THE DETECTING PROCEDURE OF AE SIGNAL UNDER HIGH BACKGROUND NOISE BASED ON ARTIFICIAL NEURAL NETWORK

NDTnet

WCNDT '96 - New Delhi
Table of Contents

AET - Artificial Neural Networks

THE DETECTING PROCEDURE OF AE SIGNAL UNDER HIGH BACKGROUND NOISE BASED ON ARTIFICIAL NEURAL NETWORK

Ma Yukuan, Li Jialin, Dong Yunzhao
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, Filter, Artificial Neural Network (ANN)

ABSTRACT

This paper present a new method to detect acoustic emission (AE) signal under high background noise employing Artificial Neural Network (ANN). The Method is: when signal-to-noise ratio (SNR) is poor, AE signal can be filtered through ANN. Thus the weak signal can be detected. There is two stages in realizing the neural network. Firstly, we'll study the method using software to simulate the neural network. Secondly, we'll develop the way to realize the neural network using hardware. Part of the content discussed in this paper is the work we've done, the other is what we'll do in the future.

AET - Artificial Neural Networks

Publication Source: Trends in NDE Science & Technology; Proceedings of the 14th World Conference on Non-Destructive Testing, New Delhi, 8-13 December 1996.Vol. 3, pages 1853 - 1858
Publisher: Ashgate Publishing Company