Mathematical Modeling for Shape Estimation during Waterjet Machining

Abstract

Technology of abrasive waterjet (AWJ) became one of the most important processes for machining. It has some advantages as compared to other technologies, such as no thermal distortion, small cutting forces and high flexibility. Abrasive waterjet has proved to be an efficient technology for plain milling various engineering materials; hard to soft; electrically conductive to insulators. This search introduces a new predictive model of AWJ milling of aluminum alloy. The model is developed to predict the some interested process parameters, which are surface roughness, depth of milling and material removal rate from process variables. Because the AWJ is a complicated multi input-output system, its model is developed using artificial neural network (ANN) as one of the artificial intelligent models. A feed forward neural network based on back error propagation is used. As the ANN needs training sets of data, an extensive experimental work has been carried out to generate these sets. Each set contains process variables and their related process parameters. The tests considered four process variables, which are traverse speed, waterjet pressure, standoff distance and abrasive flow rate. The analysis of the experimental results clears that the increase of the jet traverse speed yields a good surface quality but inversely decreases the depth of cut and the material removal rate. Moreover, it shows that the waterjet pressure and the standoff distance have no effect in the tested range. In addition, the increase of abrasive flow rate yields an increase of the depth of cut and the material removal rate but decreases the surface roughness. Keywords: Abrasive waterjet (AWJ), Plain waterjet (PWJ) milling, controlled depth milling (CDM), Surface roughness, Depth of cut, Material removal rate, Artificial Neural Network.