descriptors closely related to target properties of the material,2) establishing accurate models between descriptors and target properties,and 3) experimental synthesis and validation. Although AI shows great potential for reverse material design,several challenges remain,including the mismatch between experimental validation cycles and algorithm iteration speeds, limited descriptor universality for complex systems, and the lack of closed-loop data feedback mechanisms. These limitations indicate that closed-loop reverse material design still has a long way to go. Furthermore, accurate prediction of battery parameters, such as capacity, state of health (SOH), service life, and remaining charge,is crucial for improving battery performance, extending lifespan, and reducing costs. ML models such as SVMs and RF algorithms in AI approaches overcome the limitations of traditional empirical and statistical prediction methods.These AI-based approaches are able to accurately cope with nonlinear conditions arising from complex working scenarios and battery aging effects.Furthermore, through deep learning techniques, models such as convolutional neural networks (CNNs)can effectively extract and utilize time-series features for improved predictions. Moreover, LIB manufacturing involves complex processes covering material preparation, electrode manufacturing, battery assembly, and formation testing. Traditional production methods often rely on manual operations and static controls, leading to inefficiency and product inconsistency. The introduction of AI technologies has revolutionized lithium battery manufacturing, emerging as a key driving force for higher production efficiency, lower costs, and better product quality. However, successful implementation of AI-assisted industrial manufacturing still requires continued advancements in terms of sensor development, digital-physical world integration infrastructure, and accurate battery health state estimation. Additionally, accurate real-time monitoring of battery parameters, such as temperature, charge/discharge rates, cycle count, and environmental conditions,is critical for ensuring battery safety, extending lifespan, and optimizing performance. Traditional monitoring methods often rely on one-dimensional measurements of voltage, current, and temperature, making it difficult to capture the complex dynamic processes within batteries. AI technologies bring new solutions through powerful data processing, machine learning, and prediction. Despite these benefits, challenges remain in terms of data quality, model interpretability, computational resources, and practical applications. Finally, charge or discharge control and thermal management during battery operation directly impact its efficiency, safety, lifespan, and overall performance. Traditional methods based on fixed rules and simple algorithms cannot address the various complexities and uncertainties in battery operation. AI technologies,with their excellent data processing, machine learning, and optimization, enable more sophisticated battery system management. As AI technologies and battery management systems continue to develop, future research directions may focus on intelligent control strategies, refined management,and personalized battery management solutions.

Conclusions and Prospects AI has demonstrated significant potential to upgrade manufacturing quality and efficiency while accelerating the research and development of LIB technologies. However, persistent challenges remain in four critical domains,i.e., data quality and availability, model performance, computational resource, and practical implementation. Future research endeavors should focus on several key priority areas:acquiring, sharing, and standardizing high-quality datasets, improving model interpretability and generalizability, integrating and enhancing other technologies,expanding AI applications to new battery systems, and fostering interdisciplinary cooperation and development.

Keywords: artificial intelligence; lithium-ion battery; battery performance;material development; battery life monitoring


参考文献(References)

[1]NIRI M F, ASLANSEFAT K, HAGHI S, et al. A review of the applications of explainable machine learning for lithium-ion batteries: from production to state and performance estimation[J].Energies,2023,16(17): 6360.

[2]LOMBARDO T, DUQUESNOY M, EL-BOUYSIDY H, et al. Artificial intelligence applied to battery research: hype or reality?[J].Chemical Reviews, 2022, 122(12): 10899-10969.

[3]ZHU Z W, QIU J Y, WANG L, et al. Application of artificial intelligence to lithium-ion battery research and development[J].Journal of Electrochemistry, 2022, 28(12): 2219003.

[4]ETACHERI V, MAROM R, ELAZARI R, et al.Challenges in the development of advanced Li-ion batteries: a review[J].Energy & Environmental Science, 2011, 4(9): 3243-3262.

[5]MANTHIRAM A, YU X W, WANG S F. Lithium battery chemistries enabled by solid-state electrolytes[J]. Nature Reviews Materials, 2017, 2(4): 16103.

[6]WALSH A. The quest for new functionality[J].Nature Chemistry, 2015, 7(4): 274-275.

[7]BELSKY A, HELLENBRANDT M, KARENV L, et al.New developments in the inorganic crystal structure database (ICSD):accessibility in support of materials research and design[J].Acta Crystallographica Section B:Structural Science Crystal Engineering and Materials, 2002, B58(1/2/3): 364-369.

[8]GROOMC R, BRUNOI J, LIGHTFOOTM P, et al. The cambridge structural database[J]. Acta Crystallographica Section B:Structural Science Crystal Engineering and Materials, 2016, 72(Part 2): 171-179.

