Metamodels to describe the high temperature deformation behaviour of Al 2014+2wt%TiB2 composite

Isothermal hot compression tests were conducted on Al 2014 + 2 wt% TiB2 composite at different temperatures (300-450 degrees C) and strain rates (0.001-1 s(-1)). Using the compression test data, an attempt was made to establish the constitutive relationship for the material by employing three different metamodels such as response surface methodology (RSM), genetic algorithm (GA), and artificial neural network (ANN). The capability of these metamodels towards establishing the non-linear constitutive relation between the process parameters such as temperature, strain rate and strain and flow stress has been ascertained using standard statistical parameters, namely, correlation coefficient (R) and average absolute relative error (AARE). A cubic regression equation was suggested by RSM, while 3-15-1 neural network architecture provided a better correlation. The results obtained are presented and discussed here.

Automated summary

Isothermal hot compression tests were conducted on Al 2014 + 2 wt% TiB2 composite to establish the constitutive relationship using different metamodels. The results showed that the 3-15-1 neural network architecture provided a better correlation.