Neural networks approach for characterisation of viscoelastic polymers

Abstract

AbstractRecent developments in computer‐aided polymer processing have brought along the need for accurate description of the behaviour of viscoelastic materials under the conjugated effect of applied stress and temperature. In order to serve this purpose, in this study, experimental data provided by bubble inflation tests for flat circular thermoplastic membranes (CTPM) comprised of viscoelastic materials when subjected to combined effects of applied stress and temperature are coupled with numerical simulations to obtain the required parameters for the characterisation of such materials. First, pressure inside the bubble and height at the hemispheric pole are recorded during bubble inflation experiment. Thereafter, Christensen's theory of viscoelasticity is employed to define the constitutive model of flat circular thermoplastic membranes (CTPM) and nonlinear equilibrium equations of the inflation process are solved using finite difference method with deferred corrections. As a last step, a neuronal algorithm (ANN model) is employed to minimise the difference between calculated and measured parameters to determine material constants for Christensen viscoelastic model. Although the developed procedure can be applied to several polymeric materials, in this paper, this technique is successfully implemented for high‐impact polystyrene (HIPS) and acrylonitrile–butadiene–styrene (ABS), at typical thermoforming temperatures, 150 and 145°C, respectively. Using these coefficients, the material behaviours of HIPS and ABS with Christensen's constitutive law are reproduced. The material model obtained in this study for HIPS and ABS can be implemented into industrial and academic softwares for applications and design purposes.