Publication
Title
X-ray tube with artificial neural network model as a promising alternative for radioisotope source in radiation based two phase flowmeters
Author
Abstract
In this paper, X-ray tube is introduced as a potential alternative for radioisotope sources used in radiation based liquid-gas two-phase flowmeters. X-ray tubes have lots of advantages over the radioisotope sources such as having an adjustable emitting photon's energy, being safer from point of view of radiation health physics during the transportation of the source, having ability to generate a high flux photon beam, and etc. The proposed radiation based system in this study composes an X-ray tube with a tube voltage of 150 kV and a 2.5 mm aluminum filter as the radiation source and one sodium iodide crystal as the photon detector. A pipe was positioned between the X-ray tube and the detector. Two main flow regimes of annular and stratified with different void fractions were modelled inside the pipe. Artificial neural network model of multi-layer perceptron (MLP) was also used in this study for analyzing the obtained data. The output spectrum of sodium iodide detector with 150 samples was applied as the input of multi-layer perceptron network and void fraction was considered as its output. The root mean squared error of proposed measuring system was 4.13 which shows the X-ray tube can be implemented as a promising alternative for radioisotope in radiation based two phase flow meters.
Language
English
Source (journal)
Applied radiation and isotopes. - Oxford, 1993, currens
Publication
Oxford : 2020
ISSN
0969-8043
DOI
10.1016/J.APRADISO.2020.109255
Volume/pages
164 (2020) , 8 p.
Article Reference
109255
ISI
000560918800014
Pubmed ID
32819501
Medium
E-only publicatie
Full text (Publisher's DOI)
Full text (publisher's version - intranet only)
UAntwerpen
Research group
Publication type
Subject
External links
Web of Science
Record
Identifier
Creation 30.10.2020
Last edited 06.10.2024
To cite this reference