Identification of fatigue damage evaluation using entropy of acoustic emission waveform

Abstract

Acoustic emission (AE) is a passive nondestructive testing (NDT) technique which is employed to identify critical damage in structures before failure can occur. Currently, AE monitoring is carried out by calculating the features of the signal received by the AE sensor. User-defined acquisition settings (i.e., timing and threshold) significantly affect many traditional AE features such as count, energy, centroid frequency, rise time and duration. In AE monitoring, AE features are strongly related to the damage sources. Therefore, AE features that are calculated due to inaccurate user-defined acquisition settings can result in inaccurately classified damage sources. This work presents a new feature of the signal based on the measure of randomness calculated using second-order Renyi's entropy. The new feature is computed from its discrete amplitude distribution making it independent of acquisition settings. This can reduce the need for human judgement in measuring the feature of the signal. To investigate the effectiveness of the presented feature, fatigue testing is conducted on an un-notched steel sample with simultaneous AE monitoring. Digital image correlation (DIC) is measured alongside AE monitoring to correlate both monitoring methods with material damage. The results suggest that the new feature is sensitive in identifying critical damages in the material.

Acoustic emission waveforms for damage monitoring in composite materials: Shifting in spectral density, entropy and wavelet packet transform

“…22 Several researchers have used Shannon's entropy in processing acoustic waveforms but Renyi's entropy is seldom used. Tanvir et al 23 have used Renyi's entropy of the acoustic waveform in defining the characteristics of the fatigue damage evolution. But the application in which they used Renyi's entropy is debatable.…”

“…But the application in which they used Renyi's entropy is debatable. Tanvir et al 23 used cumulative Renyi's entropy, which is necessarily the sum of the measures of randomness, as an acoustic descriptor for damage evolution.…”

A New Fracture Detection Algorithm of Low Amplitude Acoustic Emission Signal Based on Kalman Filter-Ripple Voltage

“…Hence, an appropriate threshold voltage is set for a typically processed signal, as described in the introduction. It is common to analyze the aforementioned AE parameters based on the threshold voltage [ 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 ].…”

“…The analysis method of such an AE signal can be largely divided into waveform analysis related to the activity and the shape of the signal [ 10 , 15 ]. The AE activity analysis method evaluates various parameters such as the number of events, ring-down count, and maximum amplitude.…”

Statistical analysis of acoustic emission signals and tensile deformation measurement for three dimensional carbon fiber braided composites

“…In regard to the signal processing, AE method requires effective calibration and analysis to discriminate between all sorts of damage modes and failure mechanisms [20][21][22]. Meanwhile, the quantification of irreversible damage is a challenge and crucial aspect to understand the properties of materials, so it is necessary to find ways of quantifying the dynamic behaviors and micro-damage mechanisms [23][24][25]. Qi et al [26] studied the randomly microscopic events (RME) of acrylic bone cement materials under pure tension and three-point bending tests.…”