Mechanical properties prediction of dual phase steels using machine learning
The use of artificial intelligence techniques, with the increase of data generation capacity and the advancement of computational resources, has enabled the industries to develop and improve products without compromising laboratory and industrial resources. In this paper, a supervised machine learning (ML) based technique was used to predict the yield strength (YS), ultimate tensile strength (UTS), and elongation (EL) of dual phase steels with minimum tensile strengths of 590 and 780 MPa. The computational analysis was done from industrial data information containing the chemical composition and the thermomechanical processing parameters of the referred materials. The proposed ML model reached values of coefficient of determination above 0.94, with an accuracy of ±30 MPa for YS and UTS, and ±3% for EL. These results demonstrated the rationality and reliability of the tested method, allowing its application in future research works and in decision making that aim to optimize the steels industrial processing parameters.
1 Ramazani A, Mukherjee K, Prahl U, Bleck W. Modelling the effect of microstructural banding on the flow curve behavior of dual-phase (DP) steels. Computational Materials Science. 2012;52:46-54.
2 Arruda MVP, Melo TMF, Costa FS, Santos DB. Microstructural evolution during continuous annealing of a 980 MPa cold rolled steel grade. Journal of Physics: Conference Series. 2019;1270(012020):1-6.
3 Xu Z, Liu X, Zhang K. Mechanical properties prediction for hot rolled alloy steel using convolutional neural network. IEEE Access: Practical Innovations, Open Solutions. 2019;7:47068-47078.
4 Guo S, Yu J, Liu X, Wang C, Jiang Q. A predicting model for properties of steel using the industrial big data based on machine learning. Computational Materials Science. 2019;160:95-104.
5 Lalam S, Tiwari PK, Sahoo S, Dalal AK. Online prediction and monitoring of mechanical properties of industrial galvanized steel coils using neural networks. Ironmaking & Steelmaking. 2019;46(1):89-96.
6 Liu Y, Zhao T, Ju W, Shi S. Materials discovery and design using machine learning. Journal of Materiomics. 2017;3:159-177.
7 Rahaman M, Mu W, Odqvist J, Hedström P. Machine learning to predict the martensite start temperature in steels. Metallurgical and Materials Transactions. A, Physical Metallurgy and Materials Science. 2019;50A:2081-2091.
8 Kuhn M. Building predictive models in R using the Caret package. Journal of Statistical Software. 2008;28(5):1-26.
9 Zhou J, Li E, Wei H, Li C, Qiao Q, Armaghani DJ. Random forests and cubist algorithms for predicting shear strengths of rockfill materials. Applied Sciences (Basel, Switzerland). 2019;9(1621):1-16.
10 Noi PT, Degener J, Kappas M. Comparison of multiple linear regression, cubist regression, and random forest algorithms to estimate daily air surface temperature from dynamic combinations of MODIS LST data. Remote Sensing. 2017;9(398):1-23.
11 Chheda AM, Nazro L, Sen FG, Hegadekatte V. Prediction of forming limit diagrams using machine learning. IOP Conference Series. Materials Science and Engineering. 2019;651(012107):1-8.
12 Wijnand HP, Velde RV. Mann-Whitney/Wilcoxon’s nonparametric cumulative probability distribution. Computer Methods and Programs in Biomedicine. 2000;63:21-28.