机器学习辅助材料筛选研究进展

doi:10.16085/j.issn.1000-6613.2024-1978

摘要/Abstract

摘要：

机器学习凭借其高效性和准确性，正在革新传统模式的材料筛选和设计，极大地推动了材料科学的快速发展。本文综述了机器学习在材料筛选中的应用，探讨了其在不同材料体系中的策略和方法。首先，介绍了机器学习在材料科学中的一般工作流程；然后详细讨论了机器学习在电池材料、热电材料、催化材料以及合金材料等领域的应用案例，展示了机器学习如何加速材料的发现和优化过程。此外，本文还探讨了现阶段机器学习在材料领域应用中面临的一些挑战，指出目前材料领域数据的缺乏仍然是机器学习在材料领域应用的一大难题，以及一些新型的基于神经网络的模型无法适用于小型数据集且其缺乏直观的解释机制。未来随着高通量技术的不断进步，数据获取将更加高效，为机器学习在材料领域的应用提供更丰富的数据支持。与此同时，深度学习与迁移学习等技术的持续发展将大幅提升材料智能模型的泛化能力和预测精度，推动材料研发向智能化、精准化方向迈进。

关键词: 机器学习, 材料筛选, 数据驱动, 智能模型, 预测精度

Abstract:

Machine learning, with its high efficiency and accuracy, is revolutionizing the traditional models of material screening and design, greatly promoting the rapid development of materials science. This article reviewed the application of machine learning in material screening and discussed its strategies and methods in different material systems. Firstly, the general workflow of machine learning in materials science was introduced, and then, detailed discussions were given on the application cases of machine learning in battery materials, thermoelectric materials, catalytic materials and alloy materials, showing how machine learning accelerated the discovery and optimization process of materials. Additionally, this article also explored some challenges faced by the current application of machine learning in the field of materials, pointing out that the lack of data in the materials field remained a major obstacle for the application of machine learning in this area, as well as the fact that some new neural network-based models were not suitable for small datasets and lack intuitive interpretation mechanisms. In the future, with the continuous advancement of high-throughput technology, data acquisition would become more efficient, providing richer data support for the application of machine learning in the field of materials. Meanwhile, the ongoing development of technologies such as deep learning and transfer learning would significantly enhance the generalization ability and prediction accuracy of material intelligent models, and promote the research of materials in the direction of intelligence and precision.

Key words: machine learning, materials screening, data-driven, intelligent model, prediction accuracy

中图分类号:

黄燕, 嵇海宁, 刘东青. 机器学习辅助材料筛选研究进展[J]. 化工进展, 2025, 44(12): 7019-7033.

HUANG Yan, JI Haining, LIU Dongqing. Research advances in screening of machine learning-assisted materials[J]. Chemical Industry and Engineering Progress, 2025, 44(12): 7019-7033.

图/表 11

图1 机器学习在材料领域应用的一般流程框架

图2 电池结构示意图[50]

图3 模型预测掺杂NCM正极的首次放电容量和第50个循环末放电容量的预测值与实验值的拟合散点图[57]

图4 模型框架与预测效果图[63]

图5 各模型在总声子带中心预测上的性能对比[65]

图6 各模型对热电材料的分类效果对比[67]

图7 ANN模型预测值与真实值拟合图[68]

图8 催化剂筛选流程图[74]

图9 机器学习在高强度铝合金筛选上的应用[81]

图10 各模型对硬度和电导率的预测效果[82]

图11 高熵合金主动学习筛选流程图[83]

