Non-Invasive Estimation of Machining Parameters during End-Milling Operations Based on Acoustic Emission

Abstract

This work presents a non-invasive and low-cost alternative to traditional methods for measuring the performance of machining processes directly on existing machine tools. A prototype measuring system has been developed based on non-contact microphones, a custom designed signal conditioning board and signal processing techniques that take advantage of the underlying physics of the machining process. Experiments have been conducted to estimate the depth of cut during end-milling process by means of the measurement of the acoustic emission energy generated during operation. Moreover, the predicted values have been compared with well established methods based on cutting forces measured by dynamometers.

Improved Estimation of End-Milling Parameters from Acoustic Emission Signals Using a Microphone Array Assisted by AI Modelling

“…In a previous work, a single microphone implementation was developed based on traditional signal processing and a geometric model of the process [ 14 ]. This method required a simpler setup and calculation algorithm, but the signal was highly contaminated by other sources of noise and reverberation, limiting the system’s precision.…”

“…In a previous study, it was found that acoustic emission waves picked up by microphones have mostly low frequency components [ 14 ]. Since the highest frequency component that would be measured was 200 Hz, then, its wavelength would be about 1720 mm.…”

“…In this context, techniques based on the sensing of acoustic emission (AE) phenomena have gained increasing interest during the last few years due to their simplicity of use and the ability to monitor a wide variety of features, such as teeth breakage [ 7 , 8 ], runout [ 9 , 10 ] and chattering [ 11 , 12 ]. In particular, those techniques based on the use of noncontact sensors in the range of audible sound, such as different types of microphones [ 13 , 14 ], are particularly attractive due to the reduced intrusiveness, so they have been used for detecting the chattering phenomenon [ 15 , 16 , 17 ] and tool condition monitoring in conventional machining [ 18 , 19 ], high-speed machining [ 20 ] and abrasion processes [ 21 ].…”

Non-Invasive Inspections: A Review on Methods and Tools

“…The method has also been used to detect On Load Tap Change and amplified the received recording signal. The signal is analysed through extraction of Power Spectrum Density (PSD) using the Welch method (Welch PSD), and the detail of the energy coefficient is determined by using a Haar wavelet [74,75]. Furthermore, AE can be used to estimate cutting depth during the end-milling process by measuring the acoustic emissions generated during the drilling operation.…”

In Situ Measurement of the Machining State in Small-Diameter Drilling by Acoustic Emission Sensing

“…Numerous studies have been conducted in which AE sensing was used to monitor conditions during various machining operations, such as turning [8][9][10][11], milling [12][13][14][15], reaming [16], honing [17,18], grinding [19][20][21][22][23], or polishing [24][25][26]. Several AE sensing studies have also been reported for drilling [27][28][29]; for example, Gómez and Ferrari described correlations between various AE parameters and thrust force and tool wear [30,31], whereas Patra reported the usefulness of an artificial-neural-network model based on wavelet-packet features for evaluating flank wear [32].…”