Research and Application of a Data-Driven Intelligent Design Platform for Materials

doi:10.11871/jfdc.issn.2096-742X.2023.02.007

Abstract

Abstract:

[Objective] Under the new scientific paradigm, artificial intelligence (AI) based on big data has provided new methods and perspectives for accelerating new material design and discovery. Pro-viding a useful intelligent design platform for material researchers has significant implications for im-proving the discovery efficiency and performance optimization of new materials. [Methods] This article proposes an overall architecture of a data-driven intelligent design platform for materials, elaborating on the key technologies and related tools used to develop and optimize new materials on the platform. Additionally, the article provides an application case for the platform in material science. [Results] The material intelligent design platform and its application have accelerated the process of new material design and performance optimization, while also providing researchers with an interactive and plugin-based development environment. [Limitations] The characteristics of the material data, such as multiple heterogeneous sources, small sample sizes, and complex relationships, have a particular impact on the training effect of the models. In the future, more exploration will be done on data standardization and small sample training. [Conclusions] The material design platform proposed in this article provides a theoretical basis and demonstration for the transformation of the scientific research paradigm in material science.

Key words: data-driven, artificial intelligence model, feature calculation, materials design platform, scientific paradigm

WANG Zongguo,WAN Meng,CHEN Ziyi,LI Kai,WANG Xiaoguang,LIU Miao,MENG Sheng,WANG Yangang. Research and Application of a Data-Driven Intelligent Design Platform for Materials[J]. Frontiers of Data and Computing, 2023, 5(2): 86-96, https://cstr.cn/32002.14.jfdc.CN10-1649/TP.2023.02.007.

