A Study for Extracting the Information about Flaws in Ultrasonic Detection Based on NNT Cancellation

DOI

10.2991/smont-19.2019.54How to use a DOI?

Keywords

ultrasonic detection; signal cancellation; neural network technology; cancellation rate (CR)

Abstract

How to extract the information about flaws in ultrasonic detection has been paid attention all the time. Because of strong initial echo, side echo and bottom echo (called “clusters”) existing in the ultrasonic echoes, traditional detection methods are difficult to cancel all clusters at the same time, and the clusters left affect detecting the tiny flaws in the object. In this paper, the neural network technology in machine learning is studied, and applied in signal cancellation. A method for extracting the information about flaws is put forward, in which zero mean, wavelet-denoising and adaptive cancellation based on neural network technology are integrated together. The experimental results show that, compared with traditional detection methods, the method can cancel strong clusters such as initial echo, side echo and bottom echo at the same time, and the CR is bigger, the information about the flaws is reserved better, and it has the real time.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Meiming Feng
AU  - Yicheng Zhang
AU  - Shusheng Liao
AU  - Wenbin Wei
PY  - 2019/04
DA  - 2019/04
TI  - A Study for Extracting the Information about Flaws in Ultrasonic Detection Based on NNT Cancellation
BT  - Proceedings of the 2019 International Conference on Modeling, Simulation, Optimization and Numerical Techniques (SMONT 2019)
PB  - Atlantis Press
SP  - 243
EP  - 247
SN  - 1951-6851
UR  - https://doi.org/10.2991/smont-19.2019.54
DO  - 10.2991/smont-19.2019.54
ID  - Feng2019/04
ER  -