Modelling of hot closed die forging of an automotive piston with ANN for intelligent manufacturing

This study presents the finite element modelling and analysis of an automotive piston. Several simulations are carried out using a range of process parameters such as, billet temperature, velocity of die, and friction factor. Investigations are carried out for two sets of dies for hot closed die forging of automotive piston by including all realistic process parameters. The final forging load required for manufacturing the piston is estimated along with the maximum equivalent strain rate in the final product. A generic Artificial Neural Network (ANN) model for hot closed die forging of an automotive piston made of an aluminium alloy is then developed with the help of training data obtained from finite element simulations. ANN model intelligently determines: (i) The maximum equivalent strain rate to assess the quality of forged piston; and (ii) The final forging load to determine the selection of forging machine for a given set of input process parameters, i.e., ram velocity, billet temperature, and friction coefficient. The results from these models assist in achieving energy and material saving, quality improvement and in the development of sound automotive pistons. As this approach is generic in nature, other complex industrial processes can be similarly modelled.