Develop In-Flight Acoustic Emission Monitoring of Aircraft to Detect Fatigue Crack Growth

Pattern Recognition Analysis of Acoustic Emission Data from 7075-T651 Aluminum Simulated Joint Specimens

“…We followed the procedure established in our analysis of the data from plate specimens and generated statistical measures of the time domain waveform (including the standard deviation, skewness, and kurtosis), statistical measures of the power spectrum (including the first three moments), the frequency of maximum response, the level of response at that frequency, and the first fifty autocorrelation lags (Hutton, et al, 1981). These features were all computed using the first 4000 points of the sensor P waveforms.…”

“…At least year's Review of Progress we presented data analysis results for data from a plate specimen of 2024-T85l aluminum (Hutton, et al, 1981). We reported a success rate of between 83 and 90% in correctly classifying crack growth acoustic emissions (AE) and noise signals.…”

“…Our data analysis effort followed the general approach described previously (Doctor, et al, 1979;Hutton et al, 1981). Briefly, that approach is to analyze independent subsets of a large training set to determine variation in feature selection and decision rule performance.…”

SHM reliability and implementation – A personal military aviation perspective

“…The earliest documentation of the concept of SHM for aviation can be found in publications from 1980 [7]. Most early work focused on the use of acoustic emission (AE) techniques to detect crack growth.…”

Acoustic Emission from a Crack Growth Event

“…Previous laboratory studies [1) have successfully characterized acoustic emission signals using statistical pattern recognition algorithms to characterize signals on the basis of features observed in a large number of events which can be assumed valid. Normally such techniques extract features which are dependent both upon the emission event and the response of the surrounding structure to that event.…”