CUTTING TOOL WEAR PREDICTION BY USING THE GROUP METHOD OF DATA HANDLING (GMDH)

In machining area, the use of conventional approach to develop a reliable method predicting tool wear with a mathematical model based on the plastic deformation of the work material cannot always deal to satisfactory results. Sometimes the conventional model gives rather large prediction errors by the disturbance into the cutting process. This paper deals with the prediction of the tool flank wear in a turning operation using the powerful technique called Group Method of data Handling (GMDH). As a process modeling tool, the GMDH algorithm determines a mathematical representation between tool flank wear and the measured variables involved. The GMDH method is said useful for forming a statistical model of a complex multi-variable system using a few process data. The tool wear model obtained by applying GMDH has considerably high prediction accuracy and indicates the influence of input variables on the cutting tool life. Special care was taken to avoid the influence of the dynamic phenomenon of turning process on the obtained experimental data. The derived model reveals that tool wear and consequently tool life is a complex function according to cutting parameters: speed, feed and depth of cut.DOI: http://dx.doi.org/10.5755/j01.mech.18.5.2696