Prediction and Optimization of Surface Roughness in Minimum Quantity Coolant Lubrication Applied Turning of High Hardness Steel

Abstract

Dry machining is accredited as sustainable manufacturing; however, in case of machining a high hardness steel, an elevated temperature and eventual alteration of machined surface entail thermal pacification. As such, the minimum quantity coolant lubrication (MQCL) bridges the improved productivity with sustainability in machining. Likewise, the prediction and optimization of machining characteristics uphold sustainability by conserving resources. In this regard, this study presents the modeling and investigation of average surface roughness parameter ($R_a$) with respect to spindle speed ($N$), feed rate ($f$), depth of cut ($a_p$) and time ($t$) gap between MQCL pulsing in turning $∼60 R_c$ steel. The least-square support vector machine (LS-SVM) method for the prediction and the interior point method (IPM) for the optimization have been employed. The devised LS-SVM model predicted $R_a$ with $4.96%$ MAPE; while, IPM exhibited improvement in $R_a$ at optimum $N = 259$ rpm, $f=0.18$ mm/rev, $a_p=0.25$ mm and MQCL impingement within the smallest time interval possible. Moreover, the obtained results revealed that the augmented feed due to enhanced straining and wider peak-to-peak distance increased $R_a$ while $1s$ interval based pulsing reduced $R_a$ owing to increased lubrication and cooling.