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

Title: Automated detection of corrosion in used nuclear fuel dry storage canisters using residual neural networks

Abstract

Nondestructive evaluation methods play an important role in ensuring component integrity and safety in many industries. Operator fatigue can play a critical role in the reliability of such methods. This is important for inspecting high value assets or assets with a high consequence of failure, such as aerospace and nuclear components. Recent advances in convolution neural networks can support and automate these inspection efforts. Here, we propose using residual neural networks (ResNets) for real-time detection of corrosion, including iron oxide discoloration, pitting and stress corrosion cracking, in dry storage stainless steel canisters housing used nuclear fuel. The proposed approach crops nuclear canister images into smaller tiles, trains a ResNet on these tiles, and classifies images as corroded or intact using the per-image count of tiles predicted as corroded by the ResNet. The results demonstrate that such a deep learning approach allows to detect the locus of corrosion via smaller tiles, and at the same time to infer with high accuracy whether an image comes from a corroded canister. Thereby, the proposed approach holds promise to automate and speed up nuclear fuel canister inspections, to minimize inspection costs, and to partially replace human-conducted onsite inspections, thus reducing radiation doses to personnel.

Authors:
; ; ; ; ; ; ; ; ; ;
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Nuclear Physics (NP)
OSTI Identifier:
1644334
Alternate Identifier(s):
OSTI ID: 1787978
Grant/Contract Number:  
AC05-00OR22725
Resource Type:
Published Article
Journal Name:
Nuclear Engineering and Technology
Additional Journal Information:
Journal Name: Nuclear Engineering and Technology Journal Volume: 53 Journal Issue: 2; Journal ID: ISSN 1738-5733
Publisher:
Elsevier
Country of Publication:
Korea, Republic of
Language:
English
Subject:
73 NUCLEAR PHYSICS AND RADIATION PHYSICS; 97 MATHEMATICS AND COMPUTING; convolutional neural networks; corrosion; deep learning; dry storage canisters; feature detection; residual neural networks

Citation Formats

Papamarkou, Theodore, Guy, Hayley, Kroencke, Bryce, Miller, Jordan, Robinette, Preston, Schultz, Daniel, Hinkle, Jacob, Pullum, Laura, Schuman, Catherine, Renshaw, Jeremy, and Chatzidakis, Stylianos. Automated detection of corrosion in used nuclear fuel dry storage canisters using residual neural networks. Korea, Republic of: N. p., 2021. Web. https://doi.org/10.1016/j.net.2020.07.020.
Papamarkou, Theodore, Guy, Hayley, Kroencke, Bryce, Miller, Jordan, Robinette, Preston, Schultz, Daniel, Hinkle, Jacob, Pullum, Laura, Schuman, Catherine, Renshaw, Jeremy, & Chatzidakis, Stylianos. Automated detection of corrosion in used nuclear fuel dry storage canisters using residual neural networks. Korea, Republic of. https://doi.org/10.1016/j.net.2020.07.020
Papamarkou, Theodore, Guy, Hayley, Kroencke, Bryce, Miller, Jordan, Robinette, Preston, Schultz, Daniel, Hinkle, Jacob, Pullum, Laura, Schuman, Catherine, Renshaw, Jeremy, and Chatzidakis, Stylianos. Mon . "Automated detection of corrosion in used nuclear fuel dry storage canisters using residual neural networks". Korea, Republic of. https://doi.org/10.1016/j.net.2020.07.020.
@article{osti_1644334,
title = {Automated detection of corrosion in used nuclear fuel dry storage canisters using residual neural networks},
author = {Papamarkou, Theodore and Guy, Hayley and Kroencke, Bryce and Miller, Jordan and Robinette, Preston and Schultz, Daniel and Hinkle, Jacob and Pullum, Laura and Schuman, Catherine and Renshaw, Jeremy and Chatzidakis, Stylianos},
abstractNote = {Nondestructive evaluation methods play an important role in ensuring component integrity and safety in many industries. Operator fatigue can play a critical role in the reliability of such methods. This is important for inspecting high value assets or assets with a high consequence of failure, such as aerospace and nuclear components. Recent advances in convolution neural networks can support and automate these inspection efforts. Here, we propose using residual neural networks (ResNets) for real-time detection of corrosion, including iron oxide discoloration, pitting and stress corrosion cracking, in dry storage stainless steel canisters housing used nuclear fuel. The proposed approach crops nuclear canister images into smaller tiles, trains a ResNet on these tiles, and classifies images as corroded or intact using the per-image count of tiles predicted as corroded by the ResNet. The results demonstrate that such a deep learning approach allows to detect the locus of corrosion via smaller tiles, and at the same time to infer with high accuracy whether an image comes from a corroded canister. Thereby, the proposed approach holds promise to automate and speed up nuclear fuel canister inspections, to minimize inspection costs, and to partially replace human-conducted onsite inspections, thus reducing radiation doses to personnel.},
doi = {10.1016/j.net.2020.07.020},
journal = {Nuclear Engineering and Technology},
number = 2,
volume = 53,
place = {Korea, Republic of},
year = {2021},
month = {2}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
https://doi.org/10.1016/j.net.2020.07.020

