王治涵,徐骏,王婧,李传维,顾剑锋

• 1:上海交通大学材料科学与工程学院材料改性与数值模拟研究所
• 2:上海交通大学上海市激光制造与材料改性重点实验室

摘要(Abstract)：

热处理模拟技术可以高效、低成本地设计和优化热处理工艺，但受限于现有的实验测试技术和计算手段，目前热处理模拟所需的材料数据库普遍面临数据完整性不足的问题。为此，基于B/S架构设计并开发了一款适用于热处理模拟的材料数据库系统，并提供了一种基于机器学习技术的材料参数获取方法。结果表明：所设计的一种自动优化算法，能解决数据库数据更新、机器学习模型再次训练过程中超参数人工调节所存在的盲目性和随机性等难题；经验证，利用机器学习模型预测的材料参数，误差相对于传统线性插值法可降低40%以上。

关键词(KeyWords)： 热处理模拟;材料数据库;B/S架构;机器学习;超参数优化

Abstract:

Keywords:

基金项目(Foundation): 国家重点研发计划(2023YFB3408000);; 宁波市重点技术研发(2023T002);; 国家自然科学基金(52171042)

作者(Author): 王治涵,徐骏,王婧,李传维,顾剑锋

DOI: 10.13289/j.issn.1009-6264.2025-0168

参考文献(References)：

• [1] 周玉龙，张振，周永松，等.热处理过程智能信息化生产调度技术[J].金属热处理，2020,45(12):242-247.ZHOU Yu-long,ZHANG Zhen,ZHOU Yong-song,et al.Intelligent information production scheduling technology of heat treatment process[J].Heat Treatment of Metals,2020,45(12):242-247.
• [2] 高志玉，薛维华.金属热处理计算机辅助工艺设计系统之核心数据库设计[J].金属热处理，2012,37(9):128-131.GAO Zhi-yu,XUE Wei-hua.Core database design of metal heat treatment computer aided process planning system[J].Heat Treatment of Metals,2012,37(9):128-131.
• [3] 徐跃明，李俏，罗新民，等.热处理技术进展[J].金属热处理，2015,40(9):1-15.XU Yue-ming,LI Qiao,LUO Xin-min,et al.Technology progress in heat treatment[J].Heat Treatment of Metals,2015,40(9):1-15.
• [4] 任鑫.热处理数值模拟技术的研究进展[J].材料导报，2021,35(19):19186-19194.REN Xin.Research progress of numerical simulation in heat treatment[J].Materials Reports,2021,35(19):19186-19194.
• [5] 顾剑锋，潘健生.智能热处理及其发展前景[J].金属热处理，2013,38(2):1-9.GU Jian-feng,PAN Jian-sheng.Intelligent heat treatment and its development prospects[J].Heat Treatment of Metals,2013,38(2):1-9.
• [6] 潘健生，王婧，顾剑锋.热处理数值模拟进展之一——扩展求解域热处理数值模拟[J].金属热处理，2012,37(1):7-13.PAN Jian-sheng,WANG Jing,GU Jian-feng.One of progress in heat treatment numerical simulation——Numerical model of heat treatment with expanded solution domain[J].Heat Treatment of Metals,2012,37(1):7-13.
• [7] Alberi K,Nardelli M B,Zakutayev A,et al.The 2019 materials by design roadmap[J].Journal of Physics D-Applied Physics,2019,52(1):013001.
• [8] 何高，程晓敏，何小于.常用钢的淬火回火热处理数据库系统设计[J].热处理技术与装备，2009,30(1):58-61.HE Gao,CHENG Xiao-min,HE Xiao-yu.Design of database system of heat treatment about quenching and tempering for the commonly used steels[J].Heat Treatment Technology and Equipment,2009,30(1):58-61.
• [9] 朱利民.基于VBA的常用材料及热处理查询系统开发[J].煤炭技术，2020,39(12):132-135.ZHU Li-min.Development of common materials and heat treatment inquiry system based on VBA[J].Coal Technology,2020,39(12):132-135.
• [10] 高宁，邵周俊，徐跃明.热处理与表面改性技术数据共享网的建设[J].金属热处理，2009,34(1):11-13.GAO Ning,SHAO Zhou-jun,XU Yue-ming.Website construction of heat treatment and surface modification data sharing[J].Heat Treatment of Metals,2009,34(1):11-13.
• [11] Toenjes A,Von Hehl A.A fast method for predicting the mechanical properties of precipitation-hardenable aluminum alloys[J].Metals,2019,9(2):147-155.
• [12] Schmid-Fetzer R,Zhang F.The light alloy Calphad databases PanAl and PanMg[J].Calphad,2018,61:246-263.