Figures/Tables 7

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

References 31

[1]	赵继成. 材料基因组计划简介[J]. 自然, 2013, 36(2):89-104.
[2]	Anubhav Jain, Shyue Ping Ong, Geoffroy Hautier, et al. Commentary: The Materials Project: A materials genome approach to accelerating materials innovation[J]. APL Materials, 2013, 1:011002. doi: 10.1063/1.4812323
[3]	Stefano Curtarolo, Gua L. W. Hart, Marco Buongiorno Nardelli, et al. The high-throughput highway to comput-ational materials design[J]. Nature Materials, 2013, 12: 191-201. doi: 10.1038/nmat3568 pmid: 23422720
[4]	James E. Saal, Scott Kirklin, Muratahan Aykol, et al. Materials Design and Discovery with High-Throughput Density Functional Theory: The Open Quantum Materials Database (OQMD)[J]. JOM, 2013, 65:1501-1509. doi: 10.1007/s11837-013-0755-4
[5]	Claudia Draxl and Matthias Scheffler. NOMAD: The FAIR concept for big data-driven materials science[J]. MRS Bulletin, 2018, 43(9): 676-682. doi: 10.1557/mrs.2018.208
[6]	Leopold Talirz, Snehal Kumbhar, Elsa Passaro, et al. Materials Cloud, a platform for open computational sci-ence[J]. Scientific Data, 2020, 7: 299. doi: 10.1038/s41597-020-00637-5 pmid: 32901046
[7]	Giovanni Pizzi, Andrea Cepellotti, Riccardo Sabatini, et al. AiiDA: automated interactive infrastructure and data-base for computational science[J]. Computational Mater-ials Science, 2016, 111: 218-230.
[8]	Xiaoyu Yang, Zongguo Wang, Xushan Zhao, et al. MatCloud: A high-throughput computational infrast-ructure for integrated management of materials simul-ation, data and resources[J]. Computational Materials Science, 2018, 146:319-333. doi: 10.1016/j.commatsci.2018.01.039
[9]	刘淼, 孟胜. Atomly.net数据平台及其在无机化学中的应用[J]. 中国科学: 化学, 2023, 53(1): 19-25.
[10]	Lili Xi, Shanshan Pan, Xin Li, et al. Discovery of high-performance thermoelectric chalcogenides through reli-able high-throughput material screening[J]. Journal of the American Chemical Society, 2018, 140(34): 10785-10793. doi: 10.1021/jacs.8b04704 pmid: 30086638
[11]	Paul Raccuglia, Katherine C. Elbert, Philip D. F. Adler, et al. Machine-learning-assisted materials discovery using failed experiments[J]. Nature, 2016, 533:73-76. doi: 10.1038/nature17439
[12]	Yiming Chen, Chi Chen, Chen Zhang, et al. Database of ab initio L-edge X-ray absorption near edge structure[J]. Scientific Data, 2021, 8:153. doi: 10.1038/s41597-021-00936-5 pmid: 34117266
[13]	Nathan C. Frey, Matthew K. Horton, Jason M. Munro, et al. High-throughput search for magnetic and topological order in transition metal oxides[J]. Science Advances, 2020, 6: eabd1076. doi: 10.1126/sciadv.abd1076
[14]	Yilei Wu, Shuaihua Lu, Minggang Ju, et al. Accelerated design of promising mixed lead-free double halide organ-ic-inorganic perovskites for photovoltaics using machine learning[J]. Nanoscale, 2021, 13: 12250-12259. doi: 10.1039/d1nr01117k pmid: 34241606
[15]	Yang Liu, Haiyou Huang, Jie Yuan, et al. Upper limit of the transition temperature of superconducting materials[J]. Patterns, 2022, 3:100609. doi: 10.1016/j.patter.2022.100609
[16]	Ze Yu, Tao Bo, Bo Liu, et al. Superconductive materials with MgB2-like structures from data-driven screening[J]. Physical Review B, 2022, 105:214517. doi: 10.1103/PhysRevB.105.214517
[17]	Wei Wu and Qiang Sun. Screening Topological Quantum Materials for Na-Ion Battery Cathode[J]. ACS Materials Letters, 2022, 4(1): 175-180. doi: 10.1021/acsmaterialslett.1c00545
[18]	Xia Cai, Yiming Zhang, Zejiao Shi, et al. Discovery of Lead-Free Perovskites for High-Performance Solar Cells via Machine Learning: Ultrabroadband Absorption, Low Radiative Combination, and Enhanced Thermal Condu-ctivities[J]. Advanced Science, 2022, 9: 2103648. doi: 10.1002/advs.v9.4
[19]	Rupp M, Tkatchenko A, Muller KR, von Lilienfeld OA. Fast and accurate modeling of molecular atomization energies with machine learning[J]. Phys Rev Lett, 2012, 108:058301. doi: 10.1103/PhysRevLett.108.058301
[20]	R. Carhart, Dennis H. Smith, and R. Venkataraghavan. Atom pairs as molecular features in structure-activity studies: definition and applications[J]. Journal of Chemi-cal Information and Computer Sciences, 1985, 25:64-73.
[21]	Jörg Behler. Atom-centered symmetry functions for constructing high-dimensional neural network potentials[J]. Journal of Chemical Physics, 2011, 134(7):074106. doi: 10.1063/1.3553717
[22]	Artrith N, Urban A, Ceder G. Efficient and accurate ma-chine-learning interpolation of atomic energies in comp-ositions with many species[J]. Physical Review B, 2017, 96(1): 1-7.
[23]	Han Wang, Linfeng Zhang, Jiequn Han, Weinan E. DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics[J]. Computer Physics Communications, 2018, 228:178-184. doi: 10.1016/j.cpc.2018.03.016
[24]	Xie T. and J.C. Grossman, Crystal graph convolutional neural networks for an accurate and interpretable pre-diction of material properties[J]. Physical review letters, 2018, 120(14): 145301. doi: 10.1103/PhysRevLett.120.145301
[25]	Li-Yuan Xue, Feng Guo, Yu-Shi Wen, et al. Reaxff-mpnn machine learning potential: a combination of reactive force field and message passing neural networks[J]. Physical Chemistry Chemical Physics, 2021, 23:19457-19464. doi: 10.1039/d1cp01656c pmid: 34524283
[26]	Zhenqin Wu, Bharath Ramsundar, Evan N. Feinberg, et al. MoleculeNet: A benchmark for molecular machine learning[J]. Chemical Science, 2018, 9:513-530. doi: 10.1039/c7sc02664a pmid: 29629118
[27]	Chi Chen, Weike Ye, Yunxing Zuo, et al. Graph net-works as a universal machine learning framework for molecules and crystals[J]. Chemistry of Materials, 2019, 31(9): 3564-3572. doi: 10.1021/acs.chemmater.9b01294
[28]	Jianjun Hu, Stanislav Stefanov, Yuqi Song, et al. Materia-lsAtlas.org: a materials informatics web app platform for materials discovery and survey of state-of-the art[J]. npj Computational Materials, 2022, 8:65. doi: 10.1038/s41524-022-00750-6
[29]	Zhiwei Liu, Jialong Guo, Ziyi Chen, et al. Swarm intelli-gence for new materials[J]. Computational Materials Science, 2022, 214:111699. doi: 10.1016/j.commatsci.2022.111699
[30]	Yingzong Liang, Mingwei Chen, Yanan Wang, et al. A universal model for accurately predicting the formation energy of inorganic compounds[J]. Science China Mat-erials, 2023, 66(1):343-351.
[31]	Jialong Guo, Ziyi Chen, Zhiwei Liu, et al. Neural net-work training method for materials science based on multi-source databases[J]. Scientific Reports, 2022, 12: 15326. doi: 10.1038/s41598-022-19426-8 pmid: 36096926