Application of Artificial Neural Networks for Modeling Deformation Forces in Open-Die Forging

Abstract

This paper presents an experimental and modeling study of the deformation force in bulk forming processes conducted in open dies. The research was performed on stepped axisymmetric aluminum alloy (AlMgSi0.5) specimens under hot-working conditions, with forming temperatures ranging from 420 °C to 460 °C. A full factorial experimental design with five input factors—geometrical parameters of the die and workpiece, as well as forming temperature—was applied to investigate their influence on the deformation force.

The experimental data were used to develop a predictive model based on Artificial Neural Networks (ANN) using the Deep Learning Toolbox in MATLAB. A feedforward neural network with one hidden layer consisting of ten neurons was trained, validated, and tested. The obtained results showed excellent agreement between the experimental and predicted values, with very high values of the correlation coefficients for all datasets.

The high accuracy and generalization ability of the network confirm its suitability for modeling nonlinear relationships in metal forming. The proposed ANN model provides a reliable alternative to complex numerical simulations, enabling efficient prediction of forming forces in open-die forging processes.