基于物理信息引导的磷酸铁锂电池荷电状态估计方法

骆可; 强栋; 翁伟; 姚绒; 付俊成; 张轩铭; 孟锦豪

doi:10.19799/j.cnki.2095-4239.2026.0420

您当前的位置：

首页 >

文章列表页 >

基于物理信息引导的磷酸铁锂电池荷电状态估计方法

储能XXXX | 更新时间：2026-07-14

- 基于物理信息引导的磷酸铁锂电池荷电状态估计方法
- Physics-Informed Adaptive Filtering Method for State-of-Charge Estimation of Lithium Iron Phosphate Batteries
- 储能科学与技术 2026年页码：1-15
- 作者机构：
  
  1.特变电工新疆新能源股份有限公司，新疆维吾尔自治区乌鲁木齐 830011
  2.西安交通大学国家储能技术产教融合创新平台（中心），陕西西安 710049
  3.西安交通大学电气学院，陕西西安 710049
- 作者简介：
  
  骆可（1985—），男，学士，中级工程师，研究方向为储能系统集成与运行优化，E-mail：luoke@tbea.com；
  孟锦豪，副教授，研究方向为储能系统设计与能量管理、电池建模与SOX估计、数据-机理混合的大数据电池诊断预警、电池系统测试方法与性能评价，E-mail：jinhao@xjtu.edu.cn。
- 基金信息：
  
  工信部制造业高质量发展专项(gx23225)
- DOI：10.19799/j.cnki.2095-4239.2026.0420
  中图分类号： TM 912
- 收稿：2026-05-15，
  
  修回：2026-06-12，
  
  网络首发：2026-07-14，
- 稿件说明：
移动端阅览
骆可, 强栋, 翁伟, 等. 基于物理信息引导的磷酸铁锂电池荷电状态估计方法[J]. 储能科学与技术, XXXX, XX(XX): 1-15.

Luo Ke, Qiang Dong, Weng Wei, et al. Physics-Informed Adaptive Filtering Method for State-of-Charge Estimation of Lithium Iron Phosphate Batteries[J]. Energy Storage Science and Technology, XXXX, XX(XX): 1-15.
骆可, 强栋, 翁伟, 等. 基于物理信息引导的磷酸铁锂电池荷电状态估计方法[J]. 储能科学与技术, XXXX, XX(XX): 1-15. DOI： 10.19799/j.cnki.2095-4239.2026.0420.

Luo Ke, Qiang Dong, Weng Wei, et al. Physics-Informed Adaptive Filtering Method for State-of-Charge Estimation of Lithium Iron Phosphate Batteries[J]. Energy Storage Science and Technology, XXXX, XX(XX): 1-15. DOI： 10.19799/j.cnki.2095-4239.2026.0420.

摘要

磷酸铁锂（lithium iron phosphate，LFP）电池在电动汽车与储能系统中应用广泛，但其电压平台区可观测性不足，表面力信号又易受温度漂移、机械滞后与老化效应耦合干扰，导致复杂工况下荷电状态（state of charge，SOC）估计精度下降。针对上述问题，本文提出一种物理信息引导熵权自适应扩展卡尔曼滤波（physics-informed entropy-weighted adaptive extended Kalman filter，PI-EWAEKF）方法。首先，基于准静态标定结果构建开路电压（open-circuit voltage，OCV）-SOC 与表面力-SOC 物理映射关系；其次，采用增量力分析（incremental force analysis，IFA）提取抗漂移力学特征，并结合 OCV 梯度构建多源物理特征集；进一步，利用香农熵和物理梯度共同表征电压与表面力信号的实时置信度，并将其映射为AEKF测量噪声协方差矩阵的自适应调节因子，从而实现多源观测权重的动态分配。实验在25℃、35℃、45℃以及健康状态约为 100% 和 91% 条件下开展，并采用联邦城市行驶工况（federal urban driving schedule，FUDS）、动态应力测试（dynamic stress test，DST）和新欧洲行驶工况（new European driving cycle，NEDC）进行验证。结果表明，相较于单电压观测模型，所提方法在 25℃动态工况下将 FUDS、DST 和 NEDC 工况的均方根误差分别由 4.13%、3.70% 和 4.10% 降低至 0.96%、0.97% 和 0.79%，最大降幅约为 80.7%。在高温与老化耦合条件下，所提方法仍能保持较低误差水平，验证了其在复杂工况下的估计精度与鲁棒性。