[9]JAIN A, ONG S P, HAUTIER G, et al.Commentary: the materials project: a materials genome approach to accelerating materials innovation[J].APL Materials, 2013, 1(1): 011002.

[10]CURTAROLO S, SETYAWAN W, WANG S D, et al.AFLOWLIB.ORG: a distributed materials properties repository from high-throughput ab initio calculations[J].Computational Materials Science, 2012, 58: 227-235.

[11]SAALJ E, KIRKLIN S, AYKOL M, et al.Materials design and discovery with high-throughput density functional theory: the open quantum materials database (OQMD)[J].JOM, 2013, 65: 1501-1509.

[12]HACHMANN J, OLIVARES-AMAYA R, ATAHAN-EVRENK S, et al.The Harvard clean energy project: large-scale computational screening and design of organic photovoltaics on the world community grid[J]. The Journal of Physical Chemistry Letters, 2011, 2(17): 2241-2251.

[13]ORTIZ C, ERIKSSON O, KLINTENBERG M.Data mining and accelerated electronic structure theory as a tool in the searchfor new functional materials[J].Computational Materials Science, 2009, 44(4): 1042-1049.

[14]JIE J S, WENG M Y, LI S N, et al. A new material go database and its comparison with other high-throughput electronic structure databases for their predicted energy band gaps[J]. Science China Technological Sciences, 2019, 62: 1423-1430.

[15]CHEN C, ZUO Y X, YE W K, et al.Acritical review of machine learning of energy materials[J]. Advanced Energy Materials, 2020, 10(8): 1903242.

[16]SHA W X, GUO Y Q, YUAN Q, et al. Artificial intelligence to power the future of materials science and engineering[J].Advanced Intelligent Systems, 2020, 2(4): 1900143.

[17]AGRAWAL A, CHOUDHARY A.Perspective:materials informatics and big data: realization of the “fourth paradigm” of science in materials science[J].APL Materials, 2016, 4(5): 053208.

[18]VEGGE T, TARASCON J M, EDSTRÖ M K. Toward better and smarter batteries by combining AI with multisensory and self-healing approaches[J]. Advanced Energy Materials, 2021, 11(23): 2100362.

[19]EL-BOUSIYD H, LOMBARDO T, PRIMOE N, et al. What can text mining tell us about lithium-ion battery researchers’ habits?[J].Batteries & Supercaps, 2021, 4(5): 758-766.

[20]DUQUESNOY M, LOMBARDO T, CHOUCHANE M, et al. Data-driven assessment of electrode calendaring process by combining experimental results, in silico mesostructures generation and machine learning[J]. Journal of Power Sources, 2020, 480: 229103.

[21]GAO X L, LIU X H, HE R, et al. Designed high-performance lithium-ion battery electrodes using a novel hybrid model-data driven approach[J].Energy Storage Materials, 2021, 36(18): 435-458.

[22]ZHOU T, SONG Z, SUNDMACHER K.Big datacreates new opportunities for materials research: a review on methods and applications of machine learning for materials design[J].Engineering, 2019, 5(6): 1017-1026.

[23]YANG K K, WU Z, ARNOLDF H.Machine-learning-guided directed evolution for protein engineering[J]. Nature Methods,2019, 16(8): 687-694.

[24]WEI J, CHU X, SUN X Y, et al. Machine learning in materials science[J]. InfoMat, 2019, 1(3): 338-358.

[25]LI J G, TU Y X, LIU R, et al. Toward “on-demand” materials synthesis and scientific discovery through intelligent robots[J]. Advanced Science, 2020, 7(7): 1901957.

[26]BUTLER K T, DAVIES D W, CARTWRIGHT H, et al. Machine learning for molecular and materials science[J].Nature, 2018, 559(7715): 547-555.

[27]CHEN A, ZHANG X, ZHOU Z. Machine learning: accelerating materials development for energy storage and conversion[J]. InfoMat, 2020, 2(3): 553-576.

[28]GUG H, NOH J, KIM I, et al. Machine learning for renewable energy materials[J]. Journal of Materials Chemistry A, 2019, 7: 17096-17117.

[29]SEVERSON K A, ATTIA P M, JIN N, et al. Data-driven prediction of battery cycle life before capacity degradation[J].Nature Energy, 2019, 4(5): 383-391.

[30]PETRICH L, WESTHOFF D, FEINAUER J, et al. Crack detection in lithiumion cells using machine learning[J]. Computational Materials Science, 2017, 136: 297-305.

[31]AMIRI M N, HÅKANSSON A, BURHEIM O S, et al.Lithium-ion battery digitalization: combining physics-based models and machine learning[J]. Renewable and Sustainable Energy Reviews, 2024, 200: 114577.

[32]IRWINJ J, SHOICHETB K. ZINC:a free database of commercially available compounds for virtual screening[J]. Journal of Chemical Information and Modeling, 2005, 45(1): 177-182.