参考文献 89

[1]	AGRAWAL Ankit, CHOUDHARY Alok. Perspective: Materials informatics and big data: Realization of the “fourth paradigm” of science in materials science[J]. APL Materials, 2016, 4(5): 053208.
[2]	CHEN CHI, ZUO YUNXING, YE WEIKE, et al. A critical review of machine learning of energy materials[J]. Advanced Energy Materials, 2020, 10(8): 1903242.
[3]	LU Mingying, JI Haining, ZHAO Yong, et al. Machine learning-assisted synthesis of two-dimensional materials[J]. ACS Applied Materials & Interfaces, 2023, 15(1): 1871-1878.
[4]	WANG Jingting, LU Mingying, CHEN Yongxing, et al. Machine learning-assisted large-area preparation of MoS₂ materials[J]. Nanomaterials, 2023, 13(16): 2283.
[5]	XIONG Gaoyang, JI Haining, CHEN Yongxing, et al. Preparation of thermochromic vanadium dioxide films assisted by machine learning[J]. Nanomaterials, 2024, 14(13): 1153.
[6]	CHEN Yongxing, JI Haining, LU Mingying, et al. Machine learning guided hydrothermal synthesis of thermochromic VO₂ nanoparticles[J]. Ceramics International, 2023, 49(18): 30794-30800.
[7]	BELLE Carl E, AKSAKALLI Vural, RUSSO Salvy P. A machine learning platform for the discovery of materials[J]. Journal of Cheminformatics, 2021, 13(1): 42.
[8]	KALIDINDI Surya R, BROUGH David B, LI Shengyen, et al. Role of materials data science and informatics in accelerated materials innovation[J]. MRS Bulletin, 2016, 41(8): 596-602.
[9]	SUN Zhehao, YIN Hang, YIN Zongyou. Leveraging machine learning in the innovation of functional materials[J]. Matter, 2023, 6(8): 2553-2555.
[10]	封俊杰, 周琨, 李木琛, 等. 数据驱动研究范式下材料数据库的构建与应用[J]. 科学通报, 2025, 70(24): 4044-4065.
	FENG Junjie, ZHOU Kun, LI Muchen, et al. Construction and application of material database under data-driven research paradigm[J]. Chinese Science Bulletin, 2025, 70(24): 4044-4065.
[11]	BUTLER Keith T, DAVIES Daniel W, CARTWRIGHT Hugh, et al. Machine learning for molecular and materials science[J]. Nature, 2018, 559: 547-555.
[12]	WEI Jing, CHU Xuan, SUN Xiangyu, et al. Machine learning in materials science[J]. InfoMat, 2019, 1(3): 338-358.
[13]	CHEN Yongxing, LONG Peng, LIU Bin, et al. Development and application of Few-shot learning methods in materials science under data scarcity[J]. Journal of Materials Chemistry A, 2024, 12(44): 30249-30268.
[14]	WANG Anthony Yu-Tung, MURDOCK Ryan J, KAUWE Steven K, et al. Machine learning for materials scientists: An introductory guide toward best practices[J]. Chemistry of Materials, 2020, 32(12): 4954-4965.
[15]	TAO Qiuling, YU Jinxin, MU Xiangyu, et al. Machine learning strategies for small sample size in materials science[J]. Science China Materials, 2025, 68(2): 387-405.
[16]	BOONPALIT Kajjana, KINCHAGAWAT Jiramet, NAMUANGRUK Supawadee. Expanding the applicability domain of machine learning model for advancements in electrochemical material discovery[J]. ChemElectroChem, 2024, 11(10): e202300681.
[17]	RAMPRASAD Rampi, BATRA Rohit, PILANIA Ghanshyam, et al. Machine learning in materials informatics: Recentapplications and prospects[J]. NPJ Computational Materials, 2017, 3: 54.
[18]	何林, 贺常晴, 隋红. 人工智能驱动新型界面分离材料的创制[J]. 化工进展, 2024, 43(4): 1649-1654.
	HE Lin, HE Changqing, SUI Hong. Prospects for the creation of novel interfacial separation materials driven by artificial intelligence[J]. Chemical Industry and Engineering Progress, 2024, 43(4): 1649-1654.
[19]	MORGAN Dane, JACOBS Ryan. Opportunities and challenges for machine learning in materials science[J]. Annual Review of Materials Research, 2020, 50: 71-103.
