面向复合材料带隙预测的两段式集成学习模型构建

doi:10.12066/j.issn.1007-2861.2389

上海大学学报(自然科学版) ›› 2022, Vol. 28 ›› Issue (3): 504-511.doi: 10.12066/j.issn.1007-2861.2389

面向复合材料带隙预测的两段式集成学习模型构建

徐燕¹(), 胡红青², 刘茜², 张玉凤¹, 丁广太², 张惠然²^,³^,⁴

1.上海电力大学数理学院, 上海 201306
2.上海大学计算机工程与科学学院, 上海 200444
3.上海大学材料基因组工程研究院材料信息与数据科学中心, 上海 200444
4.之江实验室, 浙江杭州 311100

收稿日期:2022-04-09 出版日期:2022-06-30 发布日期:2022-05-27
通讯作者: 徐燕 E-mail:xuyan@shiep.edu.cn
作者简介:徐燕(1982—), 女, 博士, 研究方向为钙钛矿材料、机器学习等. E-mail: xuyan@shiep.edu.cn
基金资助:
国家重点研发计划资助项目(2018YFB0704400);云南省重大科技专项资助项目(202002AB080001-2);云南省重大科技专项资助项目(202102AB080019-3);之江实验室科研攻关资助项目(2021PE0AC02);上海张江国家自主创新示范区专项发展资金重大资助项目(ZJ2021-ZD-006)

Two-stage ensemble learning model for predicting band gaps of composites

XU Yan¹(), HU Hongqing², LIU Xi², ZHANG Yufeng¹, DING Guangtai², ZHANG Huiran²^,³^,⁴

1. College of Mathematics and Physics, Shanghai University of Electric Power, Shanghai 201306, China
2. School of Computer Engineering and Science, Shanghai University, Shanghai 200444, China
3. Center of Materials Informatics and Data Science, Materials Genome Institute, Shanghai University, Shanghai 200444, China
4. Zhejiang Laboratory, Hangzhou 311100, Zhejiang, China

Received:2022-04-09 Online:2022-06-30 Published:2022-05-27
Contact: XU Yan E-mail:xuyan@shiep.edu.cn

摘要/Abstract

摘要：

带隙是钙钛矿型复合氧化物材料重要的特征参数, 对材料的物理化学性质起决定性作用, 如导电性能和光电性能等. 为了寻找适合不同应用领域的钙钛矿型材料, 利用机器学习进行带隙预测是一种重要的研究手段. 构建了一个两阶段异质集成学习模型, 在第一阶段使用多种不同的基础机器学习器(回归模型)进行预测; 在第二阶段把对预测结果影响较大的描述子和基础机器学习器进行集成学习.利用该模型对 210 种钙钛矿型复合氧化物材料的带隙进行预测, 并与多种独立的机器学习算法以及不同集成策略模型的预测性能相对比, 评估了本模型的预测性能. 结果表明, 这种两段式的集成学习模型能够更好地学习到材料数据的内在关系, 并具有较好的预测效果和较强的泛化能力.

关键词: 集成学习模型, 组合策略, 带隙预测, 钙钛矿型复合氧化物材料, 泛化能力

Abstract:

The band gap is an important parameter that can affect the physical and chemical properties of perovskite oxide composites, such as their conductivity and photo-electricity. To identify new perovskites for different applications, their band gap should be predicted via machine learning. Herein, a two-stage ensemble learning model that can predict the band gap of perovskite oxide composites is proposed by combining multiple individual base learners via a certain strategy. The first stage involves individual test functions produced by multiple regression learners. All individual base learners and some specific descriptors are aggregated into an ensemble model in the second stage. Subsequently, a dataset comprising the data of 210 ABX$_3$-type perovskites is used to evaluate the proposed ensemble learning model. Results show that the proposed two-stage ensemble methodology can improve the generalization performance. Its successful application on ABX$_3$-type perovskites indicates the effectiveness and practicability of ensemble learning in material research.

Key words: ensemble learning model, combination strategy, band gap predicting, perovskite oxides composite, generalization performance

中图分类号:

TB 331

徐燕, 胡红青, 刘茜, 张玉凤, 丁广太, 张惠然. 面向复合材料带隙预测的两段式集成学习模型构建[J]. 上海大学学报(自然科学版), 2022, 28(3): 504-511.

XU Yan, HU Hongqing, LIU Xi, ZHANG Yufeng, DING Guangtai, ZHANG Huiran. Two-stage ensemble learning model for predicting band gaps of composites[J]. Journal of Shanghai University（Natural Science Edition）, 2022, 28(3): 504-511.

