Composition Design of Light and Low Expansion Glass Curtain Walls Based on Machine Learning

Abstract

Abstract: The data-driving method for glass composition design has become a desirable way to create novel glass materials, thanks to the ongoing development of machine learning algorithms and the progressive collection of glass materials. In the present work, the machine learning approach based on random forest regression algorithm was used to develop prediction model between the composition and performance of glass materials which contains 56 different oxides. In addition, the interpretability was studied by SHAP analysis. The accurate prediction of linear expansion coefficient, density, and elastic modulus in high-dimensional composition space was realized. The obtained models were utilized to forecast about 1.18 million Si-Al-B-Ca-Mg-Na six-component oxide glass composition swiftly. Four preferred borosilicate glass samples were tested. The results show that linear expansion coefficient of sample ranges from 52.00×10^-7℃^-1 to 58.00×10^-7℃^-1, the density ranges from 2.34 g/cm³ to 2.39 g/cm³, and the elastic modulus ranges from 67.00 GPa to 74.00 GPa, respectively. These values are consistent with the predicted values and outperform the applicable standards.

Key words: machine learning, data driving, glass curtain wall, borosilicate oxide glass, composition design

CLC Number:

TB3
TP321

TIAN Jing, HUANG Yiping, MIAO Enxin, LI Yuan, LIU Junbo, ZHANG Bentao, LIU Yong, HAN Gaorong. Composition Design of Light and Low Expansion Glass Curtain Walls Based on Machine Learning[J]. BULLETIN OF THE CHINESE CERAMIC SOCIETY, 2023, 42(7): 2603-2612.

