An acoustic emission activity detection method based on short-term waveform features: Application to metallic components under uniaxial tensile test

Abstract

The Acoustic Emission (AE) phenomenon has been used as a powerful tool with the purpose to either detect, locate or assess damage for a wide range of applications. Derived from its monitoring, one major current challenge on the analysis of the acquired signal is the proper identification and separation of each AE event. Current advanced methods for detecting events are primarily focused on identifying with high accuracy the beginning of the AE wave; however, the detection of the conclusion has been disregarded in the literature. For an automatic continuous detection of events within a data stream, this lack of accuracy for the conclusion of the events generates errors in two critical aspects. In one hand, it deteriorates the accuracy of the measurement of the events duration, truncating the span of the event (undesirable in evaluation applications), and in the other hand, it causes false detections. In this work, an accurate and computationally efficient AE activity detector is presented, using a framework inspired by the area of speech processing, and which provides the required indicators to accurately detect the onset and the end of an AE event. This is achieved by means of a threshold approach that instead of directly operates with the transduced voltage signal it does so over the Short-Term Energy and the Short-Term Zero-Crossing Rate measures of the signal. The STE-ZCR method is developed for an application related to the continuous monitoring of a single AE channel derived from the characterization of metallic components by means of an uniaxial tensile test. Additionally, two experimental test-benches are implemented with the aim to quantify the accuracy and the quality of event detection of the presented method. Finally, the obtained results are compared with four different techniques, representing the current state of the art related to AE activity detection.

Signal Activity Detection for Fiber Optic Distributed Acoustic Sensing with Adaptive-Calculated Threshold

“…At first, by totaling the number of detected signals versus the correct positions (referring to a total amount of true signals presenting in the data frame), classify the properly detected signals (true positive), missed signals (false negative) and mistake detected signals (false positive). Then, these indicators are obtained: accuracy (ACC, the ratio of true-positive detections against to all detected and not detected signals); false-discovery rate (FDR, the ratio of false-positive detections to the whole detected signals); false-negative rate (FNR, the ratio of false-negative detections to the sum of false-negative detections and true positive detections); and the F1-score (the harmonic of 1-FNR and 1-FDR) [34]. Then, these indicators are obtained: accuracy (ACC, the ratio of true-positive detections against to all detected and not detected signals); false-discovery rate (FDR, the ratio of false-positive detections to the whole detected signals); false-negative rate (FNR, the ratio of false-negative detections to the sum of false-negative detections and true positive detections); and the F1-score (the harmonic of 1-FNR and 1-FDR) [34].…”

“…It can be observed that on average, the proposed method is superior to the better reference in terms of the ACC (46.31%), FDR (6.16%), FNR (48.59%) and F1-score (34.36%) due to the accurate adaptability of the threshold calculated by statistical commonality of the complex urban noise, and the bottleneck problem for current methods of calculating the threshold separating signals and environmental noise accurately without affecting the integrity of signal is solved. Figure 9 gives an example of the manual determination of the onset and the endpoint of the actual signal with the wavelet synchrosqueezed transform (WSST) method, which are chosen as the actual onset and the actual endpoint of actual signal [34], and the error calculations of the onset and endpoint are the same as the simulated data stream. Due to the limitation of the number of correct detections by the LSFM method, 100 samples are collected to calculate the errors of the onset and endpoint for a kind of signal shown in Figure 10.…”

“…The average consumption times of the STFT method (computation complexity generated by short-time Fourier transform) and the LSFM method (computation complexity generated by dividing the geometric mean of the power spectrum by the arithmetic mean of the power spectrum) are 0.0224 s and 0.09343 s, respectively, which can prove that the proposed method is very highly efficient. 9 gives an example of the manual determination of the onset and the endpoint of the actual signal with the wavelet synchrosqueezed transform (WSST) method, which are chosen as the actual onset and the actual endpoint of actual signal [34], and the error calculations of the onset and endpoint are the same as the simulated data stream. Due to the limitation of the number of correct detections by the LSFM method, 100 samples are STFT methods, for high accuracy, the initial thresholds have to choose the maximum of the beginning noise.…”

Fracture Characteristics of Polyurethane Grouting Materials Based on Acoustic Emission Monitoring

“…e acoustic emission (AE) phenomenon has been used as a powerful tool with the purpose to either detect, locate, or assess damage for a wide range of applications [37]. Deformation and crack propagation of materials under stress are important mechanisms of structural failure.…”

A Method for Pipeline Leak Detection Based on Acoustic Imaging and Deep Learning

“…A change in the AE phenomenon increases the energy of the AE signal detected by the AE sensors in the form of hits. These hits can be overwhelmed by interference noises [31]. Continuous wavelet transform (CWT) can be used to analyze leak-related useful hits.…”