图/表 6

图1

图2

图3

表1

图4

图5

参考文献 20

[1]	李梓进, 王维燕, 李红江, 等. 钙钛矿/晶硅叠层太阳电池关键材料与技术研究进展[J]. 材料导报, 2020, 34(11): 21061-21071.
[2]	杨潇, 曹成龙, 胡书, 等. 混合卤化物钙钛矿中的光致相分离效应[J]. 材料导报, 2022, 16: 1-27.
[3]	Karmakar A, Bernard G, Meldrum A, et al. Tailorable indirect to direct band-gap double perovskites with bright white-light emission: decoding chemical structure using solid-state NMR[J]. J Am Chem Soc, 2020, 142: 10780-10793. doi: 10.1021/jacs.0c02198 pmid: 32426971
[4]	Lu S, Zhou Q, Ouyang Y, et al. Accelerated discovery of stable lead-free hybrid organic-inorganic perovskites via machine learning[J]. Nat Commun, 2018, 9: 1-8. doi: 10.1038/s41467-017-02088-w
[5]	Gu T, Lu W, Bao X, et al. Using support vector regression for the prediction of the band gap and melting point of binary and ternary compound semiconductors[J]. Solid State Sci, 2006, 8(2): 129-136. doi: 10.1016/j.solidstatesciences.2005.10.011
[6]	Jain D, Chaube S, Khullar P, et al. Bulk and surface DFT investigations of inorganic halide perovskites screened using machine learning and materials property databases[J]. Phys Chem Chem Phys, 2019, 21: 19423-19436. doi: 10.1039/C9CP03240A
[7]	Liu H, Cheng J, Dong H, et al. Screening stable and metastable ABO$_3$ perovskites using machine learning and the materials project[J]. Comput Mater Sci, 2020, 177: 109614. doi: 10.1016/j.commatsci.2020.109614
[8]	Pilania G, Mannodi-Kanakkithodi A, Uberuaga B, et al. Machine learning bandgaps of double perovskites[J]. Sci Rep, 2016, 6: 1-10. doi: 10.1038/s41598-016-0001-8
[9]	Dey P, Bible J, Datta S, et al. Informatics-aided bandgap engineering for solar materials[J]. Comput Mater Sci, 2014, 83: 185-95. doi: 10.1016/j.commatsci.2013.10.016
[10]	Lee J, Seko A, Shitara K, et al. Prediction model of band gap for inorganic compounds by combination of density functional theory calculations and machine learning techniques[J]. Phys Rev B, 2016, 93: 1-12. doi: 10.1103/PhysRev.93.1
[11]	周钢, 郭福亮. 集成学习方法研究[J]. 计算技术与自动化, 2018, 37(4): 148-153.
[12]	徐永林, 王香蒙, 李鑫, 等. 基于高通量计算及机器学习的新材料带隙预测[J]. 中国科学: 技术科学, 2019, 49(1): 44-54.
[13]	Andrii T, Zoia D, Ivan I, et al. Sm-co alloys coercivity prediction using stacking heteroge- neous ensemble model[J]. Acta Metallurgica Slovaca, 2021, 27(4): 195-202. doi: 10.36547/ams.27.4.1173
[14]	Wang Z, Han Y, Lin X, et al. An Ensemble Learning Platform for the Large-Scale Exploration of New Double Perovskites[J]. ACS Appl Mater Interfaces, 2022, 14: 717-725. doi: 10.1021/acsami.1c18477
[15]	Liu Y, Wu J, Wang Z, et al. Predicting creep rupture life of Ni-based single crystal superalloys using divide-and-conquer approach based machine learning[J]. Acta Materialia, 2020, 195: 454467.
[16]	Zhang C, Ma Y. Ensemble machine learning: methods and applications[M]. Berlin: Springer, 2012.
[17]	Zhang Z, Tehrani A, Oliynyk A, et al. Finding the next superhard material through ensemble learning[J]. Advanced Materials, 2021, 33: 2020005112.
[18]	Stukowski A. Structure identification methods for atomistic simulations of crystalline materials[J]. Model Simul Mater Sci Eng, 2012, 20: 045021. doi: 10.1088/0965-0393/20/4/045021
[19]	Brendel A. Some comments on the evaluation of transducer performance[J]. Exp Tech, 1983, 7: 20-21.
[20]	闫岭岭, 江峰, 杜军威, 等. 基于混合采样与 Random_Stacking 的软件缺陷预测[J]. 计算机与现代化, 2021, 8: 70-76.

面向复合材料带隙预测的两段式集成学习模型构建

Two-stage ensemble learning model for predicting band gaps of composites

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 20

相关文章 1

编辑推荐

Metrics

本文评价