MODELING THE PLASMA ARC CUTTING PROCESS USING ANN

This paper deals with modeling of plasma cutting process using artificial neural network (ANN). The objective was to develop an ANN model in order to predict the ten-point height of irregularities (Rz) in terms of three input parameters, cutting current, plate thickness and cutting speed. To this aim, a feed-forward backpropagation ANN was developed and trained using the experimental data. After the prediction accuracy of the ANN was validated, ANN model was used to generate contour plots from which optimal cutting regions were identified.