Optimization of AWJC Parameters for Ti-6Al-4V Alloy Using ANN-Based GA and PSO

Abstract

AbstractThe objective of this study is to explore at the geometry of the kerf, surface topography, and surface morphology of titanium (Ti6Al4V) alloy produced by abrasive water jet cutting (AWJC) process. The three input parameters are waterjet pressure (WJP), standoff distance (SOD), and abrasive flow rate (AFR), while the two output responses are surface roughness and kerf taper angle, which are utilized as model variables. The implementation of a Taguchi L9 orthogonal array to cut trails and two computational techniques to particle swarm optimization (PSO) and genetic algorithm (GA)-based artificial neural network (ANN) to predict the best parameters to reduce surface roughness and kerf taper angle was significant in this effort. The ANN model is made up of three input neurons, ten hidden neurons, and two output neurons. Analysis of variance was performed to determine the most important factor impacting surface roughness and kerf taper angle. The observation results show that increasing WJP increases the surface roughness and decreases the kerf taper angle. For AWJC of Ti6Al4V alloy with in cutting settings, the minimum surface roughness was 3.18 μm, and the kerf taper angle was 0.964º.KeywordsAWJCTi6Al4V alloyOptimizationSurface topographySurface morphology