机器学习辅助相场模拟预测锂离子输运参数对电池枝晶最大生长高度和空间利用率的影响

doi:10.19799/j.cnki.2095-4239.2024.0513

储能科学与技术 ›› 2024, Vol. 13 ›› Issue (9): 2864-2870.doi: 10.19799/j.cnki.2095-4239.2024.0513

• AI辅助先进电池设计与应用专刊 • 上一篇下一篇

机器学习辅助相场模拟预测锂离子输运参数对电池枝晶最大生长高度和空间利用率的影响

李亚捷¹(), 王依平¹, 陈斌¹, 林海龙¹, 张更²(), 施思齐¹^,³()

^1.上海大学材料科学与工程学院，上海 200444
^2.湖南理工学院机械工程学院，湖南岳阳 414006
^3.上海大学材料基因组工程研究院，上海 200444

收稿日期:2024-06-06 修回日期:2024-06-26 出版日期:2024-09-28 发布日期:2024-09-20
通讯作者: 张更,施思齐 E-mail:liyajiejuly@shu.edu.cn;geng.zhang@kaust.edu.sa;sqshi@shu.edu.cn
作者简介:李亚捷（1990—），女，博士，研究方向为电池安全性能及相场模拟研究，E-mail：liyajiejuly@shu.edu.cn；
基金资助:
国家自然科学基金(52102280);国家重点研发计划(2021YFB3802104)

Machine learning-assisted phase-field simulation for predicting the impact of lithium-ion transport parameters on maximum battery dendrite height and space utilization rate

Yajie LI¹(), Yiping WANG¹, Bin CHEN¹, Hailong LIN¹, Geng ZHANG²(), Siqi SHI¹^,³()

^1.School of Materials Science and Engineering, Shanghai University, Shanghai 200444, China
^2.College of Mechanical Engineering, Hunan Institute of Science and Technology, Yueyang 414006, Hunan, China
^3.Materials Genome Institute, Shanghai University, Shanghai 200444, China

Received:2024-06-06 Revised:2024-06-26 Online:2024-09-28 Published:2024-09-20
Contact: Geng ZHANG, Siqi SHI E-mail:liyajiejuly@shu.edu.cn;geng.zhang@kaust.edu.sa;sqshi@shu.edu.cn

摘要/Abstract

摘要：

在锂基电池反复充放电的过程中，锂离子在负极表面的不均匀沉积会导致不可控的枝晶生长，进而严重影响电池的安全性能。相场模拟方法是描述和预测枝晶生长的有力手段，然而求解描述场变量演化的偏微分方程组对计算资源的要求较高。机器学习因能快速拟合历史数据中的潜在规律以实现材料性能的预测，已被广泛用于电池材料性能预测与筛选、电池健康状况评估等方面。本文以锂离子输运参数对电池枝晶形貌的影响为例，通过相场模拟收集不同锂离子扩散系数与离子电导率对应的枝晶图像，基于这些数据训练机器学习模型，进而预测给定离子输运参数所对应的枝晶描述因子(枝晶最大生长高度和空间利用率)。结果表明K-最邻近(K-nearest neighbors)模型可以较为精准地刻画离子输运参数与两种枝晶描述因子之间的联系(R²为0.995和0.992)，同时机器学习模型对锂离子输运参数与枝晶描述因子间构效关系的挖掘方式及枝晶描述因子的区间范围都会影响预测结果的准确性。本文能够有效降低计算成本，有助于指导高效地设计具有枝晶抑制性能的电池材料体系。

关键词: 扩散系数, 离子电导率, 枝晶描述因子, 相场模拟, 机器学习

Abstract:

During the repeated charging and discharging processes of lithium-ion batteries, the uneven Li⁺ deposition leads to uncontrollable dendrite growth on the electrode surface, severely affecting the battery's safety use. A phase-field simulation is a powerful tool for describing and predicting dendrite growth, while solving partial differential equations that describe the evolution of field variables requires high computational resources. Machine learning approaches that can quickly fit the underlying laws in historical data to predict material performance have been widely used in battery materials. This paper focuses on the effects of lithium-ion transport parameters on dendrite morphology. The corresponding dendrite images were first obtained from the phase-field simulation. The machine learning models were then trained to predict the dendrite metrics (maximum dendrite height and space utilization rate). The results show that the K-nearest neighbors model can accurately capture the relationship between lithium-ion transport parameters and dendrite metrics (R²= 0.995 and 0.992). Meanwhile, the choice of machine learning models and the intervals of dendrite metrics all affect the accuracy of the results. This study can effectively save computational costs and contribute to the effective design of battery materials with dendrite suppression performance.

