Artificial Intelligence Model for the Prediction of Cut Quality in Abrasive Water Jet Cutting

In abrasive water jet cutting, the cut quality is of great importance. In this paper, artificial intelligence model was developed for the prediction of cut quality in abrasive water jet cutting of aluminum alloy. To this aim, artificial neural network (ANN) model was developed in terms of workpiece material thickness, traverse rate and abrasive flow rate. Three-layered feedforward ANN model having four hidden neurons trained with backpropagation algorithm with momentum was used for modeling purposes. The mathematical model showed high prediction accuracy with average absolute percentage error of about 3 %. Using the developed ANN model, 3-D graphs, showing the interaction effects of the traverse rate and abrasive flow rate for three different thicknesses, were given. It was showed that ANNs may be used as a good alternative in analyzing the effects of abrasive water jet cutting parameters on the cut quality characteristics.