ANOVA and fuzzy rule based evaluation and estimation of flank wear, temperature and acoustic emission in turning

Abstract

Prediction of tool wear plays a key role in machining world due to its considerable impact on total costs in terms of fabricated part quality and disposal of cutting insert before reaching the tool life limit. In addition, comprehensive evaluation of machining characteristics provides a perspective for improved quality. In this study, during turning of AISI 5140 steel, on-line measurements of cutting tool tip temperature and acoustic emission (AE) and off-line measurement of flank wear (VB) of the cutting tool were performed. Very limited study has been published on the machinability characteristics of AISI 5140, while no research has been published on flank wear characteristics in addition to AE and temperature sensors before. Therefore, in this context, AE and tool tip temperature were measured with adapted sensor systems while VB measurement was performed with microscope when the machining was stopped. Three levels of cutting speed, feed rate, depth of cut and cutting-edge angle, experimental design was composed based on Taguchi’s L27 orthogonal array. The main aim of the paper is to predict the flank wear i.e., main criteria for tool life, cutting temperature and AE with the help of fuzzy inference model. In addition, for providing a holistic approach and comprehensive point of view to the machinability, statistical analysis and optimization of the input parameters were given in detail for temperature, AE and VB. According to results, the combined effect of depth of cut and cutting-edge angle having effect (40%) on VB while cutting edge angle and cutting speed have dominance on temperature (45.4%) and AE (46.23%) respectively. Considering the complexity of turning operations, obtained findings reveal the superiority of the fuzzy inference model which was found the estimations highly close to the test results (90–97%) for each machining characteristic. The applicability of fuzzy rule system for different types of variables in turning is promising for the new generation of materials utilized in manufacturing sector.