[20]	HUANG Guannan, GUO Yani, CHEN Ye, et al. Application of machine learning in material synthesis and property prediction[J]. Materials, 2023, 16(17): 5977.
[21]	FANG Jiheng, XIE Ming, HE Xingqun, et al. Machine learning accelerates the materials discovery[J]. Materials Today Communications, 2022, 33: 104900.
[22]	JAIN Anubhav, Shyue Ping ONG, HAUTIER Geoffroy, et al. Commentary: The Materials Project: A materials genome approach to accelerating materials innovation[J]. APL Materials, 2013, 1(1): 011002.
[23]	BELSKY Alec, HELLENBRANDT Mariette, KAREN Vicky Lynn, et al. New developments in the inorganic crystal structure database (ICSD): Accessibility in support of materials research and design[J]. Acta Crystallographica Section B, 2002, 58: 364-369.
[24]	KIRKLIN Scott, SAAL James E, MEREDIG Bryce, et al. The open quantum materials database (OQMD): Assessing the accuracy of DFT formation energies[J]. NPJ Computational Materials, 2015, 1: 15010.
[25]	DEVLIN Jacob, CHANG Mingwei, LEE Kenton, et al. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding[C]//BURSTEIN Jill, DORAN Christy, SOLORIO Thamar. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). Minnesota:Association for Computational Linguistics, 2019:4171-4186.
[26]	ZHANG Lei, HUANG Yiru, YAN Leiming, et al. Fast exploring literature by language machine learning for perovskite solar cell materials design[J]. Advanced Intelligent Systems, 2024, 6(6): 2300678.
[27]	LI Runzhao, HERREROS Jose Martin, TSOLAKIS Athanasios, et al. Machine learning-quantitative structure property relationship (ML-QSPR) method for fuel physicochemical properties prediction of multiple fuel types[J]. Fuel, 2021, 304: 121437.
[28]	ROY Kunal, KAR Supratik, Rudra Narayan DAS. A primer on QSAR/QSPR modeling: Fundamental concepts[M]. Cham: Springer International Publishing, 2015.
[29]	何婷. 基于机器学习结合定量构效关系(QSPR)的含能材料爆轰性能预估及筛选方法研究[D]. 西安: 西北大学, 2021.
	HE Ting. Study on prediction and screening method of detonation performance of energetic materials based on machine learning and quantitative structure-activity relationship (QSPR)[D]. Xi’an: Northwest University, 2021.
[30]	WARD Logan, DUNN Alexander, FAGHANINIA Alireza, et al. Matminer: An open source toolkit for materials data mining[J]. Computational Materials Science, 2018, 152: 60-69.
[31]	HIMANEN Lauri, JÄGER Marc O J, MOROOKA Eiaki V, et al. DScribe: Library of descriptors for machine learning in materials science[J]. Computer Physics Communications, 2020, 247: 106949.
[32]	WINES Daniel, GURUNATHAN Ramya, GARRITY Kevin F, et al. Recent progress in the JARVIS infrastructure for next-generation data-driven materials design[J]. Applied Physics Reviews, 2023, 10(4): 041302.
[33]	MUTHUKRISHNAN R, ROHINI R. LASSO: A feature selection technique in predictive modeling for machine learning[C]//2016 IEEE International Conference on Advances in Computer Applications (ICACA). Coimbatore: IEEE, 2016: 18-20.
[34]	BREIMAN Leo. Random forests[J]. Machine Learning, 2001,45(1): 5-32.
[35]	CHEN Tianqi, GUESTRIN Carlos. XGBoost: A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco: ACM, 2016: 785-794.
[36]	KE Guolin, MENG Qi, FINLEY Thomas, et al. LightGBM: A highly efficient gradient boosting decision tree[C]//LUXBURG Ulrike Von, GUYON Isabelle, BENGIO Samy, WALLACH Hanna, FERGUS Rob. Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, 2017: 3149-3157.
