Application of Neural Networks in Preform Design of Aluminium Upsetting Process Considering Different Interfacial Frictional Conditions
Design of the optimum preform for near net shape manufacturing is a crucial step in upsetting process design. In this study, the same is arrived at using artificial neural networks (ANN) considering different interfacial friction conditions between top and bottom die and billet interface. Back propagation neural networks is trained based on finite element analysis results considering ten different interfacial friction conditions and varying geometrical and processing parameters, to predict the optimum preform for commercial Aluminium. Neural network predictions are verified for three new problems of commercial aluminum and observed that these are in close match with their simulation counterparts.
KeywordsArtificial neural network Preform finite element upsetting deformation
- 5.Tumer, H., Sonmez, F.O.: Optimum shape design of die and preform for improved hardness distribution in cold forged parts. Journal of Materials Processing Technology (Article in the Press)Google Scholar
- 9.Meyers, M.A., Chawla, K.K.: Mechanical Behaviour of Materials. Prentice-Hall, Englewood Cliffs (1999)Google Scholar
- 21.Hertz, J., Krogh, A.: Introduction to the Theory of Neural Networks: Addison- Wesley Publishing Company. Addison-Wesley Publishing Company, Reading (1991)Google Scholar
- 22.User’s manual, MSC. Marc, MSC Software Corporation, Santa Ana, California 92707 USA (2005)Google Scholar