Prediction of the smooth-shared and burr height using neural network in the blanking process

During the sheet metal, burr, cutting smoothness, the rolling and the rupture sections are formed on the cutting plane. The sizes of these sections are defended on the material type, the sheet metal thickness, the clearance and the sharpness of the punch. The most important parameter is the cutting clearance. In this study, the samples are taken from four materials with different hardness by using alternative clearance values. Rolling values, cutting smoothness, rupture depth, and the height of the burrs are measured on the samples. Experimental data are used for improvement of the virtual nevro mesh system and ideal cutting clearances are predicted for different thickness and materials. The error is calculated as 9x10-5 and the test performance gives satisfactory results after the improvements on the mesh. The experimental results and the predicted values are tabulated to use in the die making industry. Key Words : Clearance, smooth-sheared, fracture , Burr, Neural Network, Prediction