数据驱动的铝基复合材料性能预测和逆向设计

doi:10.12066/j.issn.1007-2861.2386

上海大学学报(自然科学版) ›› 2022, Vol. 28 ›› Issue (3): 512-522.doi: 10.12066/j.issn.1007-2861.2386

数据驱动的铝基复合材料性能预测和逆向设计

陈水洲¹, 王晓书², 欧阳求保², 张瑞¹^,³^,⁴()

1.上海大学计算机工程与科学学院, 上海 200444
2.上海交通大学材料科学与工程学院金属基复合材料国家重点实验室, 上海 200240
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:2022-03-30 出版日期:2022-06-30 发布日期:2022-05-27
通讯作者: 张瑞 E-mail:ruizhang@shu.edu.cn
作者简介:张瑞(1981—), 女, 副教授, 博士, 研究方向为材料信息学、计算机网络、网络安全等. E-mail: ruizhang@shu.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)

Data-driven based properties prediction and reverse design of aluminum matrix composites

CHEN Shuizhou¹, WANG Xiaoshu², OUYANG Qiubao², ZHANG Rui¹^,³^,⁴()

1. School of Computer Engineering and Science, Shanghai University, Shanghai 200444, China
2. State Key Laboratory of Metal Matrix Composites, School of Material Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, 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:2022-03-30 Online:2022-06-30 Published:2022-05-27
Contact: ZHANG Rui E-mail:ruizhang@shu.edu.cn

摘要/Abstract

摘要：

采用数据驱动的方法对 SiCp(0.5CNT)/7075Al 铝基复合材料的化学成分以及制备工艺进行了分析, 针对抗拉强度和延伸率两个力学性能进行了特征重要性分析, 构建了包含 8 种机器学习算法的集成框架, 自动进行模型的参数调优和最优模型选择, 并在此基础上进行了材料逆向设计. 实验结果表明, 在 470 ${^\circ}$C 固溶 40 min, 120${^\circ}$C 时效 15 h 的热处理工艺下, SiCp(0.5CNT)/7075Al-1.0Mg 复合材料抗拉强度和延伸率的预测值为 617.48 MPa 和 2.98%, 实验值为 647.0 MPa 和 3.31%, 两项物理性能的平均绝对百分比误差(mean absolute percentage errors, MAPE)较小, 依次为 4.56% 和 9.97%. 这说明本数据驱动方法对铝基复合材料的工艺优化和性能提升有一定指导意义.

关键词: 机器学习, 特征分析, 铝基复合材料, 逆向设计

Abstract:

A data-driven approach was used to analyze the chemical composition and preparation process of aluminum matrix composites SiCp(0.5CNT)/7075Al, and analyze the tensile strength and elongation. An integrated framework comprising eight machine learning algorithms was constructed to automatically perform parameters tuning and optimal model selection. Subsequently, an inverse design of the material was conducted. Experimental results showed that under the heat treatment of a solid solution at 470 ${^\circ}$C for 40 min and aging at 120 ${^\circ}$C for 15 h, the predicted tensile strength and elongation of SiCp(0.5CNT)/7075Al-1.0Mg were 617.48 MPa and 2.98%, respectively, whereas the real experimental values were 647.0 MPa and 3.31%, respectively. The mean absolute percentage errors (MAPE) of the two mechanical properties between the predicted and experimental results were 4.56% and 9.97%, respectively. It indicated the effectiveness of the data-driven method for the process optimization and property improvement of aluminum matrix composites.

Key words: machine learning, feature analysis, aluminum matrix composite, inverse design

中图分类号:

TP 391

陈水洲, 王晓书, 欧阳求保, 张瑞. 数据驱动的铝基复合材料性能预测和逆向设计[J]. 上海大学学报(自然科学版), 2022, 28(3): 512-522.

CHEN Shuizhou, WANG Xiaoshu, OUYANG Qiubao, ZHANG Rui. Data-driven based properties prediction and reverse design of aluminum matrix composites[J]. Journal of Shanghai University（Natural Science Edition）, 2022, 28(3): 512-522.

