基于机器学习模型的中锰钢临界温度A1和A3的预测

doi:10.13251/j.issn.0254-6051.2025.02.044

金属热处理 ›› 2025, Vol. 50 ›› Issue (2): 268-277.DOI: 10.13251/j.issn.0254-6051.2025.02.044

基于机器学习模型的中锰钢临界温度A₁和A₃的预测

张志业¹, 王焱², 张彪¹, 季泽¹, 刘亚良¹, 张明赫¹, 冯运莉¹

1.华北理工大学冶金与能源学院, 河北唐山 063210;
2.石家庄海关技术中心曹妃甸业务部, 河北唐山 063205

收稿日期:2024-03-15 修回日期:2024-12-06 发布日期:2025-04-10
通讯作者: 张明赫,副教授,博士,E-mail:mhzhangmse@163.com
作者简介:张志业(2001—),男,学士,主要研究方向为中锰钢的组织与性能,E-mail:15127157690@163.com。
基金资助:
国家自然科学基金(51901078);河北省自然科学基金(E2022209070);河北省中央引导地方科技发展资金(236Z1003G);唐山市市级科技计划(24130207C)

Prediction of critical temperature A₁ and A₃ of medium-Mn steel based on machine learning models

Zhang Zhiye¹, Wang Yan², Zhang Biao¹, Ji Ze¹, Liu Yaliang¹, Zhang Minghe¹, Feng Yunli¹

1. College of Metallurgy and Energy, North China University of Science and Technology, Tangshan Hebei 063210, China;
2. Shijiazhuang Customs Technology Center Caofeidian Business Department, Tangshan Hebei 063205, China

Received:2024-03-15 Revised:2024-12-06 Published:2025-04-10

摘要/Abstract

摘要： 为便于中锰钢热处理工艺的设计,开发了用于中锰钢临界温度A₁、A₃预测的机器学习模型。通过Thermal-Calc模拟软件获取496组不同成分中锰钢临界温度数据,以Mn、Al、C成分作为输入特征,以相变温度A₁和A₃作为输出目标。采用均方根误差(RMSE)、平均绝对误差(MAE)和决定系数(R²)3种指标对模型预测效果进行评价。从7种机器学习模型(LR、DT、SVM、GPR、Boosting、Bagging以及ANN)中筛选出了预测A₁的GPR模型和A₃的GPR、ANN模型。结果表明,预测A₁的GPR模型具有足够精度,为预测A₁的最优模型。采用网格搜索法对预测A₃的初步模型进行超参数调节,从而获得A₃的最优模型(单层ANN模型)。针对所应用文献中中锰钢的化学成分,利用最优模型对A₁、A₃进行预测,得到A₁、A₃预测值与实测值的总体MAE分别为9.95 ℃和13.57 ℃,最低差距分别为0.30 ℃和6.20 ℃,表明模型精准度高,可用于中锰钢临界温度的预测。

关键词: 机器学习, 中锰钢, 临界温度, 参数调节

Abstract: In order to facilitate the design of heat treatment process of medium-Mn steel, a machine learning model for predicting the critical temperature A₁ and A₃ of medium-Mn steel was optimized. The critical temperature data of 496 groups of medium-Mn steels with different compositions were obtained by Thermal-Calc simulation software. Mn, Al and C compositions were taken as input characteristics, and phase transition temperatures A₁ and A₃ were taken as output targets. Three indexes of root mean square error (RMSE), mean absolute error (MAE) and determination coefficient (R²) were used to evaluate the prediction effect of the model. From seven machine learning models (LR, DT, SVM, GPR, Boosting, Bagging and ANN), the GPR model for predicting A₁ and the GPR and ANN model for predicting A₃ were screened. The results show that the GPR model for predicting A₁ has sufficient accuracy, that is the optimal model for A₁. The grid search method is used to adjust the hyperparameters of the preliminary model for predicting A₃, and the optimal model of A₃ (single-layer ANN model) is obtained. According to the chemical composition of medium-Mn steel in the applied literature, A₁ and A₃ are predicted by using the optimal model. The overall MAE of the predicted value and measured value of A₁ and A₃ is 9.95 ℃ and 13.57 ℃, respectively, and the minimum difference is 0.30 ℃ and 6.20 ℃, respectively, indicating that the model has high accuracy and can be used to predict the critical temperature of medium-Mn steel.

Key words: machine learning, medium-Mn steel, critical temperature, parameter adjustment

中图分类号:

TG111.7

张志业, 王焱, 张彪, 季泽, 刘亚良, 张明赫, 冯运莉. 基于机器学习模型的中锰钢临界温度A₁和A₃的预测[J]. 金属热处理, 2025, 50(2): 268-277.

