Modelling of surface roughness in abrasive waterjet cutting of Kevlar 49 composite using artificial neural network

Abstract In abrasive waterjet machining (AWJM), water acts as an accelerating medium and abrasive particles are used for cutting materials. This is a cost effective and environmentally friendly technique. Kevlar 49 is one of the most commonly used composite materials and is extensively used in aerospace industries where high strength-to-weight ratio and excellent corrosion resistance are required. This material can be processed by only by abrasive waterjet machining technology due to its high strength. The quality of AWJM is depending upon the process parameters of this technology. This paper provides an experimental investigation for the performance analysis of process parameters on machining Kevlar 49 using abrasive waterjet technology. In this paper mathematical model is also being developed to predict the surface roughness to find the relation between the inputs and the outputs. The first part predicts the surface roughness based on the pressure and the traverse speed whereas the second part indicates the surface roughness variation with standoff distance and the mass flow rate. The model is developed with the back-propagation algorithm using artificial neural network (ANN). The initial weights are assumed, and the algorithm predict and update the weights until it could predict the actual value. The final model is used to verify the results and is found that it is more than 95% accurate. Hence this model is used to study surface roughness by varying all the parameters theoretically.