Condition Monitoring of CNC Drill Bit for the Manufacturing Sector Using Wavelet Analysis and Artificial Neural Network Based on Feedforward Multilayer Perceptron (MLP)

Real-time condition monitoring and precision health assessment systems are necessary for effective maintenance programs in the industrial sector. Rapid advancements in information technology and other engineering technologies have invited more proactive attention from research and development in industrial sectors, particularly in condition monitoring of machines and related Industrial processes. In this work, drill bit condition monitoring techniques have been developed based on wavelet analysis and Artificial Neural Networks (ANN) for automatic drill bit fault detection and classification. An experimental work has been conducted to capture the vibration signals for analysis. The CNC drill machine uses a high-carbon steel drill bit and mild steel material as work pieces in this experiment. The cutting condition parameters are kept constant, and the wear level varies from 0.2 to 0.6 mm. Using an accelerometer, the Data Acquisition system (DAQ) with Lab VIEW software captures the vibration signals for drill bits with different wear conditions. The captured vibration data are analyzed using Continuous Wavelet Transform (CWT) with Morlet and Daubechies wavelet as prime functions. The CWT coefficient is generally used to generate the input features to ANN for automatic tool condition classification, with two outputs (0, 1) for healthy and (1, 0) for faulty. The outcome of ANN showed 98% accuracy in the wear prediction, and these results show the effectiveness of the combed WT and ANN for the automatic classification of tool wear conditions with a high success rate.