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

Title: On the vulnerability of data-driven structural health monitoring models to adversarial attack

Abstract

Many approaches at the forefront of structural health monitoring rely on cutting-edge techniques from the field of machine learning. Recently, much interest has been directed towards the study of so-called adversarial examples; deliberate input perturbations that deceive machine learning models while remaining semantically identical. This article demonstrates that data-driven approaches to structural health monitoring are vulnerable to attacks of this kind. In the perfect information or ‘white-box’ scenario, a transformation is found that maps every example in the Los Alamos National Laboratory three-storey structure dataset to an adversarial example. Also presented is an adversarial threat model specific to structural health monitoring. The threat model is proposed with a view to motivate discussion into ways in which structural health monitoring approaches might be made more robust to the threat of adversarial attack.

Authors:
ORCiD logo [1];  [2];  [2]; ORCiD logo [2]; ORCiD logo [2]
  1. Industrial Doctorate Centre in Machining Science, The University of Sheffield – Advanced Manufacturing Research Centre with Boeing (AMRC), Rotherham, UK, Dynamics Research Group, The University of Sheffield, Sheffield, UK
  2. Los Alamos National Laboratory, Los Alamos, NM, USA
Publication Date:
Sponsoring Org.:
USDOE
OSTI Identifier:
1630947
Grant/Contract Number:  
89233218CNA000001
Resource Type:
Published Article
Journal Name:
Structural Health Monitoring
Additional Journal Information:
Journal Name: Structural Health Monitoring; Journal ID: ISSN 1475-9217
Publisher:
SAGE Publications
Country of Publication:
United Kingdom
Language:
English

Citation Formats

Champneys, Max David, Green, Andre, Morales, John, Silva, Moisés, and Mascarenas, David. On the vulnerability of data-driven structural health monitoring models to adversarial attack. United Kingdom: N. p., 2020. Web. doi:10.1177/1475921720920233.
Champneys, Max David, Green, Andre, Morales, John, Silva, Moisés, & Mascarenas, David. On the vulnerability of data-driven structural health monitoring models to adversarial attack. United Kingdom. doi:https://doi.org/10.1177/1475921720920233
Champneys, Max David, Green, Andre, Morales, John, Silva, Moisés, and Mascarenas, David. Tue . "On the vulnerability of data-driven structural health monitoring models to adversarial attack". United Kingdom. doi:https://doi.org/10.1177/1475921720920233.
@article{osti_1630947,
title = {On the vulnerability of data-driven structural health monitoring models to adversarial attack},
author = {Champneys, Max David and Green, Andre and Morales, John and Silva, Moisés and Mascarenas, David},
abstractNote = {Many approaches at the forefront of structural health monitoring rely on cutting-edge techniques from the field of machine learning. Recently, much interest has been directed towards the study of so-called adversarial examples; deliberate input perturbations that deceive machine learning models while remaining semantically identical. This article demonstrates that data-driven approaches to structural health monitoring are vulnerable to attacks of this kind. In the perfect information or ‘white-box’ scenario, a transformation is found that maps every example in the Los Alamos National Laboratory three-storey structure dataset to an adversarial example. Also presented is an adversarial threat model specific to structural health monitoring. The threat model is proposed with a view to motivate discussion into ways in which structural health monitoring approaches might be made more robust to the threat of adversarial attack.},
doi = {10.1177/1475921720920233},
journal = {Structural Health Monitoring},
number = ,
volume = ,
place = {United Kingdom},
year = {2020},
month = {5}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
DOI: https://doi.org/10.1177/1475921720920233

Save / Share:

Works referenced in this record:

An approach to decision‐making analysis for implementation of structural health monitoring in bridges
journal, March 2019

  • Neves, Ana C.; Leander, John; González, Ignacio
  • Structural Control and Health Monitoring, Vol. 26, Issue 6
  • DOI: 10.1002/stc.2352

Self-adaptive Differential Evolution Algorithm for Numerical Optimization
conference, January 2005


Structural Health Monitoring With Autoregressive Support Vector Machines
journal, February 2009

  • Bornn, Luke; Farrar, Charles R.; Park, Gyuhae
  • Journal of Vibration and Acoustics, Vol. 131, Issue 2
  • DOI: 10.1115/1.3025827

ICAS: the center of diagnostics and prognostics for the United States Navy
conference, July 2001

  • Finley, Brian; Schneider, Eric A.
  • Aerospace/Defense Sensing, Simulation, and Controls, SPIE Proceedings
  • DOI: 10.1117/12.434237

Vibration–based structural damage identification
journal, January 2001

  • Farrar, Charles R.; Doebling, Scott W.; Nix, David A.
  • Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, Vol. 359, Issue 1778
  • DOI: 10.1098/rsta.2000.0717

The fundamental axioms of structural health monitoring
journal, March 2007

  • Worden, Keith; Farrar, Charles R.; Manson, Graeme
  • Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 463, Issue 2082
  • DOI: 10.1098/rspa.2007.1834

Reducing the Dimensionality of Data with Neural Networks
journal, July 2006


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

Experimental Validation of a Structural Health Monitoring Methodology: part iii. Damage Location on an Aircraft wing
journal, January 2003

  • Manson, G.; Worden, K.; Allman, D.
  • Journal of Sound and Vibration, Vol. 259, Issue 2
  • DOI: 10.1006/jsvi.2002.5169

Damage Assessment Using Neural Networks
journal, January 2003

  • Zapico, J. L.; GonzÁLez, M. P.; Worden, K.
  • Mechanical Systems and Signal Processing, Vol. 17, Issue 1
  • DOI: 10.1006/mssp.2002.1547

Structural health monitoring of offshore wind turbines: A review through the Statistical Pattern Recognition Paradigm
journal, October 2016

  • Martinez-Luengo, Maria; Kolios, Athanasios; Wang, Lin
  • Renewable and Sustainable Energy Reviews, Vol. 64
  • DOI: 10.1016/j.rser.2016.05.085

An outlier analysis framework for impedance-based structural health monitoring
journal, August 2005

  • Park, Gyuhae; Rutherford, Amanda C.; Sohn, Hoon
  • Journal of Sound and Vibration, Vol. 286, Issue 1-2
  • DOI: 10.1016/j.jsv.2004.10.013

The application of machine learning to structural health monitoring
journal, December 2006

  • Worden, Keith; Manson, Graeme
  • Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 365, Issue 1851
  • DOI: 10.1098/rsta.2006.1938