Prediction of the process parameters of metal powder preform forging using artificial neural network (ANN)

The powder forging process of die pressing, sintering and upsetting is a convenient way of reducing or eliminating the porosity from the traditional powder metallurgy (PM) products. Forging of metal powder preform enhances the demanding high tensile, impact and fatigue strength of PM products. In this paper, an artificial neural network (ANN) approach for exploring the prediction of the PM process parameters, particularly sinter-forged density of metal powder preform, is derived. The model is based on three layer neural network with back propagation learning algorithm. The training data are collected by the experimental setup in the laboratory. The predicted density of neural network model, coincide well with the experimental density. Therefore a new way of optimizing process parameters and enhancing forging quality of metal powder preform has been provided by the author.