Key words: diffusion coefficient, ionic conductivity, dendrite metrics, phase-field simulation, machine learning

中图分类号:

TM 911

李亚捷, 王依平, 陈斌, 林海龙, 张更, 施思齐. 机器学习辅助相场模拟预测锂离子输运参数对电池枝晶最大生长高度和空间利用率的影响[J]. 储能科学与技术, 2024, 13(9): 2864-2870.

Yajie LI, Yiping WANG, Bin CHEN, Hailong LIN, Geng ZHANG, Siqi SHI. Machine learning-assisted phase-field simulation for predicting the impact of lithium-ion transport parameters on maximum battery dendrite height and space utilization rate[J]. Energy Storage Science and Technology, 2024, 13(9): 2864-2870.

图/表 6

图1

图2

图3

表1

图4

图5

参考文献 32

1	CHENG X B, ZHANG R, ZHAO C Z, et al. Toward safe lithium metal anode in rechargeable batteries: A review[J]. Chemical Reviews, 2017, 117(15): 10403-10473. DOI: 10.1021/acs.chemrev.7b00115.
2	CHEN L, ZHANG H W, LIANG L Y, et al. Modulation of dendritic patterns during electrodeposition: A nonlinear phase-field model[J]. Journal of Power Sources, 2015, 300: 376-385. DOI: 10.1016/j.jpowsour.2015.09.055.
3	LIU Z, LI Y S, JI Y Z, et al. Dendrite-free lithium based on lessons learned from lithium and magnesium electrodeposition morphology simulations[J]. Cell Reports Physical Science, 2021, 2(1): 100294. DOI: 10.1016/j.xcrp.2020.100294.
4	HONG Z J, VISWANATHAN V. Phase-field simulations of lithium dendrite growth with open-source software[J]. ACS Energy Letters, 2018, 3(7): 1737-1743. DOI: 10.1021/acsenergylett.8b01009.
5	SHI S Q, GAO J, LIU Y, et al. Multi-scale computation methods: Their applications in lithium-ion battery research and development[J]. Chinese Physics B, 2016, 25(1): 018212. DOI: 10.1088/1674-1056/25/1/018212.
6	施思齐. 电化学储能中的计算、建模与仿真[M]. 北京: 化学工业出版社, 2023: 11.
	SHI S Q. Calculation, modeling and simulation in electrochemical energy storage[M]. Beijing: Chemical Industry Press, 2023: 11.
7	LIANG L Y, QI Y, XUE F, et al. Nonlinear phase-field model for electrode-electrolyte interface evolution[J]. Physical Review E, Statistical, Nonlinear, and Soft Matter Physics, 2012, 86(5 Pt 1): 051609. DOI: 10.1103/PhysRevE.86.051609.
8	JÄCKLE M, GROß A. Microscopic properties of lithium, sodium, and magnesium battery anode materials related to possible dendrite growth[J]. The Journal of Chemical Physics, 2014, 141(17): 174710. DOI: 10.1063/1.4901055.
9	JING H X, XING H, DONG X L, et al. Nonlinear phase-field modeling of lithium dendritic growth during electrodeposition[J]. Journal of the Electrochemical Society, 2022, 169(3): 032511. DOI: 10.1149/1945-7111/ac5fed.
10	WANG Q, ZHANG G, LI Y J, et al. Application of phase-field method in rechargeable batteries[J]. NPJ Computational Materials, 2020, 6: 176. DOI: 10.1038/s41524-020-00445-w.
11	李亚捷, 张更, 沙立婷, 等. 可充电电池中枝晶问题的相场模拟[J]. 储能科学与技术, 2022, 11(3): 929-938. DOI: 10.19799/j.cnki.2095-4239.2022.0049.
	LI Y J, ZHANG G, SHA L T, et al. Phase-field simulation of dendrite growth in rechargeable batteries[J]. Energy Storage Science and Technology, 2022, 11(3): 929-938. DOI: 10.19799/j.cnki.2095-4239.2022.0049.
12	李亚捷, 施思齐, 张更, 等. 锂离子电池隔膜结构与枝晶形貌分析软件: 2022SR0147340[P]. 2022.
13	李亚捷, 施思齐, 张更, 等. 基于相场模型模拟电池枝晶生长计算程序: 2022SR0147443[P]. 2022.
14	LI Y J, ZHANG G, CHEN B, et al. Understanding the separator pore size inhibition effect on lithium dendrite via phase-field simulations[J]. Chinese Chemical Letters, 2022, 33(6): 3287-3290. DOI: 10.1016/j.cclet.2022.03.065.
15	LI Y J, SHA L T, ZHANG G, et al. Phase-field simulation tending to depict practical electrodeposition process in lithium-based batteries[J]. Chinese Chemical Letters, 2023, 34(2): 107993. DOI: 10.1016/j.cclet.2022.107993.
16	LI Y J, CHEN B, WANG Y P, et al. Inhibiting dendrite growth by customizing electrolyte or separator to achieve anisotropic lithium-ion transport: A phase-field study[J]. Acta Physico Chimica Sinica, 2023: 2305053. DOI: 10.3866/pku.whxb202305053.
