Strain hardening prediction of materials using genetic algorithm and artificial neural network

A Machine Learning Approach for Modelling Cold-Rolling Curves for Various Stainless Steels

“…In the field of austenitic stainless steel, a number of works have been published during the last few years. Some examples of these are the approach conducted by Susmikanti and Sulistyo [ 6 ] to predict the strain hardening of the austenitic stainless steel under several cold strain levels using genetic algorithm and artificial neural networks (ANNs); the modelling of flow stress curves of austenitic stainless steel AISI 304 and 316 in dynamic strain ageing regimen made by Krishnamunthy et al [ 7 ], Bahrami et al [ 8 ] and Kumar et al [ 9 ] applying the predictive methodology of ANNs; the design of models by Wand et al [ 10 ] to predict the mechanical properties at room temperature including the Rm and Rp of the austenitic stainless steel AISI 304, 316, 321, and 347 as a function of the chemical composition, heat treatment, and test temperature; or the development of the model by Ono and Miyoshi [ 11 ] to predict the Rm and A from other mechanical properties like E or Rp among others for the austenitic AISI 304L and 316L. In addition to these, another important demonstrated use of the machine learning techniques is the prediction of the mechanical properties at elevated temperatures, as shown in the studies about the austenitic stainless steel AISI 304 performed by Kanumuri et al [ 12 ] and the grades AISI 304L and 316L made by Desu et al [ 13 ], both of which estimate the value of Rm , Rp , A , strain hardening exponent ( n ) and strength coefficient ( K ) as a function of the temperature, in the range from 50 °C to 650 °C, and the strain rate, for the three values of 0.0001, 0.001, and 0.01 s −1 .…”