[33]CURTAROLO S, MORGAN D, PERSSON K, et al. Predicting crystal structures with data mining of quantum calculations[J].Physical Review Letters, 2003, 91(13): 135503.

[34]SENDEKA D, YANG Q, CUBUKE D, et al. Holistic computational structure screening of more than 12 000 candidates for solid lithium-ion conductor materials[J]. Energy & Environmental Science, 2017, 10(1): 306-320.

[35]BHOWMIK A, CASTELLII E, GARCIA-LASTRAJ M, et al. A perspective on inverse design of battery interphases using multi-scale modelling, experiments and generative deep learning[J]. Energy Storage Materials, 2019, 21: 446-456.

[36]ZHU S, HE C N, ZHAO N Q, et al. Data-driven analysis on thermal effects and temperature changes of lithium-ion battery[J]. Journal of Power Sources, 2021, 482: 228983.

[37]GAO Y C, YUAN Y H, HUANG S Z, et al. A knowledge-data dual-driven framework for predicting the molecular properties of rechargeable battery electrolytes[J]. Angewandte Chemie International Edition, 2025, 64(4): e202416506.

[38]SHODIEV A, DUQUESNOY M, ARCELUS O, et al. Machine learning 3D-resolved prediction ofelectrolyte infiltration in battery porous electrodes[J]. Journal of Power Sources, 2021, 511: 230384.

[39]ZHANG Y W, TANG Q C, ZHAN G Y, et al. Identifying degradation patterns of lithium ion batteries from impedance spectroscopy using machine learning[J]. Nature Communications, 2020, 11(1): 1706.

[40]WU B, HAN S, SHIN K G, et al. Application of artificial neural networks in design of lithium-ion batteries[J]. Journal of Power Sources, 2018, 395: 128-136.

[41]SHANDIZM A, GAUVIN R. Application of machine learning methods for the prediction of crystal system of cathode materials in lithium-ion batteries[J]. Computational Materials Science, 2016, 117: 270-278.

[42]KIREEVA N, PERVOVV S, TSIVADZEA Y. Machine learning-based evaluation of functional characteristics of Li-rich layered oxide cathode materials using the data of XPS and XRD spectra[J].Computational Materials Science, 2024, 231: 112591.

[43]JOSHIR P, EICKHOLT J, LIL L, et al. Machine learning the voltage of electrode materials in metal-ion batteries[J].ACS Applied Materials & Interfaces, 2019, 11(20): 18494-18503.

[44]ALLAM O,CHO B W,KIM K C, et al.Application of DFT-based machine learning for developing molecular electrode mater-ials in Li-ion batteries[J]. RSC Advances, 2018, 8(69): 39414-39420.

[45]XU G S, JIANG M X, LI J L, et al. Machine learning-accelerated discovery and design of electrode materials and electrolytes for lithium ion batteries[J]. Energy Storage Materials, 2024, 72: 103710.

[46]FITZHUGH W, GHEN X, WANG Y C, et al. Solid electrolyte-interphase design in constrained ensemble for solid-state batteries[J]. Energy& Environmental Science, 2021, 14(8): 4574-4583.

[47]WANG Z H, WANG L, ZHANG H, et al. Materials descriptors of machine learning to boost development of lithium-ion batteries[J]. Nano Convergence, 2024, 11: 8.

[48]LYU C D, ZHOU X, ZHONG L X, et al. Machine learning: an advanced platform for materials development and state prediction in lithium-ion batteries[J]. Advanced Materials, 2022, 34: 2101474.

[49]OLIVARES-AMAYA R, AMADOR-BEDOLLA C, HACHMANN J, et al. Accelerated computational discovery of high-performance materials for organic photovoltaics by means of cheminformatics[J]. Energy & Environmental Science, 2011, 4(12): 4849-4861.

[50]ZHOU X, XU C, GUO X L, et al. Computational and machine learning-assisted discovery and experimental validation of conjugated sulfonamide cathodes for lithium-ion batteries[J]. Advanced Energy Materials, 2024: 2401658.

[51]WANG Y M, XIE T, FRANCE-LANORD A, et al. Toward designing highly conductive polymer electrolytes by machine learning assisted coarse-grained molecular dynamics[J]. Chemistry of Materials, 2020, 32(10): 4144-4151.

[52]ZENG S B, CHEN S, ALKALI B. Lithium-ion battery capacity estimation based on incremental capacity analysis and deep convolutional neural network[J]. Energies, 2024, 17(6): 1272.

[53]ZOU M, WANG J L, YAN D, et al. Novel state of charge estimation method of containerized lithium-ion battery energy storage system based on deep learning[J]. Journal of Power Sources,2024, 624: 235609.

