机械滥用场景下锂离子电池温度多模态神经网络预测方法

李杰; 李倩; 覃正鹏; 黄鑫蓉

doi:10.7652/xjtuxb202602004

您当前的位置：

首页 >

文章列表页 >

机械滥用场景下锂离子电池温度多模态神经网络预测方法

专题新能源电动车辆热管理 | 更新时间：2026-02-09

- 机械滥用场景下锂离子电池温度多模态神经网络预测方法
- Multimodal Neural Network-Based Temperature Prediction Method for Lithium-Ion Batteries Under Mechanical Abuse Scenarios
- 西安交通大学学报 2026年60卷第2期页码：38-48
- 作者机构：
  
  长安大学能源与电气工程学院，710064，西安
- 作者简介：
  
  作者简介：李杰（1984—），男，副教授，博士生导师。
- 基金信息：
  
  国家自然科学基金资助项目。(52407236)
- DOI：10.7652/xjtuxb202602004
  中图分类号： TM734
- 收稿：2025-07-03，
  
  纸质出版：2026-02-10
- 稿件说明：
移动端阅览
李杰, 李倩, 覃正鹏, 等. 机械滥用场景下锂离子电池温度多模态神经网络预测方法[J]. 西安交通大学学报, 2026,60(2):38-48.

LI Jie, LI Qian, QIN Zhengpeng, et al. Multimodal Neural Network-Based Temperature Prediction Method for Lithium-Ion Batteries Under Mechanical Abuse Scenarios[J]. Journal of Xi'an Jiaotong University, 2026, 60(2): 38-48.
李杰, 李倩, 覃正鹏, 等. 机械滥用场景下锂离子电池温度多模态神经网络预测方法[J]. 西安交通大学学报, 2026,60(2):38-48. DOI： 10.7652/xjtuxb202602004.

LI Jie, LI Qian, QIN Zhengpeng, et al. Multimodal Neural Network-Based Temperature Prediction Method for Lithium-Ion Batteries Under Mechanical Abuse Scenarios[J]. Journal of Xi'an Jiaotong University, 2026, 60(2): 38-48. DOI： 10.7652/xjtuxb202602004.

摘要

针对锂离子电池机械滥用场景下单一模态建模、简单融合策略和物理约束不足导致的温度预测精度低的问题，提出了一种基于多模态神经网络的最高温度预测方法。首先，选用1200 mA·h的18650型锂离子电池，在10%～90%荷电状态范围内开展机械挤压实验，采集热成像图、荷电状态、电压、载荷和变形等多源数据，构建超过200组的多模态数据集。然后，利用多尺度卷积模块、注意力机制和双向长短期记忆网络提取热成像时空特征，结合一维卷积神经网络与双向长短期记忆网络提取电化学与力学时序特征，并通过变换器融合机制实现跨模态特征自适应融合。最后，引入温度变化率约束的物理损失函数，以提升预测的合理性和鲁棒性。研究结果表明：多模态神经网络模型短期预测（1～3步）的决定系数达到0.972～0.990，中长期预测（6～15步）的决定系数稳定在0.900以上，能够实现机械滥用条件下锂离子电池的多步预测，并在不同荷电状态和预测步长下表现出显著的精度优势和良好的适应性，为锂离子电池热失控的多级预警和安全管理提供了技术支持。

Abstract

To address the issue of low temperature prediction accuracy in lithium-ion batteries under mechanical abuse scenarios caused by single-modal modeling，simplistic fusion strategies，and insufficient physical constraints，a maximum temperature prediction method based on multimodal neural networks is proposed.First，18650-type lithium-ion batteries with a capacity of 1200 mA·h were selected，and mechanical compression experiments were conducted within a state of charge（SOC）range of 10% to 90%.Multi-source data including thermal images，SOC，voltage，load，and deformation were collected to construct a multimodal dataset comprising over 200 sample groups.Subsequently，multiscale convolutional modules，attention mechanisms，and bidirectional long short-term memory（Bi-LSTM）networks were used to extract spatiotemporal features from thermal images，while one-dimensional convolutional neural networks combined with Bi-LSTM were employed to extract electrochemical and mechanical spatial-temporal features. Cross-modal adaptive feature fusion was achieved through a transformer fusion mechanism. Finally，a physics-informed loss function incorporating temperature change rate constraints was introduced to improve prediction rationality and robustness.The results show that the multimodal neural network model achieves coefficients of determination of 0.972 to 0.990 for short-term predictions（1—3 steps）and remains stable above 0.900 for medium-to long-term predictions（6—15 steps）. The model enables effective multi-step temperature prediction under mechanical abuse conditions and demonstrates significant accuracy advantages and good adaptability across different SOC levels and prediction horizons，providing technical support for multi-level early warning and safety management of lithium-ion battery thermal runaway.

关键词

Keywords

references

ZHANG Xi, CHEN Shun, ZHU Jingzhe, et al.A critical review of thermal runaway prediction and earlywarning methods for lithium-ion batteries[J].Energy Material Advances, 2023, 4:0008.

ZHOU Yuxin, DING Yifei, CHEN Yuying, et al. Thermal degradation of lithium-ion battery cathodes:a machine learning prediction of stability and safety[J]. Energy Materials, 2025 , 5 :500077 .

WANG Zilong, SADEGHI H, HUANG Xinyan, et al.Thermal runaway and flame propagation in battery packs:numerical simulation and deep learning prediction[J].Engineering Applications of Computational Fluid Mechanics, 2025, 19(1):2445160.