图/表 12

表1

图1

图2

图3

图4

表2

图5

图6

表3

表4

表5

图7

参考文献 27

[1]	Dai D, Xu T, Wei X, et al. Using machine learning and feature engineering to characterize limited material datasets of high-entropy alloys[J]. Computational Materials Science, 2020, 175: 109618. doi: 10.1016/j.commatsci.2020.109618
[2]	Sun W, Zheng Y, Yang K, et al. Machine learning-assisted molecular design and efficiency prediction for high-performance organic photovoltaic materials[J]. Science Advances, 2019, 5(11): eaay4275. doi: 10.1126/sciadv.aay4275
[3]	Stanev V, Oses C, Kusne A G, et al. Machine learning modeling of superconducting critical temperature[J]. Computational Materials Science, 2018(1): 405-418.
[4]	Miracle D B. Metal matrix composites-from science to technological significance[J]. Composites Science and Technology, 2005, 65(15/16): 2526-2540. doi: 10.1016/j.compscitech.2005.05.027
[5]	Bakshi S R, Agarwal A. An analysis of the factors affecting strengthening in carbon nanotube reinforced aluminum composites[J]. Carbon, 2011, 49(2): 533-544. doi: 10.1016/j.carbon.2010.09.054
[6]	Jiang L, Li Z, Fan G, et al. The use of flake powder metallurgy to produce carbon nanotube (CNT)/aluminum composites with a homogenous CNT distribution[J]. Carbon, 2012, 50(5): 1993-1998. doi: 10.1016/j.carbon.2011.12.057
[7]	Zhang X, Li S, Pan D, et al. Microstructure and synergistic-strengthening efficiency of CNTs-SiCp dual-nano reinforcements in aluminum matrix composites[J]. Composites Part A: Applied Science and Manufacturing, 2018, 105: 87-96. doi: 10.1016/j.compositesa.2017.11.013
[8]	Tjong S C. Recent progress in the development and properties of novel metal matrix nanocomposites reinforced with carbon nanotubes and graphene nanosheets[J]. Materials Science and Engineering R: Reports, 2013, 74(10): 281-350. doi: 10.1016/j.mser.2013.08.001
[9]	Habibi M K, Hamouda A S, Gupta M. Hybridizing boron carbide (B$_4$C) particles with aluminum (Al) to enhance the mechanical response of magnesium based nano-composites[J]. Journal of Alloys and Compounds, 2013, 550: 83-93. doi: 10.1016/j.jallcom.2012.09.128
[10]	Rahimian M, Ehsani N, Parvin N, et al. The effect of particle size, sintering temperature and sintering time on the properties of Al-Al$_{2}$O$_{3}$ composites, made by powder metallurgy[J]. Journal of Materials Processing Technology, 2009, 209(14): 5387-5393. doi: 10.1016/j.jmatprotec.2009.04.007
[11]	Umasankar V, Anthony X M, Karthikeyan S. Experimental evaluation of the influence of processing parameters on the mechanical properties of SiC particle reinforced AA6061 aluminium alloy matrix composite by powder processing[J]. Journal of Alloys and Compounds, 2014, 582: 380-386. doi: 10.1016/j.jallcom.2013.07.129
[12]	Wu Y, Kim G Y, Anderson I E, et al. Fabrication of Al6061 composite with high SiC particle loading by semi-solid powder processing[J]. Acta Materialia, 2010, 58(13): 4398-4405. doi: 10.1016/j.actamat.2010.04.036
[13]	Rezayat M, Bahremand M R, Parsa M H, et al. Modification of as-cast Al-Mg/B$_{4}$C composite by addition of Zr[J]. Journal of Alloys and Compounds, 2016, 685: 70-77. doi: 10.1016/j.jallcom.2016.05.226
[14]	Amouri K, Kazemi S, Momeni A, et al. Microstructure and mechanical properties of Al-nano/micro SiC composites produced by stir casting technique[J]. Materials Science and Engineering A, 2016, 674: 569-578. doi: 10.1016/j.msea.2016.08.027
[15]	Slipenyuk A, Kuprin V, Milman Y, et al. Properties of P/M processed particle reinforced metal matrix composites specified by reinforcement concentration and matrix-to-reinforcement particle size ratio[J]. Acta Materialia, 2006, 54(1): 157-166. doi: 10.1016/j.actamat.2005.08.036
[16]	Tang F, Hagiwara M, Schoenung J M. Microstructure and tensile properties of bulk nanostructured Al-5083/SiCp composites prepared by cryomilling[J]. Materials Science and Engineering A, 2005, 407(1/2): 306-314. doi: 10.1016/j.msea.2005.07.056
[17]	Ma G N, Wang D, Liu Z Y, et al. An investigation on particle weakening in T6-treated SiC/Al-Zn-Mg-Cu composites[J]. Materials Characterization, 2019, 158: 109966. doi: 10.1016/j.matchar.2019.109966
[18]	Lee H S, Cho S Y, Kwon H. Reaction products of Al-Mg/B$_4$C composite fabricated by pressureless infiltration technique[J]. Materials Science and Engineering A, 2001, 302(2): 227-234. doi: 10.1016/S0921-5093(00)01831-1
[19]	Kalkanli A, Yilmaz S. Synthesis and characterization of aluminum alloy 7075 reinforced with silicon carbide particulates[J]. Materials & Design, 2008, 29(4): 775-780. doi: 10.1016/j.matdes.2007.01.007
[20]	Sharma M M, Amateau M F, Eden T J. Aging response of Al-Zn-Mg-Cu spray formed alloys and their metal matrix composites[J]. Materials Science and Engineering A, 2006, 424(1/2): 87-96. doi: 10.1016/j.msea.2006.02.047
[21]	Dasgupta R, Meenai H. SiC particulate dispersed composites of an Al-Zn-Mg-Cu alloy: property comparison with parent alloy[J]. Materials Characterization, 2005, 54(4/5): 438-445. doi: 10.1016/j.matchar.2005.01.012
[22]	Kumar S R, Gudimetla K, Venkatachalam P, et al. Microstructural and mechanical properties of Al 7075 alloy processed by equal channel angular pressing[J]. Materials Science and Engineering A, 2012, 533: 50-54. doi: 10.1016/j.msea.2011.11.031
[23]	Manoharan J J L. Effect of reinforcement size and matrix microstructure on the fracture properties of an aluminum metal matrix composite[J]. Materials Science and Engineering A, 1992, 150(2): 179-186. doi: 10.1016/0921-5093(92)90110-M
[24]	Dasgupta H M. Effect of matrix strength on the mechanical properties of Al-Zn-Mg/SiCp composites[J]. Composites Part A: Applied Science and Manufacturing, 2000, 31(10): 1139-1145. doi: 10.1016/S1359-835X(00)00062-2
[25]	Min K H, Lee B H, Chang S Y, et al. Mechanical properties of sintered 7xxx series AI/SiCp composites[J]. Materials Letters, 2007, 61(11/12): 2544-2546. doi: 10.1016/j.matlet.2006.09.062
[26]	Ma W C, Zhang G J, Wang M M. Effect of SiC particles on ageing behaviour of SiCp/7075 composites[J]. Journal of Materials Science Letters, 1997, 16: 1867-1869. doi: 10.1023/A:1018501709583
[27]	Lundberg S M, Lee S I. Consistent feature attribution for tree ensembles[DB/OL]. [2022-03-14]. https://arxiv.org/pdf/1706.06060.pdf.

