Prediction and optimization of machining parameters in turning of Ti-6Al-4V super alloy using machine learning techniques

Abstract

The machining of high temperature metals is rising in various industries. Despite its difficulty in machining, titanium and its alloys are widely used in various industries due to its exceptional strength-to-weight ratio and corrosion resistance. Titanium's high chemical reactivity, low modulus of elasticity, and thermal conductivity make it difficult to machine. These issues drastically reduce titanium alloy machining productivity. Surface roughness has become increasingly important in improving product quality and machining efficiency. A good surface finish can improve fatigue strength, corrosion resistance, and temperature resistance. On the other hand, lower cutting force helps to enhance tool life and lower the overall power consumptions of machining. This study focuses on development of predictive and optimization models to analyze the influence of machining parameters on surface roughness and cutting force to obtain the optimal machining parameters leading to minimum surface roughness and cutting force during turning of Ti-6Al-4V. Predictive models for surface roughness and cutting force are developed using response surface methodology and support vector regression. The developed models have been compared using relative error and mean relative error or mean absolute percentage error. Optimization models have been developed using response surface methodology and particle swarm optimization. The results have been validated using the confirmation experimental tests. Predicted values for surface roughness and cutting force are closely related to experimental values, the results show. In order to reduce cutting force and surface roughness, the most important machining parameter is feed, which is followed by depth of cut and cutting speed. The optimization model outcomes will to lead to the application of excellent surface finish, lower tool wear, less cutting force and lower power consumption as well more environment friendly.