Quick search
Go!

CONTROL OF SPRINGBACK BY USING AN ARTIFICIAL NEURAL NETWORK MODEL


AXINTE CRINA
UNIVERSITY OF BACĂU

Issue:

MOCM, Number 12, Volume I

Section:

MOCM - Volume I (2006)

Abstract:

One of the greatest challenges of manufacturing sheet metal parts is to obtain consistent parts dimensions. Springback, the elastic strain recovery in the material when the tools are removed, is the major cause of variations and inconsistencies in the final part geometry. Obtaining a consistent and desirable amount of springback is extremely difficult due to the non-linear effects and interactions of process and material parameters. In this work, the ability of an artificial neural network model to predict optimum process parameters and tool geometry which allow to obtain minimum amount of springback is demonstrated, for a simulated cylindrical deep-drawing process.

Keywords:

springback, artificial neural network, cylindrical deep-drawing process.

Code [ID]:

MOCM200612V01S01A0004 [0000701]


Copyright (c) 1995-2007 University of Bacău