Investigating Machinability in Hard Turning of AISI 52100 Bearing Steel Through Performance Measurement: QR, ANN and GRA Study

Abstract

The existing endeavor investigates on machinability characteristics through performance measurement of flank wear, surface quality and chip morphology during finish turning of AISI 52100 bearing steel (55 ± 1 HRC) under dry environment employing carbide insert
coated along with various layers (TiN/TiCN/Al2O3). Secondly the influence of machining variables viz. cutting speed, feed rate and depth of cut on responses are assessed by ANOVA and modeled through quadratic regression and artificial neural network. Multiparametric optimization of cutting conditions has been obtained through Taguchi based grey relational analysis. Finally, tool life at ideal conditions has been evaluated through experiment. Based on the study, it is disclosed that coated carbide with multilayer insert outperformed during hard machining as wear at the flank surface and surface quality are within the benchmark cap of 0.3 mm and 1.6 microns respectively. From the chip morphology analysis, multilayer coated carbide insert generates lower temperature and maintains cutting edge sharpness and delays the growth of tool wear. ANN model using multilayered feed forward network yields accurate prediction of responses with minimum error percentage compared to QR model. The optimal parametric combination through GRA approach is found to be d1 (0.1 mm)-f1 (0.04 mm/rev)-v2 (110 m/min) and is greatly improved. Feed is the compelling aspect for multi-responses pursued by cutting speed. The tool life at optimized cutting condition is found to be approximately 19
minutes.