Performance assessment and optimization of Ti6Al4V helical hole milling process

Abstract

The superior metallurgical properties exhibited by Ti6Al4V alloy make hole drilling a tedious task. Therefore, the helical milling process was analyzed as an alternative for machining Ti6Al4V alloy in the present work. Using Analysis of Variance (ANOVA), the effects of helical milling parameters such as axial feed, cutting speed, and tangential feed on surface roughness (SR), cutting forces such as thrust force (TF) and radial force (RF), and machining temperature (MT) were investigated. Metrics like R2, root mean square error (RMSE), and error percentage were used to assess the predictive models that were developed using Response Surface Methodology (RSM) and Back Propagation Artificial Neural Networks (BPANN). Furthermore, the multi-criteria decision-making technique, specifically Grey Relation Analysis (GRA), was engaged to ascertain the ideal machining conditions for helical milling Ti6Al4V alloy. The results demonstrate that the chosen process variables influence the performance of the helical milling operation. In the case of predictive models, BPANN performs better than the RSM technique in accurately predicting the data, with errors of 2.31%, 2.32%, 1.56%, and 1.24% for TF, RF, MT, and SR, respectively. Moreover, for dry helical milling of Ti6Al4V alloy, a 75 m min−1 cutting speed, tangential feed of 0.03 mm z−1, and axial feed of 0.2 mm rev−1 were determined as the optimal setting for achieving lower TF, RF, MT, and Sr The findings show that the proposed predictive models and optimization technique can achieve the best outcomes when machining difficult-to-machine material like Ti6Al4V alloy.