[37]	THENG Dipti, BHOYAR Kishor K. Feature selection techniques for machine learning: A survey of more than two decades of research[J]. Knowledge and Information Systems, 2024, 66(3): 1575-1637.
[38]	ARLOT Sylvain, CELISSE Alain. A survey of cross-validation procedures for model selection[J]. Statistics Surveys, 2010, 4: 40-79.
[39]	EFRON Bradley, TIBSHIRANI Robert. Improvements on cross-validation: The 632+ bootstrap method[J]. Journal of the American Statistical Association, 1997, 92(438): 548-560.
[40]	POWERS David M W. Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation[EB/OL]. 2020: 2010.16061. .
[41]	FAWCETT Tom. An introduction to ROC analysis[J]. Pattern Recognition Letters, 2006, 27(8): 861-874.
[42]	PEDREGOSA Fabian, VAROQUAUX Gaël, GRAMFORT Alexandre, et al. Scikit-learn: Machine learning in Python[J]. Journal of Machine Learning Research, 2011, 12: 2825-2830.
[43]	李金栋, 郝永勤, 孙旭, 等. 机器学习在永磁材料研究中的应用进展[J]. 功能材料, 2025, 56(1): 1064-1074.
	LI Jindong, HAO Yongqin, SUN Xu, et al. Application of machine learning in permanent magnetic materials[J]. Journal of Functional Materials, 2025, 56(1): 1064-1074.
[44]	章本本, 缪林昌, 郑海忠, 等. 机器学习在声学超材料中的应用进展[J]. 振动与冲击, 2024, 43(23): 280-293.
	ZHANG Benben, MIAO Linchang, ZHENG Haizhong, et al. Application Progress of machine learning in acoustic metamaterials[J]. Journal of Vibration and Shock, 2024, 43(23): 280-293.
[45]	王璐, 张磊. 机器学习高效筛选用于CO₂/N₂选择性吸附分离的沸石材料[J]. 化工进展, 2023, 42(1):148-158.
	WANG Lu, ZHANG Lei. Lei. High-throughput screening of zeolite materials for CO₂/N₂ selective adsorption separation by machine learning[J]. Chemical Industry and Engineering Progress, 2023, 42(1): 148-158.
[46]	KIM Jaehyun, KANG Donghoon, KIM Sangbum, et al. Catalyze materials science with machine learning[J]. ACS Materials Letters, 2021, 3(8): 1151-1171.
[47]	WANG Ziman, YANG Ming, XIE Xixi, et al. Applications of machine learning in perovskite materials[J]. Advanced Composites and Hybrid Materials, 2022, 5(4): 2700-2720.
[48]	JUAN Yongfei, DAI Yongbing, YANG Yang, et al. Accelerating materials discovery using machine learning[J]. Journal of Materials Science & Technology, 2021, 79: 178-190.
[49]	MANTHIRAM Arumugam, FU Yongzhu, SU Yusheng. Challenges and prospects of lithium-sulfur batteries[J]. Accounts of Chemical Research, 2013, 46(5): 1125-1134.
[50]	GOODENOUGH John B, PARK Kyu-Sung. The Li-ion rechargeable battery: A perspective[J]. Journal of the American Chemical Society, 2013, 135(4): 1167-1176.
[51]	LU Jun, CHEN Zhongwei, PAN Feng, et al. High-performance anode materials for rechargeable lithium-ion batteries[J]. Electrochemical Energy Reviews, 2018, 1(1): 35-53.
[52]	SEVERSON Kristen A, ATTIA Peter M, JIN Norman, et al. Data-driven prediction of battery cycle life before capacity degradation[J]. Nature Energy, 2019, 4(5): 383-391.
[53]	AYKOL Muratahan, HERRING Patrick, ANAPOLSKY Abraham. Machine learning for continuous innovation in battery technologies[J]. Nature Reviews Materials, 2020, 5(10): 725-727.
[54]	ALLAM Omar, CHO Byung Woo, KIM Ki Chul, et al. Application of DFT-based machine learning for developing molecular electrode materials in Li-ion batteries[J]. RSC Advances, 2018, 8(69): 39414-39420.
[55]	JOSHI Rajendra P, EICKHOLT Jesse, LI Liling, et al. Machine learning the voltage of electrode materials in metal-ion batteries[J]. ACS Applied Materials & Interfaces, 2019, 11(20): 18494-18503.
[56]	MOSES Isaiah A, JOSHI Rajendra P, OZDEMIR Burak, et al. Machine learning screening of metal-ion battery electrode materials[J]. ACS Applied Materials & Interfaces, 2021, 13(45): 53355-53362.
