Open AccessUsing deep neural network model to predict the plastic behaviour of DP780 steel under complex loading
Author(s) -
Zemin Fu,
Pengpeng Xiong
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1684/1/012010
Subject(s) - sheet metal , deep drawing , artificial neural network , hardening (computing) , strain hardening exponent , materials science , flow stress , plasticity , structural engineering , tensile testing , test data , ultimate tensile strength , strain rate , composite material , computer science , engineering , artificial intelligence , layer (electronics) , programming language
In the forming process of sheet metal, the sheet would be subjected to a complicated loading history. The change of pre-strain and the change of strain path might be concentrated in the sheet metal forming process. Through the tensile-tensile experiment of dual-phase (DP780) sheet steel, the flow stress-strain curve obtained shows the cross effect and permanent hardening behaviour. In order to predict the flow stress of DP780 sheet steel under different pre-strain, strain and strain paths, a deep neural network model is established. The data set is divided into training set, validation set and test set to train, verify and test the deep neural network model. The correlation coefficient between the test set prediction results and the experimental results is 98.88%, and the deep neural network model has great predictive ability. Simultaneously, the deep neural network accurately predicts the cross effect and permanent hardening behaviour of DP780 steel.