[1]付佳博,王晨充,MATEO Carlos Gracia,等.物理冶金信息指导机器学习的镍基单晶高温合金蠕变寿命预测[J].中国材料进展,2023,42(09):722-731.[doi:10.7502/j.issn.1674-3962.202212004]
　FU Jiabo,WANG Chenchong,MATEO Carlos Gracia,et al.Creep Life Prediction of Ni-Based Single Crystal Superalloys by Physical Metallurgy Information Guided Machine Learning[J].MATERIALS CHINA,2023,42(09):722-731.[doi:10.7502/j.issn.1674-3962.202212004]
点击复制

物理冶金信息指导机器学习的镍基单晶高温合金蠕变寿命预测()
分享到：

中国材料进展[ISSN:1674-3962/CN:61-1473/TG]

卷:
42
期数:
2023年第09期
页码:
722-731
栏目:
出版日期:
2023-09-30

文章信息/Info

Title:
Creep Life Prediction of Ni-Based Single Crystal Superalloys by Physical Metallurgy Information Guided Machine Learning
文章编号:
1674-3962(2023)09-0722-10
作者:
付佳博1; 王晨充1; MATEO Carlos Gracia2;  CARABALLO Isaac Toda2;  CABALLERO Francisca Garcia2; 于皓1
1. 东北大学 轧制技术及连轧自动化国家重点实验室，辽宁 沈阳 110819 2. 西班牙国家冶金研究中心物理冶金系，西班牙 马德里 28040
Author(s):
FU Jiabo1;  WANG Chenchong1;  MATEO Carlos Gracia2;  CARABALLO Isaac Toda2; CABALLERO Francisca Garcia2;  YU Hao1
1. The State Key Laboratory of Rolling and Automation, Northeastern University, Shenyang 110819, China 2. Department of Physical Metallurgy, National Centre for Metallurgical Research (CENIM-CSIC),Avda. Gregorio del Amo, Madrid 28040, Spain
关键词:
镍基单晶高温合金; 机器学习; 蠕变寿命; 高温低应力
Keywords:
Ni-based single crystal superalloy;  machine learning;  creep life;  high temperature and low stress
分类号:
TP181;TG132
DOI:
10.7502/j.issn.1674-3962.202212004
文献标志码:
A
摘要:
蠕变寿命是影响镍基单晶高温合金材料服役寿命和力学性能的关键材料参数。因此，如何准确有效地预测合金的蠕变寿命具有重要现实意义。尽管多年来许多研究学者已经建立起多种蠕变寿命的预测模型，但是由于不同温度应力下的蠕变机制复杂且蠕变过程涉及长时间的显微组织演化，已有模型尚难以实现有效预测。对此，采用物理冶金原理指导下的数据挖掘结合机器学习这一研究策略，通过文献调研建立起了高温低应力下的镍基单晶合金的高质量蠕变数据集，在物理冶金原理指导下对原始数据进行挖掘，提高了原始数据的内在质量，并基于Pearson系数和随机森林平均精确度降低值分别对原始数据特征进行了相关性分析和重要性评估，表明所建立的数据集符合基本的物理冶金学机制，同时阐明了引入的三维物理冶金信息对于蠕变寿命预测的重要意义。随后，基于机器学习方法在数据挖掘后的数据集上对合金的蠕变寿命进行了预测，并根据平方相关系数（R2）、平均绝对误差（MAE）和过拟合程度评估了不同的机器学习模型。结果表明，支持向量回归（SVR)模型在本研究中具有较好的泛化能力且不容易过拟合，同时结合了物理冶金信息的机器学习模型拥有更好的预测准确性和泛化能力。最终成功地建立起了高温低应力下镍基单晶高温合金成分、工艺以及引入的物理冶金参数和蠕变寿命之间的关系，能够实现对镍基单晶高温合金蠕变寿命的有效预测，并有望应用于基于合金服役条件的成分工艺的反向设计。
Abstract:
Creep life is a key material parameter affecting the service life and mechanical properties of Ni-based single crystal superalloys. Therefore, how to predict the creep life of alloys accurately and effectively is critically important for engineering. To address this issue, a physical metallurgy (PM)-guided machine learning (ML) model is developed. Firstly based on literature research, a high quality creep dataset of single crystal at high temperature and low stress is established. Under the guidance of the principles of physical metallurgy, three dimensional physical metallurgy information (volume fraction of γ′ phase Vf,lattice misfit δ, diffusion coefficient DL) are added to the original dataset as extra dimensions to guide the training process. Additionally, the correlation analysis and importance evaluation of the original data features are made based on the Pearson correlation coefficient and the mean accuracy decrease (MDA) value of a random forest model respectively. As a result, the dataset is basically consistent with the physical metallurgy mechanism, and the threedimensional physical metallurgy information is of great significance for the creep life prediction. The creep life of the alloy is predicted on the dataset after data mining based on the machine learning method, and different machine learning models are evaluated according to the squared correlation coefficient (R2), mean absolute error (MAE) and the degree of overfitting. Finally, the optimal model is determined as support vector regression (SVR) model. The relationship between the composition, process, physical metallurgical parameters and creep rupture life of Ni-based single crystal superalloy under high temperature and low stress is successfully established, which can effectively predict the creep life and is expected to serve the reverse design of the alloy.

