Identification of the prefailure phase in microdrilling operations using multiple sensors

Abstract The life of microdrills with a diameter of less than 1 mm is short and unpredictable. Estimation of tool wear and the prefailure phase are more difficult than conventional drilling operations since most special machine tools for microdrilling have stepping motors that create fluctuating forces which consequently produce vibration. In this paper, a new method is proposed for detection of the prefailure phase of microdrilling, just 0.2 to 1 s before breakage occurs. The system measures the thrust force and microdrill velocity by using a dynamometer and a laser vibrometer, respectively. Seven characteristic features of the signals are obtained by describing their shape and spike characteristics. Adaptive Resonance Theory (ART2)-based neural networks are used for interpretation of the signal features. The proposed system accurately classified all the cases studied when a vigilance parameter of 0.995 of the ART2-type neural network was selected.