Mechanical Properties Prediction for Hot Roll Steel Using Convolutional Neural Network

Prediction the mechanical properties is very important in many real-life industry fields. In this paper, we proposed an efficient convolutional neural network (CNN) to predict the mechanical properties of hot roll steel. In this study, 20,000 sets of data are collected from the hot roll factory, where 16,000 sets of data were used for training the CNN model, and 4,000 sets of data were used for testing the performance of the model. Compared with Support Vector Machine (SVM) and Artificial Neural Network (ANN), The experimental results have been demonstrated to provide a competitive and higher prediction accuracy.

[1]  Rey-Chue Hwang,et al.  Artificial intelligent analyzer for mechanical properties of rolled steel bar by using neural networks , 2010, Expert Syst. Appl..

[2]  Yong Ren,et al.  Prediction of Mechanical Properties of Hot Rolled Strips by BP Artificial Neural Network , 2011, 2011 International Conference of Information Technology, Computer Engineering and Management Sciences.

[3]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[4]  Shahram Abbasi,et al.  Artificial Neural Network Modeling the Tensile Strength of Hot Strip Mill Products , 2009 .

[5]  Itishree Mohanty,et al.  Online mechanical property prediction system for hot rolled IF steel , 2014 .

[6]  Y. Duan,et al.  Classification of steel materials by laser-induced breakdown spectroscopy coupled with support vector machines. , 2014, Applied optics.

[7]  Kao-Shing Hwang,et al.  Modeling and optimizing tensile strength and yield point on steel bar by artificial neural network with evolutionary algorithm , 2015, 2015 IEEE International Conference on Automation Science and Engineering (CASE).

[8]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[9]  P. Saravanakumar,et al.  Prediction of Mechanical Properties of Low Carbon Steel in Hot Rolling Process Using Artificial Neural Network Model , 2012 .

[10]  L. Gardner,et al.  The continuous strength method for the design of hot-rolled steel cross-sections , 2018 .

[11]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[12]  Matthias Schonlau,et al.  Support Vector Machines , 2016 .

[13]  Jyh-Horng Chou,et al.  Modeling and Optimizing Tensile Strength and Yield Point on a Steel Bar Using an Artificial Neural Network With Taguchi Particle Swarm Optimizer , 2016, IEEE Access.

[14]  Xiang Li,et al.  Understanding the Disharmony Between Dropout and Batch Normalization by Variance Shift , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  Marco Beghini,et al.  Experimental parameter sensitivity analysis of residual stresses induced by deep rolling on 7075-T6 aluminium alloy , 2014 .