[54]LU J H, XIONG R, TIAN J P, et al. Deep learning to estimate lithium-ion battery state of health without additional degradation experiments[J]. Nature Communications, 2023, 14(1): 2760.

[55]ZHAO B, ZHANG W G, ZHANG Y R, et al. Research on the remaining useful life prediction method for lithium-ion batteries by fusion of feature engineering and deep learning[J]. Applied Energy, 2024, 358: 122325.

[56]TU H, BORAH M, MOURA S, et al. Remaining discharge energy prediction for lithium-ion batteries over broad current ranges: a machine learning approach[J]. Applied Energy, 2024, 376: 124086.

[57]NAZARI A, KAVIAN S, NAZARI A. Lithium-ion batteries’ energy efficiency prediction using physics-based and state-of-the-art artificial neural network-based models[J]. Journal of Energy Resources Technology, 2020, 142(10): 102001.

[58]THON C, FINKE B, KWADE A, et al. Artificial intelligence in process engineering[J]. Artificial Intelligence in Process Engineering, 2021, 3(6): 2000261.

[59]HAGHI S, HIDALGOM F V, NIRIM F, et al. Machine learning in lithium-ion battery cell production: a comprehensive mapping study[J]. Batteries & Supercaps, 2023, 6(7): e202300046.

[60]TURETSKYY A, THIEDE S, THOMITZEK M, et al. Toward data-driven applications in lithium-ion battery cell manufacturing[J]. Energy Technology, 2020, 8(2): 1900136.

[61]DRAKOPOULOS S, GHOLAMIPOUR-SHIRAZI A, MACDONALD P, et al. Formulation and manufacturing optimization of lithium-ion graphite-based electrodes via machine learning[J].Cell Reports Physical Science, 2021, 2(12): 100683.

[62]CUI X J, GARG A, THAON T, et al. Machine learning approach for solving inconsistency problems of Li-ion batteries during the manufacturing stage[J]. International Journal of Energy Research, 2020, 44(11): 9194-9204.

[63]LIU K L, HU X S, MENG J H, et al. RUBoost-based ensemble machine learning for electrode quality classification in Li-ion battery manufacturing[J]. IEEE/ASME Transactions on Mechatronics, 2022, 27(5): 2474-2483.

[64]YANG S J, ZHANG C P, JIANG J C, et al. Fast screening of lithium-ion batteries for second use with pack-level testing and machine learning[J]. eTransportation, 2023, 17: 100255.

[65]SAMANTA A,CHOWDHURI S,WILLIAMSONS S.Machine learning-based data-driven fault detection/diagnosis of lithium-ion battery: a critical review[J]. Electronics, 2021, 10(11): 1309.

[66]LIU Y Y, LIAO C L, ZHANG W J, et al. Internal short circuit diagnosis of lithium-ion battery based on mechanism model and deep learning[J]. Journal of the Electrochemical Society, 2022, 169(10): 100514.

[67]SONG Y, YU J S, ZHOU J H, et al. Detection of voltage fault in lithium-ion battery based on equivalent circuit model-informed neural network[J].IEEE Transactions on Instrumentation and Measurement, 2024, 73: 3508010.

[68]FU T Y, MONACO F, LI J Z, et al. Deep-learning-enabled crack detection and analysis in commercial lithium-ion battery cathodes[J]. Advanced Functional Materials, 2022, 32(39): 2203070.

[69]YAO L, ZHENG J, XIAO Y Q, et al. An intelligent fault diagnosis method for lithium-ion battery pack based on empirical mode decomposition and convolutional neural network[J]. Journal of Energy Storage, 2023, 72: 108181.

[70]ZHUANG W C, LIU Z T, SU H Y, et al. An intelligent thermal management system for optimized lithium-ion battery pack[J]. Applied Thermal Engineering, 2021, 189: 116767.

[71]ZHANG X C, XIANG H X, XIONG X, et al. Benchmarking core temperature forecasting for lithium-ion battery using typical recurrent neural networks[J].Applied Thermal Engineering, 2024, 248: 123257.

[72]ZHAO H Q, CHEN Z, SHU X, et al. Online surface temperature prediction and abnormal diagnosis of lithium-ion batteries based on hybrid neural network and fault threshold optimization[J]. Reliability Engineering & System Safety, 2024, 243: 109798.

[73]GOSWAMIB R D, ABDISOBBOUHI Y, DU H, et al. Advancing battery safety: integrating multiphysics and machine learning for thermal runaway prediction in lithium-ion battery module[J]. Journal of Power Sources, 2024, 614: 235015.

[74]YANG Y K, HE J R, CHEN C L, et al. Balancing awareness fast charging control for lithium-ion battery pack using deep reinforcement learning[J]. IEEE Transactions on Industrial Electronics, 2024, 71(4): 3718-3727.