• [13] Saal J E,Kirklin S,Aykol M,et al.Materials design and discovery with high-throughput density functional theory:The open quantum materials database (OQMD)[J].JOM,2013,65:1501-1509.
• [14] Rumor L,Andrade-Campos A.On the need for material model databases:A state-of-the-art review[J].Advances in Mechanical Engineering,2022,14(10):1-12.
• [15] Wang S,Su X,Zhou L,et al.A thermodynamic database of drill pipe steel and its application in high strength drill pipe design[C]//Advances in Energy and Environmental Materials:Proceedings of Chinese Materials Conference 2017 18th,Singapore,2018.
• [16] Xu G,Zhang L,Liu L,et al.Thermodynamic database of multi-component Mg alloys and its application to solidification and heat treatment[J].Journal of Magnesium and Alloys,2016,4(4):249-264.
• [17] 杨丽，苏航，柴锋，等.材料数据库和数据挖掘技术的应用现状[J].中国材料进展，2019,38(7):672-681.YANG Li,SU Hang,CHAI Feng,et al.Material database and application status of data mining technology[J].Materials China,2019,38(7):672-681.
• [18] 文成.基于机器学习的高熵合金成分设计与性能优化[D].北京：北京科技大学，2022.WEN Cheng.Composition design and property optimization of high entropy alloys based on machine learning[D].Beijing:University of Science and Technology Beijing,2022.
• [19] 贾丹.润滑材料数据库平台设计及机器学习性能预测方法研究[D].北京：机械科学研究总院，2021.JIA Dan.Research on method of database platform design and machine learning performance prediction of lubricating materials[J].Beijing:China Academy of Machinery Science and Technology,2021.
• [20] Hautier G,Fischer C C,Jain A,et al.Finding nature’s missing ternary oxide compounds using machine learning and density functional theory[J].Chemistry of Materials,2010,22(12):3762-3767.
• [21] Moot T,Isayev O,Call R W,et al.Material informatics driven design and experimental validation of lead titanate as an aqueous solar photocathode[J].Materials Discovery,2016,6:9-16.
• [22] Faber F A,Hutchison L,Huang B,et al.Prediction errors of molecular machine learning models lower than hybrid DFT error[J].Journal of chemical theory and computation,2017,13(11):5255-5264.
• [23] Chicco D,Warrens M J,Jurman G.The coefficient of determination R-squared is more informative than SMAPE,MAE,MAPE,MSE and RMSE in regression analysis evaluation[J].Peerj Computer Science,2021,7:e623.
• [24] Bergstra J,Bengio Y.Random search for hyper-parameter optimization[J].The Journal of Machine Learning Research,2012,13(1):281-305.
• [25] Bergstra J,Bardenet R,Bengio Y,et al.Algorithms for hyper-parameter optimization[J].Advances in Neural Information Processing Systems,2011,24:2546-2554.
• [26] Hansen N,Ostermeier A.Completely derandomized self-adaptation in evolution strategies[J].Evolutionary Computation,2001,9(2):159-195.
• [27] Press W H.Numerical Recipes 3rd Edition:The Art of Scientific Computing[M].Cambridge:Cambridge University Press,2007.

文章评论(Comment)：