References

[1] 田英良, 孙诗兵. 新编玻璃工艺学[M]. 北京: 中国轻工业出版社, 2009: 69, 86-122.
TIAN Y L, SUN S B. New glass technology[M]. Beijing: China Light Industry Press, 2009: 69, 86-122 (in Chinese).
[2] HONMA T, MAEDA K, NAKANE S, et al. Unique properties and potential of glass-ceramics[J]. Journal of the Ceramic Society of Japan, 2022, 130(8): 545-551.
[3] ROSCH W, BEALL L, MAXON J, et al. Corning ULE glass can meet P-37 specifications[C]//SPIE Advanced Lithography. Proc SPIE 6517, Emerging Lithographic Technologies XI, San Jose, California, USA. 2007, 6517: 622-629.
[4] SABIA R, EDWARDS M J, VANBROCKLIN R, et al. Corning 7972 ULE material for segmented and large monolithic mirror blanks[C]//SPIE Astronomical Telescopes Instrumentation. Proc SPIE 6273, Optomechanical Technologies for Astronomy, Orlando, Florida, USA. 2006, 6273: 12-19.
[5] SMEDSKJAER M M, YOUNGMAN R E, MAURO J C. Principles of Pyrex^ glass chemistry: structure-property relationships[J].Applied Physics A, 2014, 116(2): 491-504.
[6] JACKSON MJ, MILLS B. Thermal expansion of alumino-alkalisilicate and alumino-borosilicate glasses- comparison of empirical models[J]. Journal of Materials Science Letters, 1997, 16(15): 1264-1266.
[7] 干福熹. 现代玻璃科学技术[M]. 上海: 上海科学技术出版社, 1990: 3.
GAN F X. Modern glass science and technology[M]. Shanghai: Shanghai Science and Technology Press, 1990: 3 (in Chinese).
[8] ABD EL-MONEIM A, YOUSSOF I M, SHOAIB M M. Elastic moduli prediction and correlation in SiO₂-based glasses[J]. Materials Chemistry and Physics, 1998, 52(3): 258-262.
[9] ZENG H D, YE F, LI X, et al. Calculation of thermal expansion coefficient of glasses based on topological constraint theory[J]. Chemical Physics Letters, 2016, 662: 268-272.
[10] 曾惠丹, 邓逸凡, 李响, 等. 基于拓扑结构束缚理论的玻璃性质计算方法[J]. 硅酸盐学报, 2018, 46(1): 1-10.
ZENG H D, DENG Y F, LI X, et al. Calculations method for glass properties based on topological constraint theory[J]. Journal of the Chinese Ceramic Society, 2018, 46(1): 1-10 (in Chinese).
[11] LI N, SAKIDJA R, ARYAL S, et al. Densification of a continuous random network model of amorphous SiO₂ glass[J]. Physical Chemistry Chemical Physics, 2014, 16(4): 1500-1514.
[12] MACHACEK J, GEDEON O, LISKA M. Elastic properties of soda-lime silica glass from first principles[J]. Ceramics-Silikaty, 2009, 53(2): 137-140.
[13] ZHANG Q Y, ZHANG W J, WANG W C, et al. Calculation of physical properties of glass via the phase diagram approach[J]. Journal of Non-Crystalline Solids, 2017, 457: 36-43.
[14] JIANG Z H, ZHANG Q Y. The structure of glass: a phase equilibrium diagram approach[J]. Progress in Materials Science, 2014, 61: 144-215.
[15] ZANOTTO E D, COUTINHO F A B. How many non-crystalline solids can be made from all the elements of the periodic table?[J]. Journal of Non-Crystalline Solids, 2004, 347(1/2/3): 285-288.
[16] 董国平, 万天择, 吴敏波, 等. 玻璃基因工程在激光玻璃等光功能玻璃领域的研究进展[J]. 激光与光电子学进展, 2022, 59(15): 1516002.
DONG G P, WAN T Z, WU M B, et al. Recent applications of glass genetic engineering in laser glasses and other advanced optical glasses[J]. Laser & Optoelectronics Progress, 2022, 59(15): 1516002 (in Chinese).
[17] 刘梓葵. 关于材料基因组的基本观点及展望[J]. 科学通报, 2013, 58(35): 3618-3622.
LIU Z K. Basic viewpoints and prospects of material genome project[J]. Chinese Science Bulletin, 2013, 58(35): 3618-3622 (in Chinese).
[18] 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.
[19] 段志强, 裴小龙, 郭庆伟, 等. 多模态混合输入模拟实验过程实现新型Al-Si-Mg系合金设计[J]. 物理学报, 2023, 72(2): 307-317.
DUAN Z Q, PEI X L, GUO Q W, et al. Design of new Al-Si-Mg alloys by multi-modal mixed input simulation experiment[J]. Acta Physica Sinica, 2023, 72(2): 307-317 (in Chinese).
[20] DE PABLO J J, JACKSON N E, WEBB M A, et al. New frontiers for the materials genome initiative[J]. Computational Materials, 2019, 5: 41.
[21] LIU H, FU Z P, YANG K, et al. Machine learning for glass science and engineering: a review[J]. Journal of Non-Crystalline Solids, 2021, 557: 119419.
[22] MAURO J C, TANDIA A, VARGHEESE K D, et al. Accelerating the design of functional glasses through modeling[J]. Chemistry of Materials, 2016, 28(12): 4267-4277.
[23] CASSAR D R, DE CARVALHO A C P L F, ZANOTTO E D. Predicting glass transition temperatures using neural networks[J]. Acta Materialia, 2018, 159: 249-256.
[24] CASSAR D R, MASTELINI S M, BOTARI T, et al. Predicting and interpreting oxide glass properties by machine learning using large datasets[J]. Ceramics International, 2021, 47(17): 23958-23972.
[25] DENG B H. Machine learning on density and elastic property of oxide glasses driven by large dataset[J]. Journal of Non-Crystalline Solids, 2020, 529: 119768.
[26] YANG K, XU X Y, YANG B, et al. Predicting the Young's modulus of silicate glasses using high-throughput molecular dynamics simulations and machine learning[J]. Scientific Reports, 2019, 9: 8739.