参考文献/References:

\[1\]REED R C. The Superalloys: Fundamentals and Applications\[M\]. London: Cambridge University Press, 2008. \[2\]XIA W, ZHAO X, YUE L, et al. Journal of Alloys and Compounds\[J\], 2020, 819: 152954. \[3\]CARON P, KHAN T. Aerospace Science and Technology\[J\], 1999, 3(8): 513-523. \[4\]LARSON F R, MILLER J. Transactions of the American Society of Mechanical Engineers\[J\], 1952, 74(5): 765-771. \[5\]DANG Y Y, ZHAO X B, YUAN Y, et al. Materials at High Temperatures\[J\], 2016, 33(1): 1-5. \[6\]ZHU Z, BASOALTO H, WARNKEN N, et al. Acta Materialia\[J\], 2012, 60(12): 4888-4900. \[7\]REED R C, ZHU Z, SATO A, et al. Materials Science and Engineering: A\[J\], 2016, 667: 261-278. \[8\]REED R C, TAO T, WARNKEN N. Acta Materialia\[J\], 2009, 57(19): 5898-5913. \[9\]KIM Y K, KIM D, KIM H K, et al. International Journal of Plasticity\[J\], 2016, 79: 153-175. \[10\]HAN H, LI W, ANTONOV S, et al. Computational Materials Science\[J\], 2022, 205: 111229. \[11\]LIU Y, WU J, WANG Z, et al. Acta Materialia\[J\], 2020, 195: 454-467. \[12\]SMITH T M, UNOCIC R R, DEUTCHMAN H, et al. Materials at High Temperatures\[J\], 2016, 33(4/5): 372-383. \[13\]BARBA D, ALABORT E, PEDRAZZINI S, et al. Acta Materialia\[J\], 2017, 135: 314-329. \[14\]YEH A C, TIN S. Metallurgical and Materials Transactions A\[J\], 2006, 37(9): 2621-2631. \[15\]LU F, LI L, ANTONOV S, et al. Effect of Re on LongTerm Creep Behavior of NickelBased SingleCrystal Superalloys for Industrial Gas Turbine Applications\[C\]//Preceedings of Superalloys 2020. Cham:Springer, 2020: 218-227. \[16\]TAN X P, LIU J L, JIN T, et al. Materials Science and Engineering: A\[J\], 2013, 580: 21-35. \[17\]CARROLL L J, FENG Q, POLLOCK T M. Metallurgical and Materials Transactions A\[J\], 2008, 39(6): 1290-1307. \[18\]HUO J J, SHI Q Y, FENG Q. Materials Science and Engineering: A\[J\], 2017, 693: 136-144. \[19\]SHI Q, HUO J, ZHENG Y, et al. Materials Science and Engineering: A\[J\], 2018, 725: 148-159. \[20\]CHEN J, QIANG F, CAO L, et al. Progress in Natural Science: Materials International\[J\], 2010, 20: 61-69. \[21\]CHEN J Y, LI Q, ZHANG M J, et al. Influences of W and Al on the Microstructure and StressRupture Property of ReFree NiBased Single Crystal Superalloys\[C\]//Preceedings of Materials Science Forum. Switzerland: Trans Tech Publications Ltd, 2017, 898: 492-497. \[22\]CHEN J Y, CAO L M, XUE M, et al. Rare Metals\[J\], 2014, 33(2): 144-148. \[23\]SONG W, WANG X, LI J, et al. Journal of Materials Science & Technology\[J\], 2021, 89: 16-23. \[24\]SONG W, WANG X G, LI J G, et al. Materials Science and Engineering: A\[J\], 2020, 772: 138646. \[25\]YEH A