TY  - JOUR
T1  - Prediction of Geometrical Instabilities in Deep Drawing Using Artificial Neural Network
AU - , K.K. Pathak AU - , Vikas Kumar Anand AU - , Geeta Agnihotri 
JO  - Journal of Engineering and Applied Sciences
VL  - 3
IS  - 4
SP  - 344
EP  - 349
PY  - 2008
DA  - 2001/08/19
SN  - 1816-949x
DO  - jeasci.2008.344.349
UR  - https://makhillpublications.co/view-article.php?doi=jeasci.2008.344.349
KW  - Deep drawing
KW  -wrinkling
KW  -thinning
KW  -finite element
KW  -neural network
AB  - Geometrical instabilities like wrinkling and necking are 2 major defects in deep drawing process. Because of them, drawability is greatly reduced leading to huge lose of material and money. Friction has an important bearing on wrinkling and necking. Hence their prediction is of utmost importance in deep drawing process design. In past such prediction were made via trial and error approaches based on shop floor experiences. But such approaches are crude and time consuming. To overcome these difficulties, Artificial Neural Network (ANN) has been used in this study. Neural networks are trained based on finite element simulated data. Limiting strain hardening exponent for the success of deep drawing, are arrived at from FE simulations. It has been shown that proposed approach is powerful and fast in predictions of geometrical instabilities in deep drawing process.
ER  -