[27] HU Y J, ZHAO G, ZHANG M F, et al. Predicting densities and elastic moduli of SiO₂-based glasses by machine learning[J]. Computational Materials, 2020, 6: 25.
[28] RAVINDER R, SRIDHARA K H, BISHNOI S, et al. Deep learning aided rational design of oxide glasses[J]. Materials Horizons, 2020, 7(7): 1819-1827.
[29] EPAM systems. Sciglass database[DB/OL]. (2019-11-06)[2023-05-06]. https://github.com/epam/SciGlass.
[30] LUNDBERG S, LEE S I. A Unified approach to interpreting model predictions[C]// 31st Annual Conference on Neural Information Processing Systems, Long Beach, America, 2017.
[31] BREIMAN L. Random forests[J].Machine Learning, 2001, 45(1): 5-32.
[32] BIAU G, SCORNET E. A random forest guided tour[J].TEST, 2016, 25(2): 197-227.
[33] SCHONLAU M, ZOU R Y. The random forest algorithm for statistical learning[J]. The Stata Journal: Promoting Communications on Statistics and Stata, 2020, 20(1): 3-29.
[34] PEDREGOSA F, VAROQUAUX G, GRAMFORT A. Scikit-learn: machine learning in Python[J]. Journal of Machine Learning Research, 2011, 12: 2825-2830.
[35] 黄依平, 舒众众, 曹志强, 等. 复合澄清剂在TFT基板玻璃熔制过程中的作用机理[J]. 硅酸盐通报, 2021, 40(11): 3799-3806.
HUANG Y P, SHU Z Z, CAO Z Q, et al. Mechanism of compound refining agents for TFT substrate glass melting process[J]. Bulletin of the Chinese Ceramic Society, 2021, 40(11): 3799-3806 (in Chinese).
[36] PEDONE A, MALAVASI G, CORMACK A N, et al. Insight into elastic properties of binary alkali silicate glasses: prediction and interpretation through atomistic simulation techniques[J]. Chemistry of Materials, 2007, 19(13): 3144-3154.
[37] LUO Z W, LEI W C, LIANG H Z, et al. Improving sealing properties of CaO-SrO-Al₂O₃-SiO₂ glass and glass-ceramics for solid oxide fuel cells: effect of La₂O₃ addition[J]. Ceramics International, 2020, 46(11): 17698-17706.
[38] 曾麟, 黄守佳, 林鸿剑, 等. 混合碱效应对Li₂O-Al₂O₃-SiO₂系玻璃结构和热膨胀性能的影响[J]. 硅酸盐通报, 2021, 40(11): 3813-3821.
ZENG L, HUANG S J, LIN H J, et al. Influence of mixed alkali effect on structure and thermal expansion properties of Li₂O-Al₂O₃-SiO₂ glass[J]. Bulletin of the Chinese Ceramic Society, 2021, 40(11): 3813-3821 (in Chinese).
[39] KE Z K, CAO X, SHAN C L, et al. The effect of alkali metal oxide on the properties of borosilicate fireproof glass: structure, thermal properties, viscosity, chemical stability[J]. Ceramics International, 2021, 47(14): 19605-19613.
[40] KILINC E, BELL A M T, BINGHAM P A, et al. Effects of composition and phase relations on mechanical properties and crystallization of silicate glasses[J]. Journal of the American Ceramic Society, 2021, 104(8): 3921-3946.
[41] MASAI H, FUJII Y, KITAMURA N, et al. Relationship between the elastic properties and structure of BaO-TiO₂-GeO₂-SiO₂ glasses[J]. Journal of Non-Crystalline Solids, 2022, 576: 121248.
[42] CUI J D, CAO X, SHI L F, et al. The effect of substitution of Al₂O₃ and B₂O₃ for SiO₂ on the properties of cover glass for liquid crystal display: structure, thermal, visco-elastic, and physical properties[J]. International Journal of Applied Glass Science, 2021, 12(3): 443-456.
[43] NEMBRINI S, KNIG I, WRIGHT M N. The revival of the gini importance?[J]. Bioinformatics, 2018, 34: 3711-3718.
[44] 曹桃云. 基于随机森林的变量重要性研究[J]. 统计与决策, 2022, 38(4): 60-63.
CAO T Y. Study on the importance of variables based on random forest[J]. Statistics & Decision, 2022, 38(4): 60-63 (in Chinese).
[45] LUNDBERG S M, ERION G, CHEN H, et al. From local explanations to global understanding with explainable AI for trees[J]. Nature Machine Intelligence, 2020, 2(1): 56-67.
[46] 雷宁, 姜中宏, 李郁. 碲酸盐激光玻璃的结构分析[J]. 无机材料学报, 1997, 12(3): 381-385.
LEI N, JIANG Z H, LI Y. Structure analysis of tellurite laser glass[J]. Journal of Inorganic Materials, 1997, 12(3): 381-385 (in Chinese).
[47] 罗丽庆, 王海波, 林健, 等. 碲酸盐系统玻璃的研究进展[J]. 光电子技术与信息, 2005(1): 1-6.
LUO L Q, WANG H B, LIN J, et al. Research and progress on the tellurite glasses[J]. Optoelectronic Technology & Information, 2005(1): 1-6 (in Chinese).
[48] GAO C X, ZHAO X X, LI B. Influence of Y₂O₃ on microstructure, crystallization, and properties of MgO-Al₂O₃-SiO₂ glass-ceramics[J]. Journal of Non-Crystalline Solids, 2021, 560: 120728.
[49] 何峰, 房玉, 刘佳, 等. B₂O₃对硼硅酸盐玻璃结构和性能的影响[J]. 武汉理工大学学报, 2012, 34(2): 1-4.
HE F, FANG Y, LIU J, et al. Effect of B₂O₃ on structure and properties of borosilicate glass[J]. Journal of Wuhan University of Technology, 2012, 34(2): 1-4 (in Chinese).
[50] YOSHIDA S, TANAKA H, HAYASHI T, et al. Scratch resistance of sodium borosilicate glass[J]. Journal of the Ceramic Society of Japan, 2001, 109(1270): 511-515.
[51] 中华人民共和国建设部. 玻璃幕墙工程技术规范: JGJ 102—2003[S]. 北京: 中国建筑工业出版社, 2004.
Ministry of Construction of the People's Republic of China. Technical code for glass curtain wall engineering: JGJ 102—2003[S]. Beijing: China Architecture & Building Press, 2004 (in Chinese).