Analysis of CNC turning parameters and simultaneous optimisation of surface roughness and material removal rate by MOGA for AISI 4340 alloy steel

Abstract

ABSTRACTThis article has made an approach towards the multipass turning of AISI 4340 alloy steel with minimum quantity of coolant to determine the Optimal Turning Parameters for simultaneous maximisation of material removal rate and minimisation of surface roughness. Nine experimental runs were planned according to Taguchi’s Design of Experiments and performed on Computer Numerically Controlled Lathe Machine with high carbon steel as a tool. The mathematical models developed for surface roughness and material removal rate at different levels of speed, DOC & feed rate are considered as objective functions. Less than 5% error was observed when experimental results were compared with Regression and ANN Models for both responses. A multi-objective genetic algorithm and a Grey relational analysis were used to determine the optimal turning parameters and the results were validated experimentally. The MOGA provides the best optimal parameters for the simultaneous achievement of SR minimisation and MRR maximisation. Results reveal that speed and feed rate have a significant effect on responses. The Pareto Front Plots extracted from MOGA provide the minimum Ra value, 2.32 µm and for the same point, maximum MRR is 6894.22 mm3/min. However, the experimental method could not meet this Ra value, but the recorded minimum was 2.67 µm and the MRR was 6544.79 mm3/min.KEYWORDS: AISI 4340 alloy steelCNC multipass turningsurface roughnessmaterial removal rateGRAMOGAANNregression Nomenclature SR (Ra)=Surface Roughness in µmMRR=Material Removal Rate in mm3/minS (SS)=Speed in rpmDOC=Depth of Cut in mmFR=Feed Rate mm/revMOGA=Multi Objective Genetic AlgorithmGRA=Grey Relational AnalysisGRG=Grey Relation GradeANN=Artificial Neural networkMQC=Minimum quantity of coolantOF=Objective functionCNC=Computer Numerical ControlOA=Orthogonal Arrayw/p=Work pieceDisclosure statementNo potential conflict of interest was reported by the author(s).Data availability of statementThe data was collected from original research work and optimisation of data is done with a genetic algorithm and GRA process.Code availabilityThe datasets generated during and/or analysed during the current study are available from the corresponding author upon reasonable request.