C, RAE C M F, TIN S. Superalloys\[J\], 2004: 677-685. \[26\]WANG W Z, JIN T, JIA J H, et al. Materials Science and Engineering: A\[J\], 2015, 624: 220-228. \[27\]CHEN J, HUO Q, CHEN J, et al. Materials Science and Engineering: A\[J\], 2021, 799: 140163. \[28\]ZHANG J X, MURAKUMO T, KOIZUMI Y, et al. Journal of Materials Science\[J\], 2003, 38(24): 4883-4888. \[29\]ZHANG J X, KOIZUMI Y, KOBAYASHI T, et al. Metallurgical and Materials Transactions A\[J\], 2004, 35(6): 1911. \[30\]SIMONETTI M, CARON P. Materials Science and Engineering: A\[J\], 1998, 254(1/2): 1-12. \[31\]DING Q, BEI H, LI L, et al. International Journal of Mechanical System Dynamics\[J\], 2021, 1(1): 121-131. \[32\]ZHAO Y S, ZHANG J, LUO Y S, et al. Materials Science and Engineering: A\[J\], 2016, 672: 143-152. \[33\]LIU Y F, SHA J B, ZHAO Y S, et al. Rare Metals\[J\], 2018: 1-7. \[34\]RU Y, ZHANG H, PEI Y, et al. Materialia\[J\], 2019, 6: 100278. \[35\]HAN Y F, MA W Y, DONG Z Q, et al. Effect of Ruthenium on Microstructure and Stress Rupture Properties of a Single Crystal NickelBase Superalloy\[C\]//Superalloys 2008: Proceedings of the 11th International Symposium on Superalloys. Warrendale: Minerals Metals & Materials Society, 2008: 91-97. \[36\]YUE Q, LIU L, YANG W, et al. Materials Science and Engineering: A\[J\], 2019, 742: 132-137. \[37\]MACKAY R A, GABB T P, NATHAL M V. Materials Science and Engineering: A\[J\], 2013, 582: 397-408. \[38\]SHEN C, WANG C, WEI X, et al. Acta Materialia\[J\], 2019, 179: 201-214. \[39\]MURAKUMO T, KOBAYASHI T, KOIZUMI Y, et al. Acta Materialia\[J\], 2004, 52(12): 3737-3744. \[40\]VAN SLUYTMAN J S, POLLOCK T M. Acta Materialia\[J\], 2012, 60(4): 1771-1783. \[41\]VOLEK A, PYCZAK F, SINGER R F, et al. Scripta Materialia\[J\], 2005, 52(2): 141-145. \[42\]RETTIG R, RITTER N C, HELMER H E, et al. Modelling and Simulation in Materials Science and Engineering\[J\], 2015, 23(3): 035004. \[43\]YEH A C, SATO A, KOBAYASHI T, et al. Materials Science and Engineering: A\[J\], 2008, 490(1/2): 445-451. \[44\]金涛. 一种无铼第二代镍基单晶高温合金：CN200610046891\[P\]. 2009-02-11. JIN T. NonRhenium No2 Generating NickelBase Mono High Temp Alloy：CN200610046891\[P\]. 2009-02-11. \[45\]金涛. 一种高强度且组织稳定的第四代单晶高温合金及制备方法：CN103382536\[P\]. 2013-11-06. JIN T. FourthGeneration SingleCrystal High Temperature Alloy with High Strength and Stable Structure and Preparation Method Thereof：CN103382536\[P\]. 2013-11-06.

备注/Memo

备注/Memo:
收稿日期：2022-12-05修回日期：2023-04-21 基金项目：国家重点研发计划项目(2021YFB3702500) 第一作者：付佳博，男，1999年生，博士研究生 通讯作者：于皓，男，1991年生，博士后， Email:yuhao@ral.neu.edu.cn 王晨充，男，1988年生，副教授，博士生导师， Email:wangchenchong@ral.neu.edu.cn
更新日期/Last Update: 2023-08-28