Abstract

Lithium iron phosphate (LFP) batteries are widely used in electric vehicles and energy storage systems. However

their weak observability in the voltage plateau region

together with the coupled effects of temperature drift

mechanical hysteresis

and aging on surface-force signals

reduces the accuracy of state of charge (SOC) estimation under complex operating conditions. To address these issues

this paper proposes a physics-informed entropy-weighted adaptive extended Kalman filter (PI-EWAEKF) method. First

the physical mapping relationships of open-circuit voltage (OCV)-SOC and surface force-SOC are established based on quasi-static calibration results. Second

incremental force analysis is used to extract drift-resistant mechanical features

which are combined with OCV gradients to construct a multi-source physical feature set. Furthermore

Shannon entropy and physical gradients are jointly used to characterize the real-time confidence of voltage and surface-force signals

which is then mapped into adaptive adjustment factors for the measurement noise covariance matrix of the adaptive extended Kalman filter (AEKF)

thereby enabling dynamic allocation of multi-source observation weights. Experiments are conducted at 25 °C

35 °C

and 45 °C under state of health levels of approximately 100% and 91%

and are validated using the federal urban driving schedule (FUDS)

dynamic stress test (DST)

and new European driving cycle (NEDC). The results show that

compared with a voltage-only observation model

the proposed method reduces the root mean square errors under FUDS

DST

and NEDC profiles at 25 °C from 4.13%

3.70%

and 4.10% to 0.96%

0.97%

and 0.79%

respectively

with a maximum reduction of approximately 80.7%. Under coupled high-temperature and aging conditions

the proposed method still maintains low error levels

demonstrating its estimation accuracy and robustness under complex operating conditions.

关键词

Keywords

references

张磊.锂离子电池储能电站的运行状态监测与评估[J].储能科学与技术,2025,14(09):3538-3540.DOI:10.19799/j.cnki.2095-4239.2025.0536.

ZHANG L. Operation state monitoring and evaluation of lithium-ion battery energy storage power stations[J]. Energy Storage Science and Technology, 2025, 14(09): 3538-3540. DOI: 10.197 99/j.cnki.2095-4239.2025.0536.

陈勇,王俊磊,王鹏,等.锂离子电池早期剩余寿命预测方法综述[J].电池,2026,56(01):222-230.DOI:10.19535/j.1001-1579.2026.01.035.

CHEN Y, WANG J L, WANG P, et al. Review of early remaining useful life prediction methods for lithium-ion batteries[J]. Battery Bimonthly, 2026, 56(01): 222-230. DOI: 10.19535/j.1001-1579.2026.01.035.

YU K, LIU P, XU B, et al. Warning lithium-ion battery thermal runaway with 4-min relaxation voltage[J]. Applied Energy, 2025, 377: 124466. DOI: 10.1016/j.apenergy.2024.124466.

WANG T, YANG X, YANG F, et al. Expansion force distribution characteristics of prismatic LiFePO4 batteries during thermal runaway caused by overheating[J]. Journal of Energy Storage, 2026, 152: 120514. DOI: 10.1016/j.est.2026.120514.

SHEN L, LI J, MENG L, et al. Transfer learning-based state of charge and state of health estimation for Li-ion batteries: a review[J]. IEEE Transactions on Transportation Electrification, 2024, 10: 1465-1481. DOI: 10.1109/TTE.2023.3293551.

FU J, SONG Z, MENG J, et al. AI-augmented electrochemical model for lithium-ion battery: recent advances and perspectives[J]. Journal of Energy Chemistry, 2026, 113: 1056-1080. DOI: 10.1016/j.jechem.2025.10.008.