[57]	WANG Guanyu, FEARN Tom, WANG Tengyao, et al. Machine-learning approach for predicting the discharging capacities of doped lithium nickel-cobalt-manganese cathode materials in Li-ion batteries[J]. ACS Central Science, 2021, 7(9): 1551-1560.
[58]	WANG Guanyu, FEARN Tom, WANG Tengyao, et al. Insight gained from using machine learning techniques to predict the discharge capacities of doped spinel cathode materials for lithium-ion batteries applications[J]. Energy Technology, 2021, 9(5): 2100053.
[59]	PARK Sohyun, PARK Sunhyeon, PARK Young, et al. A new material discovery platform of stable layered oxide cathodes for K-ion batteries[J]. Energy & Environmental Science, 2021, 14(11): 5864-5874.
[60]	RANSOM Brandi, ZHAO Nathan, SENDEK Austin D, et al. Two low-expansion Li-ion cathode materials with promising multi-property performance[J]. MRS Bulletin, 2021, 46(12): 1116-1129.
[61]	LIU Xiaoxu, WANG Tian, JI Tianyi, et al. Using machine learning to screen non-graphite carbon materials based on Na-ion storage properties[J]. Journal of Materials Chemistry A, 2022, 10(14): 8031-8046.
[62]	ZHANG Xin, DING Bin, WANG Yao, et al. Machine learning for screening small molecules as passivation materials for enhanced perovskite solar cells[J]. Advanced Functional Materials, 2024, 34(30): 2314529.
[63]	TANG Le, ZHANG Guozhen, JIANG Jun. Machine learning approach accelerates search for solid state electrolytes[J]. Chinese Journal of Chemical Physics, 2024, 37(4): 505-512.
[64]	GUO Xingyu, WANG Zhenbin, YANG Jihui, et al. Machine-learning assisted high-throughput discovery of solid-state electrolytes for Li-ion batteries[J]. Journal of Materials Chemistry A, 2024, 12(17): 10124-10136.
[65]	JAAFREH Russlan, PEREZNIETO Santiago, JEONG Seonghun, et al. Phonon DOS-based machine learning model for designing high-performance solid electrolytes in Li-ion batteries[J]. International Journal of Energy Research, 2024, 2024(1): 2138847.
[66]	WANG Xiangdong, SHENG Ye, NING Jinyan, et al. A critical review of machine learning techniques on thermoelectric materials[J]. The Journal of Physical Chemistry Letters, 2023, 14(7): 1808-1822.
[67]	FAN Tao, OGANOV Artem R. Combining machine-learning models with first-principles high-throughput calculations to accelerate the search for promising thermoelectric materials[J]. Journal of Materials Chemistry C, 2025, 13(3): 1439-1448.
[68]	TIRYAKI Hasan, YUSUF Aminu, BALLIKAYA Sedat. Determination of electrical and thermal conductivities of n- and p-type thermoelectric materials by prediction iteration machine learning method[J]. Energy, 2024, 292: 130597.
[69]	Mohammed AL-FAHDI, YUAN Kunpeng, YAO Yagang, et al. High-throughput thermoelectric materials screening by deep convolutional neural network with fused orbital field matrix and composition descriptors[J]. Applied Physics Reviews, 2024, 11(2): 021402.
[70]	VAITESSWAR U S, BASH Daniil, HUANG Tan, et al. Machine learning based feature engineering for thermoelectric materials by design[J]. Digital Discovery, 2024, 3(1): 210-220.
[71]	PARSE Nuttawat, PINITSOONTORN Supree. Machine learning for predicting ZT values of high-performance thermoelectric materials in mid-temperature range[J]. APL Materials, 2023, 11(8): 081117.
[72]	ZHOU Pan, WANG Ming, TANG Fei, et al. Machine learning accelerates the screening of efficient metal-oxide catalysts for photocatalytic water splitting[J]. Materials Research Bulletin, 2024, 179: 112956.
