Predicting the multiaxial stress-strain behavior of polyethylene terephthalate (PET) at different strain rates and temperatures above Tg by using an Artificial Neural Network

In this paper, an Artificial Neural Network (ANN) is used to predict the stress-strain behavior of PET at conditions relevant to Stretch Blow Moulding including large equibiaxial deformation and constant width deformation at elevated temperature and high strain rate for plane stress conditions. The input vectors considered are temperature(T), delta time (Delta t), delta strain (Delta epsilon) and stress(sigma) in both the x and y directions with a corresponding output parameter of delta stress (Delta sigma) in both directions. In the present work, a feed-forward backpropagation algorithm was used to train the ANN. Predictions from the ANN were compared with experimental results showing that it was able to approximate the relationship between stress and strain during both equibiaxial and constant width stretch experiments at various strain rates & temperatures to a high degree of accuracy for all conditions tested.

Automated summary

In this study, an Artificial Neural Network (ANN) was used to accurately predict the stress-strain behavior of PET under different conditions, showing that the ANN could highly accurately approximate the relationship between stress and strain.