数据驱动的铝基复合材料性能预测和逆向设计

Data-driven based properties prediction and reverse design of aluminum matrix composites

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 27

相关文章 7

编辑推荐

Metrics

本文评价

[1]	岳溪朝, 冯燕, 刘健, 于烨泳, 席慷杰, 钱权. 材料基因组工程专用数据库[J]. 上海大学学报(自然科学版), 2022, 28(3): 399-412.
[2]	胡瑞, 刘庆, 张光捷, 李俊杰, 陈晓玉, 魏晓, 戴东波. 基于特征工程和机器学习的铝基高熵合金稳定性预测[J]. 上海大学学报(自然科学版), 2022, 28(3): 476-484.
[3]	麦嘉琪, 徐鹏程, 丁松, 孙阳庭, 陆文聪. 支持向量回归预测不锈钢的点蚀电位[J]. 上海大学学报(自然科学版), 2022, 28(3): 485-491.
[4]	武星, 胡明涛, 丁鹏. 陶瓷涂层材料多模态数据表征学习[J]. 上海大学学报(自然科学版), 2022, 28(3): 492-503.
[5]	陶立, 朱杰江, 蔡洪浩. 基于经典机器学习的墙柱施工图图像识别[J]. 上海大学学报(自然科学版), 2021, 27(5): 940-949.
[6]	李一航, 肖斌, 唐宇超, 刘馥, 王小梦, 刘轶. 尖晶石氧化物能量和结构的第一性原理计算和机器学习[J]. 上海大学学报(自然科学版), 2021, 27(4): 635-649.
[7]	游洋, 杜婉, 李惟驹, 陈竞哲. 基于机器学习方法的二维材料带隙预测[J]. 上海大学学报(自然科学版), 2020, 26(5): 824-833.