Zhang Zhiye, Wang Yan, Zhang Biao, Ji Ze, Liu Yaliang, Zhang Minghe, Feng Yunli. Prediction of critical temperature A₁ and A₃ of medium-Mn steel based on machine learning models[J]. Heat Treatment of Metals, 2025, 50(2): 268-277.

参考文献

[1] 张明赫. 基于高能X射线的中锰钢组织与力学行为研究[D]. 北京: 北京科技大学, 2019.
Zhang Minghe. High-energy X-ray diffraction studies on the microstructure and mechanical behavior of medium-Mn steels [D]. Beijing: University of Science and Technology Beijing, 2019.
[2] Lee J Y, Kim M, Lee Y K. Design of high strength medium-Mn steel using machine learning[J]. Materials Science and Engineering A, 2022, 843: 143148.
[3] 王正. 微结构调控中锰钢机械性能和氢脆性能研究[D]. 北京: 北京科技大学, 2021.
Wang Zheng. Research on mechanical properties and hydrogen embrittlement performance in medium Mn steels by microstructural control [D]. Beijing: University of Science and Technology Beijing, 2021.
[4] Kwok T W J, Dye D. A review of the processing, microstructure and property relationships in medium Mn steels[J]. International Materials Reviews, 2023, 68(8): 1098-1134.
[5] Suh D W, Kim S J. Medium Mn transformation-induced plasticity steels: Recent progress and challenges[J]. Scripta Materialia, 2017, 126: 63-67.
[6] 胡宝佳. 高强塑积中锰钢亚稳奥氏体调控的微观机理研究[D]. 合肥: 中国科学技术大学, 2022.
Hu Baojia. Microstructural mechanisms of metastable austenite control in high strength-ductility medium-Mn steels[D]. Hefei: University of Science and Technology of China, 2022.
[7] Zou Y, Xu Y B, Wang G, et al. Improved strength-ductility-toughness balance of a precipitation-strengthened low-carbon medium-Mn steel by adopting intercritical annealing-tempering process[J]. Materials Science and Engineering A, 2021, 802: 140636.
[8] Benzing J T, Kwiatkowski D S A, Morsdorf L, et al. Multi-scale characterization of austenite reversion and martensite recovery in a cold-rolled medium-Mn steel[J]. Acta Materialia, 2019, 166: 512-530.
[9] Li S S, Wen P Y, Li S L, et al. A novel medium-Mn steel with superior mechanical properties and marginal oxidization after press hardening[J]. Acta Materialia, 2021, 205: 116567.
[10] 谢建新, 宿彦京, 薛德祯, 等. 机器学习在材料研发中的应用[J]. 金属学报, 2021, 57(11): 1343-1361.
Xie Jianxin, Su Yanjing, Xue Dezhen, et al. Machine learning for materials research and development[J]. Acta Metallurgica Sinica, 2021, 57(11): 1343-1361.
[11] Zhang Y, Cheng W, Wang C X, et al. Phase prediction in high entropy alloys with a rational selection of materials descriptors and machine learning models[J]. Acta Materialia, 2020, 185: 528-539.
[12] Malinov S, Sha W, McKeown J J. Modelling the correlation between processing parameters and properties in titanium alloys using artificial neural network[J]. Computational Materials Science, 2001, 21(3): 375-394.
[13] Wang Z, Xu J P, Li J X. Mechanical properties and fracture behaviors of medium-Mn steels with and without delta-ferrite for different intercritical annealing times[J]. Materials Characterization, 2021, 172: 110730.
[14] Cai Z H, Jing S Y, Li H Y, et al. The influence of microstructural characteristics on yield point elongation phenomenon in Fe-0.2C-11Mn-2Al steel[J]. Materials Science and Engineering A, 2019, 739: 17-25.
[15] Guo Z K, Li L F, Yang W Y, et al. Microstructures and mechanical properties of high-Mn TRIP steel based on warm deformation of martensite[J]. Metallurgical and Materials Transactions A, 2015, 46: 1704-1714.
[16] Zhang R, Cao W Q, Peng Z J, et al. Intercritical rolling induced ultrafine microstructure and excellent mechanical properties of the medium-Mn steel[J]. Materials Science and Engineering A, 2013, 583: 84-88.
[17] 牛程程, 李少波, 胡建军, 等. 机器学习在材料信息学中的应用综述[J]. 材料导报, 2020, 34(23): 23100-23108.
Niu Chengcheng, Li Shaobo, Hu Jianjun, et al. Application of machine learning in material informatics: A survey[J]. Materials Reports, 2020, 34(23): 23100-23108.
[18] Wu Y M, Shang Z W, Lu T, et al. Target-directed discovery for low melting point alloys via inverse design strategy[J]. Journal of Alloys and Compounds, 2024, 971: 172664.
[19] 徐小鹏. 基于人工神经网络的硅基光子器件逆设计方法的研究[D]. 济南: 山东大学, 2023.
Xu Xiaopeng. Research on the inverse design method of silicon-based photonic devices based on artificial neural network [D]. Jinan: Shandong University, 2023.
[20] 王洪桥. 高斯过程回归在不确定性量化中的应用[D]. 上海: 上海交通大学, 2018.
Wang Hongqiao. Gaussian process regression for uncertainty quantification [D]. Shanghai: Shanghai Jiao Tong University, 2018.
[21] Bergstra J, Bengio Y. Random search for hyper-parameter optimization[J]. Journal of Machine Learning Research, 2012, 13(1): 281-305.
[22] Jasper S, Hugo L, Ryan P A. Practical Bayesian optimization of machine learning algorithms[C] //Proceedings of the 25th International Conference on Neural Information Processing Systems. 2012: 2951-2959.