17	LI Y J, ZHAO W, ZHANG G, et al. Unified picture on temperature dependence of lithium dendrite growth via phase-field simulation[J]. Energy Material Advances, 2023, 4: DOI: 10.34133/energymatadv.0053.
18	张更, 王巧, 沙立婷, 等. 相场模型及其在电化学储能材料中的应用[J]. 物理学报, 2020, 69(22): 27-39. DOI: 10.7498/aps.69.20201411.
	ZHANG G, WANG Q, SHA L T, et al. Phase-field model and its application in electrochemical energy storage materials[J]. Acta Physica Sinica, 2020, 69(22): 27-39. DOI: 10.7498/aps.69.20201411.
19	LI Y J, SHA L T, LV P L, et al. Influences of separator thickness and surface coating on lithium dendrite growth: A phase-field study[J]. Materials, 2022, 15(22): 7912. DOI: 10.3390/ma15227912.
20	HONG Z J, VISWANATHAN V. Prospect of thermal shock induced healing of lithium dendrite[J]. ACS Energy Letters, 2019, 4(5): 1012-1019. DOI: 10.1021/acsenergylett.9b00433.
21	GAO L T, GUO Z S. Phase-field simulation of Li dendrites with multiple parameters influence[J]. Computational Materials Science, 2020, 183: 109919. DOI: 10.1016/j.commatsci. 2020.109919.
22	YANG C, LI S L, WANG X T, et al. Phase-field simulation of multi-phase interactions in Fe-C peritectic solidification[J]. Computational Materials Science, 2020, 171: 109220. DOI: 10.1016/j.commatsci.2019.109220.
23	LIU Y, GUO B R, ZOU X X, et al. Machine learning assisted materials design and discovery for rechargeable batteries[J]. Energy Storage Materials, 2020, 31: 434-450. DOI: 10.1016/j.ensm.2020.06.033.
24	LIU Y, ZHAO T L, JU W W, et al. Materials discovery and design using machine learning[J]. Journal of Materiomics, 2017, 3(3): 159-177. DOI: 10.1016/j.jmat.2017.08.002.
25	施思齐, 涂章伟, 邹欣欣, 等. 数据驱动的机器学习在电化学储能材料研究中的应用[J]. 储能科学与技术, 2022, 11(3): 739-759. DOI: 10.19799/j.cnki.2095-4239.2022.0051.
	SHI S Q, TU Z W, ZOU X X, et al. Applying data-driven machine learning to studying electrochemical energy storage materials[J]. Energy Storage Science and Technology, 2022, 11(3): 739-759. DOI: 10.19799/j.cnki.2095-4239.2022.0051.
26	LAMA B, SMIRNOVA A L, PAUDEL T R. Enhanced Li-ion diffusivity of LiFePO₄ by Ru doping: Ab initio and machine learning force field results[J]. ACS Applied Energy Materials, 2023, 6(20): 10424-10431. DOI: 10.1021/acsaem.3c01429.
27	MA Y, HAN S X, SUN Y, et al. Improving ionic conductivity of garnet solid-state electrolytes using Gradient boosting regression optimized machine learning[J]. Journal of Power Sources, 2024, 604: 234492. DOI: 10.1016/j.jpowsour.2024.234492.
28	DHAKAL P, SHAH J K. A generalized machine learning model for predicting ionic conductivity of ionic liquids[J]. Molecular Systems Design & Engineering, 2022, 7(10): 1344-1353. DOI: 10.1039/D2ME00046F.
29	JAGAD H D, FU J T, FULLERTON W R, et al. A physics-based model assisted by machine-learning for sodium-ion batteries with both liquid and solid electrolytes[J]. Journal of the Electrochemical Society, 2024, 171(6): 060516. DOI: 10.1149/1945-7111/ad4a11.
30	LIU Y, WU J M, AVDEEV M, et al. Multi-layer feature selection incorporating weighted score-based expert knowledge toward modeling materials with targeted properties[J]. Advanced Theory and Simulations, 2020, 3(2): DOI: 10.1002/adts.201900215.
31	LIU Y, GE X Y, YANG Z W, et al. An automatic descriptors recognizer customized for materials science literature[J]. Journal of Power Sources, 2022, 545: 231946. DOI: 10.1016/j.jpowsour.2022.231946.
32	刘悦, 邹欣欣, 杨正伟, 等. 材料领域知识嵌入的机器学习[J]. 硅酸盐学报, 2022, 50(3): 863-876. DOI: 10.14062/j.issn.0454-5648.20220093.
	LIU Y, ZOU X X, YANG Z W, et al. Machine learning embedded with materials domain knowledge[J]. Journal of the Chinese Ceramic Society, 2022, 50(3): 863-876. DOI: 10.14062/j.issn.0454-5648.20220093.

