Optimal Design of the Austenitic Stainless-Steel Composition Based on Machine Learning and Genetic Algorithm

被引:9
作者
Liu, Chengcheng [1 ,2 ]
Wang, Xuandong [2 ]
Cai, Weidong [2 ]
Yang, Jiahui [2 ]
Su, Hang [2 ]
机构
[1] Cent Iron & Steel Res Inst, Inst Struct Steel, Beijing 100081, Peoples R China
[2] China Iron & Steel Res Inst Grp, Mat Digital R&D Ctr, Beijing 100081, Peoples R China
关键词
machine learning; austenitic stainless steel; genetic algorithm; composition optimization; NONDOMINATED SORTING APPROACH; HIGH-THROUGHPUT EXPERIMENTS; INFORMATICS APPROACH; PART I; OPTIMIZATION; PREDICTION; STRENGTH; BEHAVIOR; ALLOYS; DEFORMATION;
D O I
10.3390/ma16165633
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
As the fourth paradigm of materials research and development, the materials genome paradigm can significantly improve the efficiency of research and development for austenitic stainless steel. In this study, by collecting experimental data of austenitic stainless steel, the chemical composition of austenitic stainless steel is optimized by machine learning and a genetic algorithm, so that the production cost is reduced, and the research and development of new steel grades is accelerated without reducing the mechanical properties. Specifically, four machine learning prediction models were established for different mechanical properties, with the gradient boosting regression (gbr) algorithm demonstrating superior prediction accuracy compared to other commonly used machine learning algorithms. Bayesian optimization was then employed to optimize the hyperparameters in the gbr algorithm, resulting in the identification of the optimal combination of hyperparameters. The mechanical properties prediction model established at this stage had good prediction accuracy on the test set (yield strength: R2 = 0.88, MAE = 4.89 MPa; ultimate tensile strength: R2 = 0.99, MAE = 2.65 MPa; elongation: R2 = 0.84, MAE = 1.42%; reduction in area: R2 = 0.88, MAE = 1.39%). Moreover, feature importance and Shapley Additive Explanation (SHAP) values were utilized to analyze the interpretability of the performance prediction models and to assess how the features influence the overall performance. Finally, the NSGA-III algorithm was used to simultaneously maximize the mechanical property prediction models within the search space, thereby obtaining the corresponding non-dominated solution set of chemical composition and achieving the optimization of austenitic stainless-steel compositions.
引用
收藏
页数:17
相关论文
共 49 条
  • [1] High temperature tensile behavior of a PH stainless steel
    Aghaie-Khafri, M.
    Zargaran, A.
    [J]. MATERIALS SCIENCE AND ENGINEERING A-STRUCTURAL MATERIALS PROPERTIES MICROSTRUCTURE AND PROCESSING, 2010, 527 (18-19): : 4727 - 4732
  • [2] Perspective: Materials informatics and big data: Realization of the "fourth paradigm" of science in materials science
    Agrawal, Ankit
    Choudhary, Alok
    [J]. APL MATERIALS, 2016, 4 (05):
  • [3] Bergstra J, 2012, J MACH LEARN RES, V13, P281
  • [4] Modeling the environmental dependence of pit growth using neural network approaches
    Cavanaugh, M. K.
    Buchheit, R. G.
    Birbilis, N.
    [J]. CORROSION SCIENCE, 2010, 52 (09) : 3070 - 3077
  • [5] Chand S., 2015, Surveys in Operations Research and Management Science, V20, P35, DOI DOI 10.1016/J.SORMS.2015.08.001
  • [6] Tensile stress-strain and work hardening behaviour of 316LN austenitic stainless steel
    Choudhary, BK
    Samuel, EI
    Rao, KBS
    Mannan, SL
    [J]. MATERIALS SCIENCE AND TECHNOLOGY, 2001, 17 (02) : 223 - 231
  • [7] An Evolutionary Many-Objective Optimization Algorithm Using Reference-Point-Based Nondominated Sorting Approach, Part I: Solving Problems With Box Constraints
    Deb, Kalyanmoy
    Jain, Himanshu
    [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2014, 18 (04) : 577 - 601
  • [8] A strategy assisted machine learning to process multi-objective optimization for improving mechanical properties of carbon steels
    Diao, Yupeng
    Yan, Luchun
    Gao, Kewei
    [J]. JOURNAL OF MATERIALS SCIENCE & TECHNOLOGY, 2022, 109 : 86 - 93
  • [9] Bayesian Optimization for Adaptive Experimental Design: A Review
    Greenhill, Stewart
    Rana, Santu
    Gupta, Sunil
    Vellanki, Pratibha
    Venkatesh, Svetha
    [J]. IEEE ACCESS, 2020, 8 : 13937 - 13948
  • [10] Hernández-Lobato JM, 2016, J MACH LEARN RES, V17