Hot Deformation Flow Behaviour Modelling of Cast-Extruded Al 7075 Alloy Using Arrhenius Equations and Artificial Neural Network

Abstract

Aluminium alloys are widely employed for manufacturing structural components in automotive industries mainly due to their high strength to weight ratio. Selecting the appropriate material is the most important task in achieving the required performance for a particular application. Numerical simulation has been increasingly applied in the process of evaluation of material properties for structural applications and the reliability of such simulations greatly depends on the accuracy of material-flow stress relationship and the mathematical model that are being used. In this paper, both constitutive equations and artificial neural network (ANN) based models have been developed and presented for the cast-extruded aluminium 7075 alloy. Hot compression tests have been carried out on the cast-extruded alloy specimens over the temperature range of 573–773 K at three different strain rates of 0.1, 0.01 and 0.001 s−1 up to a strain of 0.5. Arrhenius-type constitutive equation model and ANN model have been developed to predict the hot deformation behaviour of the alloy based on the experimental data. The performance of these models is compared in terms of accuracy and its degree of correlation with experimental results. From the analysis, the ANN model is found to be more accurate in predicting the flow behaviour with more than 80 % reduction in average relative error compared to the Arrhenius-type constitutive equation model.