{"title":"A Study on Barreling Behavior during Upsetting Process using Artificial Neural Networks with Levenberg Algorithm","authors":"H.Mohammadi Majd, M.Jalali Azizpour","country":null,"institution":"","volume":54,"journal":"International Journal of Mathematical and Computational Sciences","pagesStart":841,"pagesEnd":845,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/10631","abstract":"In this paper back-propagation artificial neural network\r\n(BPANN )with Levenberg\u2013Marquardt algorithm is employed to\r\npredict the deformation of the upsetting process. To prepare a\r\ntraining set for BPANN, some finite element simulations were\r\ncarried out. The input data for the artificial neural network are a set\r\nof parameters generated randomly (aspect ratio d\/h, material\r\nproperties, temperature and coefficient of friction). The output data\r\nare the coefficient of polynomial that fitted on barreling curves.\r\nNeural network was trained using barreling curves generated by\r\nfinite element simulations of the upsetting and the corresponding\r\nmaterial parameters. This technique was tested for three different\r\nspecimens and can be successfully employed to predict the\r\ndeformation of the upsetting process","references":null,"publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 54, 2011"}