Effect of Sampling Frequency on Acoustic Emission Onset Determination using Fractal Dimension

Pal, Survinder and Akula, Aparna and Ghosh, Ripul and Kumar, Satish and Sardana, H.K. (2015) Effect of Sampling Frequency on Acoustic Emission Onset Determination using Fractal Dimension. The e-Journal of Nondestructive Testing, 20 (6). pp. 1-6. ISSN 1435-4934

Full text not available from this repository.
Official URL: http://www.ndt.net/search/docs.php3?showForm=off&i...

Abstract

Acoustic emission is the generation of transient stress waves by rapid release of energy from localized sources giving rise to a versatile non-destructive testing technique. Numerous researchers have tried to exploit the fractal nature of acoustic emission for analysis purposes. The analysis of acoustic emission signature typically consists of four stages namely damage detection, damage localization, damage characterization and health prediction. The determination of arrival of first energy of particular phase at sensor, called onset determination, is an important prerequisite for source localization and source mechanism analysis because the accuracy in determination of onset time directly translates to accuracy in source localization and source mechanism analysis. However, the effect of acquisition parameters on the analysis is not well understood. In this present work, the objective is to understand the effect of one of the important acquisition parameters, sampling frequency. Choice of low sampling frequency might lead to loss of signal information while a high sampling frequency increases storage and computation, hence identifying the appropriate sampling frequency is a critical factor for onset determination. In the present work, the effect of sampling frequency on onset determination using fractal dimension has been assessed. Two open source datasets available at Acoustic emission portal set-up by Muravin have been employed for the research work. The first dataset pertains to a concrete beam while the second data is generated from a carbon fiber woven fabric plate. The acoustic emission events have been generated by using pencil lead break tests. The sampling frequency has been varied synthetically from 0.8 to 2.0 MHz in the steps of 100 kHz. The onset is determined for each individual value of sampling acquisition, by detecting change of fractal dimension estimated using Higuchi’s method. It has been observed by analyzing the onset values determined at different sampling frequencies that the onset does not always converge to the onset having high accuracy with increase of sampling frequency. This observation has been further verified by calculating Spearman Rank correlation values. Also, the results depict the dependence of technique on material in which emissions are generated. The results may be useful for the design of instrumentation for acoustic emission as well as development of computational methods for analysis.

Item Type: Article
Uncontrolled Keywords: Acoustic Emission, Fractal Dimension, Higuchi’s method, Onset Determination,Sampling Frequency
Subjects: CSIO > Computational Instrumentation
Divisions: Computational Instrumentation
Depositing User: Ms. Jyotsana
Date Deposited: 09 Aug 2018 11:56
Last Modified: 09 Aug 2018 11:56
URI: http://csioir.csio.res.in/id/eprint/662

Actions (login required)

View Item View Item