Analysis of Acoustic Emission Signal for Crack Detection and Distance Measurement on Steel Structure

Abstract

Acoustic emission (AE) technique has been merged to a promising method for structural health monitoring in non-destructive technique. So an analysis of the AE signal is becoming a very important research component. In this paper, an algorithm is developed for detection of the crack signal among different noise signals since the AE signal is also generated by several means like any impact or rubbing action on the structure which may give erroneous results. An AE monitoring system is developed with three experimental setups to generate three types of AE signals from three dissimilar sources. Thus, an algorithm is developed to identify the crack signal by comparing the parameters of different signals acquired from different sources using some signal processing techniques such as parameter based analysis, waveform based analysis e.g. fast Fourier transform, continuous wavelet transform, cross-correlation coefficient, magnitude coherence coefficient, and energy distribution. After identification of the crack signal, the distance of the crack source has been calculated by analysing the signal in time–frequency domain also an algorithm has been designed to calculate the velocity of the acoustic wave more accurately and consequently the distance of the crack.