SHARMA S, GARG A, PANIGRAHI B K. Predicting state-of-charge using gradient-boosted SVR ensemble technique for lithium ion battery used in EVs[J]. IEEE Transactions on Transportation Electrification, 2024, 10: 4441-4454. DOI: 10.1109/TTE.2023.3310159.

刘素贞, 袁路航, 张闯, 等. 基于超声时域特征及随机森林的磷酸铁锂电池荷电状态估计[J]. 电工技术学报, 2022, 37(22): 5872-5885.

LIU S Z, YUAN L H, ZHANG C, et al. State of charge estimation of lithium iron phosphate batteries based on ultrasonic time-domain features and random forest[J]. Transactions of China Electrotechnical Society, 2022, 37(22): 5872-5885.

田佳强, 张泽培, 潘天红, 等. 基于动量反向传播神经网络的锂离子电池建模与荷电状态估计[J]. 电工技术学报, 2025, 40(24): 8156-8170.

TIAN J Q, ZHANG Z P, PAN T H, et al. Lithium-ion battery modeling and state of charge estimation based on momentum backpropagation neural network[J]. Transactions of China Electrotechnical Society, 2025, 40(24): 8156-8170.

TIAN H, LI A, LI X. SOC estimation of lithium-ion batteries for electric vehicles based on multimode ensemble SVR[J]. Journal of Power Electronics, 2021, 21: 1365-1373. DOI: 10.1007/s43236-021-00279-9.

窦元运,张成知,封居强.多因素影响下融合RNN和AUKF的矿用锂离子电池SOC估计[J].电源技术,2025,49(04):764-771.

DOU Y Y, ZHANG C Z, FENG J Q. SOC estimation of mine-used lithium-ion batteries integrating RNN and AUKF under multi-factor influences[J]. Chinese Journal of Power Sources, 2025, 49(04): 764-771.

杜伟, 王圣, 李健, 等. 基于 CNN-LSTM-AM 模型的储能锂离子电池荷电状态预测[J]. 电工技术学报, 2025, 40(9): 2982-2995.

DU W, WANG S, LI J, et al. State of charge prediction of energy storage lithium-ion batteries based on CNN-LSTM-AM model[J]. Transactions of China Electrotechnical Society, 2025, 40(9): 2982-2995.

程思涵,刘思懿,郭子旭,等.基于CNN-GRU-注意力的锂离子电池SOC估计[J].电池,2026,56(01):131-135.DOI:10.19535/j.1001-1579.2026.01.019.

CHENG S H, LIU S Y, GUO Z X, et al. SOC estimation of lithium-ion batteries based on CNN-GRU-attention[J]. Battery Bimonthly, 2026, 56(01): 131-135. DOI: 10.19535/j.1001-1579.2026.01.019.

马昊远,吴焱,王通,等.基于力-电-温度信号和CNN-BiLSTM模型的磷酸铁锂电池SOC估计[J].储能科学与技术,2025,14(07):2865-2874.DOI:10.19799/j.cnki.2095-4239.2025.0062.

MA H Y, WU Y, WANG T, et al. SOC estimation of lithium iron phosphate batteries based on force-electric-thermal signals and CNN-BiLSTM model[J]. Energy Storage Science and Technology, 2025, 14(07): 2865-2874. DOI: 10.19799/j.cnki.2095-4239.2025.0062.

FENG X, CHEN J, ZHANG Z, et al. State-of-charge estimation of lithium-ion battery based on clockwork recurrent neural network[J]. Energy, 2021, 236: 121360. DOI: 10.1016/j.energy.2021.121360.

谭必蓉,杜建华,叶祥虎,等.基于模型的锂离子电池SOC估计方法综述[J].储能科学与技术,2023,12(06):1995-2010.DOI:10.19799/j.cnki.2095-4239.2023.0016.

TAN B R, DU J H, YE X H, et al. Review of model-based SOC estimation methods for lithium-ion batteries[J]. Energy Storage Science and Technology, 2023, 12(06): 1995-2010. DOI: 10.19799/j.cnki.2095-4239.2023.0016.