WANG Gongquan, PING Ping, KONG Depeng, et al. Advances and challenges in thermal runaway modeling of lithium-ion batteries[J].The Innovation, 2024, 5(4):100624.

AZUAJE-BERBECÍB J, ERTAN H B.A model for the prediction of thermal runaway in lithium-ion batteries[J]. Journal of Energy Storage, 2024, 90(Part A):111831 .

朱振威,苗嘉伟,祝夏雨,等.基于机器学习方法的锂电池剩余寿命预测研究进展[J].储能科学与技术, 2024, 13(9):3134-3149.

ZHU Zhenwei, MIAO Jiawei, ZHU Xiayu, et al.Research progress in lithium-ion battery remaining useful life prediction based on machine learning[J].Energy Storage Science and Technology, 2024, 13(9):3134-3149.

张世超,沈泽宇,陆盈盈.金属锂电池的热失控与安全性研究进展[J].物理化学学报, 2021 , 37(1):55-72.

ZHANG Shichao, SHEN Zeyu, LU Yingying.Research progress of thermal runaway and safety for lithium metal batteries[J].Acta Physico-Chimica Sinica, 2021 , 37(1):55-72.

FENG Xuning, REN Dongsheng, HE Xiangming, et al. Mitigating thermal runaway of lithium-ion batteries[J]. Joule, 2020, 4(4):743-770.

THELEN A, HUAN Xun, PAULSON N, et al. Probabilistic machine learning for battery health diagnostics and prognostics:review and perspectives[J]. npj Materials Sustainability, 2024, 2(1):14.

KHALEGHI S, HOSEN M S, KARIMI D, et al.Developing an online data-driven approach for prognostics and health management of lithium-ion batteries[J]. Applied Energy, 2022, 308:118348.

WANG Fujin, ZHAI Zhi, ZHAO Zhibin, et al.Physics-informed neural network for lithium-ion battery degradation stable modeling and prognosis[J].Nature Communications, 2024, 15(1):4332.

LIU Hongao, LI Chang, HU Xiaosong, et al.Multimodal framework for battery state of health evaluation using open-source electric vehicle data[J].Nature Communications, 2025, 16(1):1137.

SUH S, MITTAL D A, BELLO H, et al.Remaining useful life prediction of lithium-ion batteries using spatio-temporal multimodal attention networks[J].Heliyon, 2024, 10(16):e36236.

KIM S W, KWAK E, KIM J H, et al.Modeling and prediction of lithium-ion battery thermal runaway via multiphysics-informed neural network[J].Journal of Energy Storage, 2023, 60:106654.

JEONG J, KWAK E, KIM J H, et al.Prediction of thermal runaway for a lithium-ion battery through multiphysics-informed DeepONet with virtual data[J]. eTransportation, 2024, 21 :100337.

DAS GOSWAMI B R, ABDISOBBOUHI Y, DU Hui, 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.

CHAI Zhixiong, LI Junqiu, LIU Ziming, et al.Experimental analysis and safety assessment of thermal runaway behavior in lithium iron phosphate batteries under mechanical abuse[J].Scientific Reports, 2024, 14(1):8673.

KONG Jinzhen, LIU Jie, ZHU Jingzhe, et al.Review on lithium-ion battery PHM from the perspective of key PHM steps[J].Chinese Journal of Mechanical Engineering, 2024, 37(1):71.

XIONG Rui, WANG Ju, SHEN Weixiang, et al.Coestimation of state of charge and capacity for lithiumion batteries with multi-stage model fusion method[J]. Engineering, 2021 , 7(10):1469-1482.

DING Shuya, DONG Chaoyu, ZHAO Tianyang, et al.A meta-learning based multimodal neural network for multistep ahead battery thermal runaway forecasting[J].IEEE Transactions on Industrial Informatics, 2021, 17(7):4503-4511.

LANUBILE A, BOSONI P, POZZATO G, et al.Do main knowledge-guided machine learning framework for state of health estimation in lithium-ion batteries[J]. Communications Engineering, 2024, 3(1):168.

IFTIKHAR M, SHOAIB M, ALTAF A, et al.A deep learning approach to optimize remaining useful life prediction for li-ion batteries[J].Scientific Reports, 2024, 14(1):25838.

LI Jie, ZHANG Yunlong, YUAN Boxing, et al.A data-driven thermal runaway warning method for lithium-ion batteries under mechanical abuse[J].IEEE Transactions on Transportation Electrification, 2025, 11(2):5169-5179.

PAN Zhexin, SEDLATSCHEK T, XIA Yong.Effect of state-of-charge and air exposure on tensile mechanical properties of lithium-ion battery electrodes[J].Journal of the Electrochemical Society, 2020, 167(9):090517.

CORTADA-TORBELLINO M, ELVIRA D G, EL AROUDI A, et al.Review of lithium-ion battery internal changes due to mechanical loading[J].Batteries, 2024, 10(7):258.

VALIZADEH A, AMIRHOSSEINI M H.Machine learning in lithium-ion battery:applications, challenges, and future trends[J].SN Computer Science, 2024, 5(6):717.

KIM H, JUNG T, JUNG J, et al.Accurate prediction of electrochemical degradation trajectory for lithium-ion battery using self-discharge[J].International Journal of Energy Research, 2024, 2024(1):1758578.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

电动汽车动力电池全寿命周期交流加热策略研究

基于关系提示的单模块单步骤实体关系抽取方法研究

采用轻量级卷积神经网络的H.266/通用视频编码跨分量预测

采用感受野优化与渐进特征融合的图像超分辨率算法

锂离子电池枝晶内短路的电热特性研究