skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Detectability limitations with 3-D point reconstruction algorithms using digital radiography

Abstract

The estimated impact of pores in clusters on component fatigue will be highly conservative when based on 2-D rather than 3-D pore positions. To 3-D position and size defects using digital radiography and 3-D point reconstruction algorithms in general require a lower inspection time and in some cases work better with planar geometries than X-ray computed tomography. However, the increase in prior assumptions about the object and the defects will increase the intrinsic uncertainty in the resulting nondestructive evaluation output. In this paper this uncertainty arising when detecting pore defect clusters with point reconstruction algorithms is quantified using simulations. The simulation model is compared to and mapped to experimental data. The main issue with the uncertainty is the possible masking (detectability zero) of smaller defects around some other slightly larger defect. In addition, the uncertainty is explored in connection to the expected effects on the component fatigue life and for different amount of prior object-defect assumptions made.

Authors:
 [1]
  1. Department of Materials and Manufacturing Technology, Chalmers University of Technology, Göteborg, 412 96 (Sweden)
Publication Date:
OSTI Identifier:
22391260
Resource Type:
Journal Article
Resource Relation:
Journal Name: AIP Conference Proceedings; Journal Volume: 1650; Journal Issue: 1; Conference: 41. Annual Review of Progress in Quantitative Nondestructive Evaluation, Boise, ID (United States), 20-25 Jul 2014; Other Information: (c) 2015 AIP Publishing LLC; Country of input: International Atomic Energy Agency (IAEA)
Country of Publication:
United States
Language:
English
Subject:
71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS; ALGORITHMS; COMPARATIVE EVALUATIONS; COMPUTERIZED SIMULATION; COMPUTERIZED TOMOGRAPHY; DEFECTS; FATIGUE; INSPECTION; JOINTS; MATHEMATICAL MODELS; NONDESTRUCTIVE ANALYSIS; X RADIATION

Citation Formats

Lindgren, Erik, E-mail: erik.lindgren@chalmers.se. Detectability limitations with 3-D point reconstruction algorithms using digital radiography. United States: N. p., 2015. Web. doi:10.1063/1.4914659.
Lindgren, Erik, E-mail: erik.lindgren@chalmers.se. Detectability limitations with 3-D point reconstruction algorithms using digital radiography. United States. doi:10.1063/1.4914659.
Lindgren, Erik, E-mail: erik.lindgren@chalmers.se. Tue . "Detectability limitations with 3-D point reconstruction algorithms using digital radiography". United States. doi:10.1063/1.4914659.
@article{osti_22391260,
title = {Detectability limitations with 3-D point reconstruction algorithms using digital radiography},
author = {Lindgren, Erik, E-mail: erik.lindgren@chalmers.se},
abstractNote = {The estimated impact of pores in clusters on component fatigue will be highly conservative when based on 2-D rather than 3-D pore positions. To 3-D position and size defects using digital radiography and 3-D point reconstruction algorithms in general require a lower inspection time and in some cases work better with planar geometries than X-ray computed tomography. However, the increase in prior assumptions about the object and the defects will increase the intrinsic uncertainty in the resulting nondestructive evaluation output. In this paper this uncertainty arising when detecting pore defect clusters with point reconstruction algorithms is quantified using simulations. The simulation model is compared to and mapped to experimental data. The main issue with the uncertainty is the possible masking (detectability zero) of smaller defects around some other slightly larger defect. In addition, the uncertainty is explored in connection to the expected effects on the component fatigue life and for different amount of prior object-defect assumptions made.},
doi = {10.1063/1.4914659},
journal = {AIP Conference Proceedings},
number = 1,
volume = 1650,
place = {United States},
year = {Tue Mar 31 00:00:00 EDT 2015},
month = {Tue Mar 31 00:00:00 EDT 2015}
}