基于机器学习模型的中锰钢临界温度A₁和A₃的预测

Prediction of critical temperature A₁ and A₃ of medium-Mn steel based on machine learning models

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	刘明珠, 丁桦, 邹宇明. 部分奥氏体逆转变工艺对Fe-8Mn-0.2C-3Al中锰钢组织性能的影响[J]. 金属热处理, 2025, 50(2): 90-95.
[2]	张盛豪, 王宝, 李思佳, 肖美美, 周建安. 奥氏体逆相变退火温度对含铜中锰钢组织和性能的影响[J]. 金属热处理, 2025, 50(2): 96-101.
[3]	张英凡, 陈慧琴, 党淑娥, 陈娟, 徐全, 房晓天, 石腾龙, 代云云. 基于机器学习与遗传算法的Cu-Ni-Co-Si合金性能预测及机理[J]. 金属热处理, 2025, 50(1): 255-265.
[4]	王欢欢, 卢苏君, 李渊, 徐宁, 朱廷贤, 王旭, 魏宁, 吴大莉, 彭威中. 机器学习预测退火温度和时间对SUS321不锈钢力学性能的影响[J]. 金属热处理, 2025, 50(1): 266-271.
[5]	张文, 郭雨潼, 张灵聪, 沈锐, 石绘, 包汉伟, 李刚炎. 45钢激光淬火表面非稳态传热计算方法[J]. 金属热处理, 2024, 49(8): 232-241.
[6]	安庆生, 万德成, 马少康, 李杰, 张千禧. 逆相变退火对中锰钢组织演变和力学性能的影响[J]. 金属热处理, 2024, 49(6): 36-42.
[7]	刘敏, 郑志鹏, 甘利红, 刘胤辰, 刘震, 张阳, 冯波, 刘春泉. 桁架用Nb-Mo微合金中锰钢强化机理及碳化物析出行为[J]. 金属热处理, 2024, 49(3): 198-203.
[8]	周建虎, 李岩, 定巍, 方琪. 0.2C-5.0Mn-0.5Si-1.0Al中锰钢选择性氧化的热力学及动力学分析[J]. 金属热处理, 2024, 49(11): 17-22.
[9]	余灿生, 蒋家乐, 李云杰, 常智渊, 康健, 袁国. 逆相变退火温度对低密度中锰钢组织和力学性能的影响[J]. 金属热处理, 2024, 49(11): 118-122.
[10]	张坛, 李昊彧, 定巍, 李岩. 预备组织对中锰钢临界退火组织演变和力学性能的影响[J]. 金属热处理, 2024, 49(10): 8-17.
[11]	郑昊青, 孟少博, 任武彬, 孙新军. 两相区退火时间对Fe-0.16C-0.2Si-4.78Mn-0.24Mo中锰钢组织和力学性能的影响[J]. 金属热处理, 2024, 49(10): 133-139.
[12]	张伦凤, 王治涵, 赵俊渝, 安康, 徐骏, 顾剑锋. 热处理工艺模拟专用材料数据库的设计与实现[J]. 金属热处理, 2023, 48(9): 247-252.
[13]	彭龙生, 刘春泉, 熊芬, 张颖, 黄伟, 林英华. 轧制方式及热处理工艺对中锰钢组织和性能的影响[J]. 金属热处理, 2023, 48(8): 106-112.
[14]	樊立峰, 郭芷毓, 张志朋, 岳尔斌, 肖丽俊, 高军, 何建中. 逆相变退火时间对5%Mn冷轧中锰钢显微组织和力学性能的影响[J]. 金属热处理, 2023, 48(6): 29-35.
[15]	贾涓, 江萱, 蒋建江, 宋新莉. 淬火配分时间对一种中锰钢耐磨性的影响[J]. 金属热处理, 2023, 48(5): 270-274.