Surface Roughness Analysis in High-Speed Turning of Ti-6Al-4V with Coated Carbide Inserts Using Vibration Signals

Abstract

AbstractUse of vibration signals for analysis of machining processes has been done for various purposes including surface roughness evaluation and tool wear monitoring. Ti-6Al-4V is a common Titanium-based alloy that has been utilized in a variety of applications such as aerospace, automotive, biomedical etc. As Ti is a difficult-to-machine material, there have been few investigations on using vibration signals to analyze roughness of the surface in Ti-based alloys. This research investigates vibration signals generated during high-speed turning (HST) of Ti-6Al-4V employing coated carbide inserts in this context. Experiments with HST have been carried out at high speed and different feed and depth of cuts and the resulting vibration signals have been analyzed along with surface roughness. Tool wear and cutting tool vibrations, which are the major factors in machining these alloys have been considered for the study and analysis. Further, a robust artificial neural network model called Radial Basis Function Neural Networks (RBFNN) was adopted for modeling and predicting surface roughness parameters Ra and Rt. Three RBFNN models have been developed: one that considers all parameters, including tool wear and vibration, another that ignores tool wear and considers vibrations, and a third that ignores vibrations but considers tool wear. All the models developed have achieved the same prediction accuracy of 98.3607% on the test data establishing the feasibility of using this for modeling and also establishing the significance of using vibration and tool wear in surface roughness analysis of Ti-6Al-4V alloys.KeywordsTi-6Al-4VHigh-speed turningRBFNNTool flank wearCutting tool vibration