[73]	SUN Jikai, TU Rui, XU Yuchun, et al. Machine learning aided design of single-atom alloy catalysts for methane cracking[J]. Nature Communications, 2024, 15(1): 6036.
[74]	GOLDER Rahul, Shraman PAL, SATHISH KUMAR C, et al. Machine learning-enhanced optimal catalyst selection for water-gas shift reaction[J]. Digital Chemical Engineering, 2024, 12: 100165.
[75]	Moses ABRAHAM B, SINHA Priyanka, HALDER Prosun, et al. Fusing a machine learning strategy with density functional theory to hasten the discovery of 2D MXene-based catalysts for hydrogen generation[J]. Journal of Materials Chemistry A, 2023, 11(15): 8091-8100.
[76]	ZHANG Jingzi, WANG Yuelin, ZHOU Xuyan, et al. Accurate and efficient machine learning models for predicting hydrogen evolution reaction catalysts based on structural and electronic feature engineering in alloys[J]. Nanoscale, 2023, 15(26): 11072-11082.
[77]	WANG Song, JIANG Jun. Interpretable catalysis models using machine learning with spectroscopic descriptors[J]. ACS Catalysis, 2023, 13(11): 7428-7436.
[78]	YIN Peng, NIU Xiangfu, LI Shuobin, et al. Machine-learning-accelerated design of high-performance platinum intermetallic nanoparticle fuel cell catalysts[J]. Nature Communications, 2024, 15(1): 415.
[79]	ZHANG Sheng, LU Shuaihua, ZHANG Peng, et al. Accelerated discovery of single-atom catalysts for nitrogen fixation via machine learning[J]. Energy & Environmental Materials, 2023, 6(1): e12304.
[80]	HU Mingwei, TAN Qiyang, KNIBBE Ruth, et al. Recent applications of machine learning in alloy design: A review[J]. Materials Science and Engineering: R: Reports, 2023, 155: 100746.
[81]	LI Jiaheng, ZHANG Yingbo, CAO Xinyu, et al. Accelerated discovery of high-strength aluminum alloys by machine learning[J]. Communications Materials, 2020, 1: 73.
[82]	PAN Shaobin, WANG Yongjie, YU Jinxin, et al. Accelerated discovery of high-performance Cu-Ni-Co-Si alloys through machine learning[J]. Materials & Design, 2021, 209: 109929.
[83]	RAO Ziyuan, TUNG Po-Yen, XIE Ruiwen, et al. Machine learning-enabled high-entropy alloy discovery[J]. Science, 2022, 378(6615): 78-85.
[84]	WEN Cheng, SHEN Haicheng, TIAN Yuwan, et al. Accelerated discovery of refractory high-entropy alloys for strength-ductility co-optimization: An exploration in NbTaZrHfMo system by machine learning[J]. Scripta Materialia, 2024, 252: 116240.
[85]	YAN Yonggang, LU Dan, WANG Kun. Accelerated discovery of single-phase refractory high entropy alloys assisted by machine learning[J]. Computational Materials Science, 2021, 199: 110723.
[86]	PRIYA Pikee, ALURU N R. Accelerated design and discovery of perovskites with high conductivity for energy applications through machine learning[J]. NPJ Computational Materials, 2021, 7: 90.
[87]	WU Mengfan, TIKHONOV Evgenii, TUDI Abudukadi, et al. Target-driven design of deep-UV nonlinear optical materials via interpretable machine learning[J]. Advanced Materials, 2023, 35(23): 2300848.
[88]	CHEN Fengqing, WENG Longjie, WANG Jinhe, et al. An adaptive framework to accelerate optimization of high flame retardant composites using machine learning[J]. Composites Science and Technology, 2023, 231: 109818.
[89]	FAN Qingyang, MIN Gege, LIU Li, et al. Accelerate the design of new superhard carbon allotropes in Pca21 space group: High-throughput screening and machine learning strategies[J]. Diamond and Related Materials, 2024, 143: 110928.