Machining process monitoring

Abstract

In the modern manufacturing industry, monitoring the machining process and tool condition is becoming increasingly important in order to achieve better product quality, higher productivity, higher process automation, and lower human labor costs. This chapter introduces the fundamental technologies and state-of-the-art development for machining process monitoring. After a brief introduction of the background, in section “Measurands and Sensors,” the commonly used measurands for machining process monitoring are presented, including motor power and current, force, torque, acoustic emission, vibration, image, temperature, displacement, strain, etc. The corresponding sensors for these measurands, and the requirement for signal conditioning, are also discussed. Signal conditioning includes amplification, filtering, converting, range matching, isolation, and other processes to make sensor output suitable for data acquisition and signal processing. Knowledge of data acquisition, which is the process of sampling sensor signals and converts the resulting samples into digital numeric values that can be manipulated by a computer, is provided in section “Data Acquisition.” Some key concepts such as analog-to-digital conversion, quantization, sampling rate, Nyquist sampling theorem, and aliasing are explained. Section “Signal Processing” introduces the essential signal process techniques, including the time domain analysis, frequency domain analysis, time-frequency domain analysis, and artificial intelligence approaches such as artificial neural networks, fuzzy logic, etc. Detailed machining process monitoring strategies and approaches, together with some examples and case studies, are provided in section “Monitoring Strategies and Approaches,” which covers the topics of tool wear estimation, tool breakage detection, chatter detection, surface integrity, and chip monitoring.