机器学习模型	最大枝晶生长高度		空间利用率
机器学习模型	R²	RMSE	R²	RMSE
AdaBoost	0.986	0.022	0.976	0.030
bagging	0.994	0.015	0.991	0.019
decision tree	0.983	0.024	0.980	0.028
gradient boosting	0.992	0.016	0.990	0.019
extremely randomized tree	0.995	0.013	0.990	0.020
KNN	0.995	0.013	0.992	0.017
MLP	0.983	0.024	0.979	0.028
random forest	0.994	0.014	0.991	0.019
Ridge	0.973	0.030	0.946	0.045
SVR	0.913	0.054	0.906	0.059

[1]	樊金保, 李娜, 吴宜琨, 贺春旺, 杨乐, 宋维力, 陈浩森. 能源电池单体层级数字孪生技术[J]. 储能科学与技术, 2024, 13(9): 3112-3133.
[2]	邢瑞鹤, 翁素婷, 李叶晶, 张佳怡, 张浩, 王雪锋. AI辅助电池材料表征与数据分析[J]. 储能科学与技术, 2024, 13(9): 2839-2863.
[3]	刘定宏, 董文楷, 李召阳, 张红烛, 齐昕. 基于RUN-GRU-attention模型的实车动力电池健康状态估计方法[J]. 储能科学与技术, 2024, 13(9): 3042-3058.
[4]	周国兵, 许审镇. 锂金属负极固态电解质界面膜形成和生长机理的理论研究进展[J]. 储能科学与技术, 2024, 13(9): 3150-3160.
[5]	李从心, 岳美玲, 李昕彤, 熊庆辉, 刘孝艳. 基于条件神经网络的质子交换膜燃料电池的老化性能预测[J]. 储能科学与技术, 2024, 13(9): 3094-3102.
[6]	刘子玉, 姜泽坤, 邱伟, 徐泉, 牛迎春, 徐春明, 周天航. 人工智能在长时液流电池储能中的应用：性能优化和大模型[J]. 储能科学与技术, 2024, 13(9): 2871-2883.
[7]	谢莹莹, 邓斌, 张与之, 王晓旭, 张林峰. AI for Science时代下的电池平台化智能研发[J]. 储能科学与技术, 2024, 13(9): 3182-3197.
[8]	朱振威, 苗嘉伟, 祝夏雨, 王晓旭, 邱景义, 张浩. 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.
[9]	何智峰, 陶远哲, 胡泳钢, 王其聪, 杨勇. 机器学习强化的电化学阻抗谱技术及其在锂离子电池研究中的应用[J]. 储能科学与技术, 2024, 13(9): 2933-2951.
[10]	李想, 刘德重, 袁开, 陈大鹏. 用于低温锂金属电池的固态电解质技术研究进展[J]. 储能科学与技术, 2024, 13(7): 2327-2347.
[11]	李炳金, 韩晓霞, 张文杰, 曾伟国, 武晋德. 锂离子电池剩余使用寿命预测方法综述[J]. 储能科学与技术, 2024, 13(4): 1266-1276.
[12]	赵争光, 陈振营, 翟光群, 张希, 庄小东. Sc/O掺杂硫化物固态电解质的制备及全固态电池性能[J]. 储能科学与技术, 2023, 12(8): 2412-2423.
[13]	赵悠曼, 黄旸, 熊立, 林海军. 从材料结构特征到功能特征——[J]. 储能科学与技术, 2023, 12(11): 3287-3298.
[14]	李一涛, 沈凯尔, 庞全全. 有机物辅助的硫化物电解质基固态电池[J]. 储能科学与技术, 2022, 11(6): 1902-1918.
[15]	施思齐, 涂章伟, 邹欣欣, 孙拾雨, 杨正伟, 刘悦. 数据驱动的机器学习在电化学储能材料研究中的应用[J]. 储能科学与技术, 2022, 11(3): 739-759.

机器学习辅助相场模拟预测锂离子输运参数对电池枝晶最大生长高度和空间利用率的影响

Machine learning-assisted phase-field simulation for predicting the impact of lithium-ion transport parameters on maximum battery dendrite height and space utilization rate

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 32

相关文章 15

编辑推荐

Metrics

本文评价