Save / Share:

Works referenced in this record:

A logical calculus of the ideas immanent in nervous activity
journal, December 1943

  • McCulloch, Warren S.; Pitts, Walter
  • The Bulletin of Mathematical Biophysics, Vol. 5, Issue 4
  • DOI: 10.1007/BF02478259

Gradient-based learning applied to document recognition
journal, January 1998

  • Lecun, Y.; Bottou, L.; Bengio, Y.
  • Proceedings of the IEEE, Vol. 86, Issue 11
  • DOI: 10.1109/5.726791

Vision-based detection of loosened bolts using the Hough transform and support vector machines
journal, November 2016


Improvement of Crack-Detection Accuracy Using a Novel Crack Defragmentation Technique in Image-Based Road Assessment
journal, January 2016


The perceptron: A probabilistic model for information storage and organization in the brain.
journal, January 1958


Regionally Enhanced Multiphase Segmentation Technique for Damaged Surfaces: Regionally enhanced multiphase segmentation technique for damaged surfaces
journal, September 2014

  • O'Byrne, Michael; Ghosh, Bidisha; Schoefs, Franck
  • Computer-Aided Civil and Infrastructure Engineering, Vol. 29, Issue 9
  • DOI: 10.1111/mice.12098

Deep Learning-Based Crack Damage Detection Using Convolutional Neural Networks: Deep learning-based crack damage detection using CNNs
journal, March 2017

  • Cha, Young-Jin; Choi, Wooram; Büyüköztürk, Oral
  • Computer-Aided Civil and Infrastructure Engineering, Vol. 32, Issue 5
  • DOI: 10.1111/mice.12263

Vision-Based Automated Crack Detection for Bridge Inspection: Vision-based automated crack detection for bridge inspection
journal, May 2015

  • Yeum, Chul Min; Dyke, Shirley J.
  • Computer-Aided Civil and Infrastructure Engineering, Vol. 30, Issue 10
  • DOI: 10.1111/mice.12141

Nonlinear total variation based noise removal algorithms
journal, November 1992


NB-CNN: Deep Learning-Based Crack Detection Using Convolutional Neural Network and Naïve Bayes Data Fusion
journal, May 2018

  • Chen, Fu-Chen; Jahanshahi, Mohammad R.
  • IEEE Transactions on Industrial Electronics, Vol. 65, Issue 5
  • DOI: 10.1109/TIE.2017.2764844

Approximation by superpositions of a sigmoidal function
journal, December 1989

  • Cybenko, G.
  • Mathematics of Control, Signals, and Systems, Vol. 2, Issue 4
  • DOI: 10.1007/BF02551274

Defect Detection in Reinforced Concrete Using Random Neural Architectures: Defect detection in reinforced concrete using random neural architectures
journal, August 2013

  • Butcher, J. B.; Day, C. R.; Austin, J. C.
  • Computer-Aided Civil and Infrastructure Engineering, Vol. 29, Issue 3
  • DOI: 10.1111/mice.12039

Simulation of self-organizing systems by digital computer
journal, September 1954

  • Farley, B.; Clark, W.
  • Transactions of the IRE Professional Group on Information Theory, Vol. 4, Issue 4
  • DOI: 10.1109/TIT.1954.1057468

Backpropagation Applied to Handwritten Zip Code Recognition
journal, December 1989