吕高,樊郭宇,张嘉蕾,等.温度自适应SMO算法估计锂离子电池的SOC[J].电池,2024,54(03):334-339.DOI:10.19535/j.1001-1579.2024.03.009.

LÜ G, FAN G Y, ZHANG J L, et al. SOC estimation of lithium-ion batteries using temperature-adaptive SMO algorithm[J]. Battery Bimonthly, 2024, 54(03): 334-339. DOI: 10.19535/j.1001-1579.2024.03.009.

LI H, YU H, ZHU R, et al. Distributed finite-time S℃ estimation for lithium-ion battery packs[J]. IEEE Transactions on Industrial Electronics, 2026. DOI: 10.1109/TIE.2026.3661048.

ZHANG S, WANG X, CHEN Z, et al. Antidisturbance state-of-charge estimation for lithium-ion batteries using nonlinear extended state observers[J]. IEEE Transactions on Transportation Electrification, 2025, 11: 2918-2928. DOI: 10.1109/TTE.2024.3430540.

WU X, LI X, DU J. State of charge estimation of lithium-ion batteries over wide temperature range using unscented Kalman filter[J]. IEEE Access, 2018, 6: 41993-42003. DOI: 10.1109/ACCESS.2018.2860050.

WANG Y, CHENG Y, XIONG Y, et al. Estimation of battery open-circuit voltage and state of charge based on dynamic matrix control-extended Kalman filter algorithm[J]. Journal of Energy Storage, 2022, 52: 104860. DOI: 10.1016/j.est.2022.104860.

DU J, WANG J, TAN B, et al. Estimation of battery state of charge based on changing window adaptive extended Kalman filtering[J]. Journal of Energy Storage, 2024, 103: 114325. DOI: 10.1016/j.est.2024.114325.

ZHANG K, XIONG R, LI Q, et al. A novel pseudo-open-circuit voltage modeling method for accurate state-of-charge estimation of LiFePO4 batteries[J]. Applied Energy, 2023, 347: 121406. DOI: 10.1016/j.apenergy.2023.121406.

JIANG Y, XU J, LIU M, et al. An electromechanical coupling model-based state of charge estimation method for lithium-ion pouch battery modules[J]. Energy, 2022, 259: 125019. DOI: 10.1016/j.energy.2022.125019.

XIONG R, LI Z, WANG P, et al. Advanced electrical-mechanical hybrid approach for enhanced state of charge estimation of lithium-ion batteries[J]. IEEE Transactions on Industrial Electronics, 2026, 73: 5687-5697. DOI: 10.1109/TIE.2025.3634458.

FU J, WU C, WANG J, et al. Lithium-ion battery SOH prediction based on VMD-PE and improved DBO optimized temporal convolutional network model[J]. Journal of Energy Storage, 2024, 87: 111392. DOI: 10.1016/j.est.2024.111392.

He X, Sun K, He Y, et al. Mechanistic insights into the dynamic evolution processes of LiFePO4/C batteries[J]. Chemical Engineering Journal, 2025, 515: 163447.

ZHAO H, MENG J, PENG Q. Early perception of lithium-ion battery degradation trajectory with graphical features and deep learning[J]. Applied Energy, 2025, 381: 125214. DOI: 10.1016/j.apenergy.2024.125214.

LIU Q, MA J, ZHAO X, et al. Voltage fault diagnosis and misdiagnosis analysis of battery systems using the modified Shannon entropy in real-world electric vehicles[J]. Journal of Energy Storage, 2023, 73: 109287. DOI: 10.1016/j.est.2023.109287.

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

基于物理信息神经网络的锂离子电池荷电状态估计研究综述

过充和加热滥用下大容量磷酸铁锂电池热失控特性

储能用锂电池模组的浸没式热安全性能试验研究

储能电站预制舱磷酸铁锂电池热失控燃爆危害仿真研究

磷酸铁锂单体电池热失控火灾危险性