Comparison of Signal Processing Techniques for Prediction of Optimal Process Variables to Yield Higher Productivity During Turning on CNC lathe

Abstract

Tool chatter is one of such occurrences that limits MRR in a number of industries. In the current research, a method to boost output while lowering clatter during turning operations on a CNC lathe has been presented. A microphone is used to record the vibration signals generated during turning tests. The denoised signals are analysed using local mean decomposition (LMD). Disruptions and undesirable embedded ambient noise are removed using wavelet denoising (WD). The product functions that expose chatter information are chosen using these decomposed signals. To recreate the real-time chatter, these well-known PFs are used to reconstruct the signal. A consistent range of turning parameters for greater productivity has been created using the Grey relational analysis (GRA) prediction technique. The measured Chatter Index value has been found to denote steady turning, unstable, and moderate chatter circumstances. In order to confirm the validity of the presented methodology, several tests have been conducted.