PATTERN RECOGNITION OF THE CHARACTERISTICS OF AE SOURCE USING ARTIFICIAL NEURAL NETWORK

NDTnet

WCNDT '96 - New Delhi
Table of Contents

AET - AET - Signal Processing

PATTERN RECOGNITION OF THE CHARACTERISTICS OF AE SOURCE USING ARTIFICIAL NEURAL NETWORK

Ma Yukuan, Dong Yunzhao, Li Jial in
Research Institute of Non-Destructive, Testing Department of Electrical Engineering
JiLin University of Technology, Changchun-130025, China

Keywords: Acoustic Emission (AE), AE source, AE characteristics, Pattern Recognition, Artificial Neural Network (ANN)

ABSTRACT

This paper presents a new method for identifying the characteristics of acoustic emission (AE) source employing Artificial Neural Network (ANN). The three stages of flaw's development, formation, diffusion, rupture, are taken as the AE source. Sample the AE signal. Analyse its wave form and accomplish the pattern recognition by using ANN. This method can be used to determine if the flaw will go on to diffuse or rupture. It can also be used to distinct different AE signals sent out from different AE sources such as noise, leakage and flaw, etc.

AET - AET - Signal Processing

Publication Source: Trends in NDE Science & Technology; Proceedings of the 14th World Conference on Non-Destructive Testing, New Delhi, 8-13 December 1996.Vol. 4, pages 2559 - 2564
Publisher: Ashgate Publishing Company