Loading…
Effects of noise on lamb-mode acoustic-emission arrival times determined by wavelet transform
Precise acoustic emission (AE) signal arrival times of the fundamental Lamb modes can be obtained from the arrival time of the peak wavelet transform (WT) magnitude at a particular frequency of interest. Since these arrival times are not determined from a fixed threshold, they are not affected by di...
Saved in:
Published in: | Journal of acoustic emission 2005-01, Vol.23, p.1 |
---|---|
Main Authors: | , |
Format: | Article |
Language: | English |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Precise acoustic emission (AE) signal arrival times of the fundamental Lamb modes can be obtained from the arrival time of the peak wavelet transform (WT) magnitude at a particular frequency of interest. Since these arrival times are not determined from a fixed threshold, they are not affected by dispersion, attenuation and source amplitude. They also correspond to a single group velocity and lead to source location results more accurate than those obtained by traditional AE location calculations based on threshold-based arrival times. In this research, noisefree finite-element modeled (FEM) AE signals were combined with experimental wideband electronic noise to form noisy signals. Since the noise-free signal was available, the changes in the WT-based arrival times from noise-free to noise-altered signals could be quantitatively evaluated. Several signal-to-noise (S/N) ratios were examined in a statistical fashion for three important types of AE sources at three propagation distances and at different radiation angles. The arrival times determined by WTs were obtained for the two different frequency-mode combinations (A0 at 60 kHz and S0 at 522 kHz) that represent the most energetic portions of the signals in a large 4.7 mm thick aluminum plate. From the arrival times, statistical calculations of linear source location were also studied to evaluate errors in location accuracy caused by noise. Even at S/N ratios as low as 1 to 1, the location error was 2 % or less for a large majority of the cases. The errors in arrival time were also examined using two alternative time-frequency analysis approaches to obtain the arrival times. One algorithm (Choi-Williams) provided significantly improved results for the noisy signals. |
---|---|
ISSN: | 0730-0050 0730-0050 |