Fukunaga-Koontz feature transformation for statistical structural damage detection and hierarchical neuro-fuzzy damage localisation

Simon Hoell, Piotr Omenzetter

Research output: Contribution to journalArticle

2 Citations (Scopus)
6 Downloads (Pure)

Abstract

Considering jointly damage sensitive features (DSFs) of signals recorded by multiple sensors, applying advanced transformations to these DSFs and assessing systematically their contribution to damage detectability and localisation can significantly enhance the performance of structural health monitoring systems. This philosophy is explored here for partial autocorrelation coefficients (PACCs) of acceleration responses. They are interrogated with the help of the linear discriminant analysis based on the Fukunaga-Koontz transformation using datasets of the healthy and selected reference damage states. Then, a simple but efficient fast forward selection procedure is applied to rank the DSF components with respect to statistical distance measures specialised for either damage detection or localisation. For the damage detection task, the optimal feature subsets are identified based on the statistical hypothesis testing. For damage localisation, a hierarchical neuro-fuzzy tool is developed that uses the DSF ranking to establish its own optimal architecture. The proposed approaches are evaluated experimentally on data from non-destructively simulated damage in a laboratory scale wind turbine blade. The results support our claim of being able to enhance damage detectability and localisation performance by transforming and optimally selecting DSFs. It is demonstrated that the optimally selected PACCs from multiple sensors or their Fukunaga-Koontz transformed versions can not only improve the detectability of damage via statistical hypothesis testing but also increase the accuracy of damage localisation when used as inputs into a hierarchical neuro-fuzzy network. Furthermore, the computational effort of employing these advanced soft computing models for damage localisation can be significantly reduced by using transformed DSFs.
Original languageEnglish
Pages (from-to)329-353
Number of pages15
JournalJournal of Sound and Vibration
Volume400
Early online date24 Apr 2017
DOIs
Publication statusPublished - 21 Jul 2017

Fingerprint

Damage detection
Autocorrelation
damage
Soft computing
Structural health monitoring
Sensors
Testing
Discriminant analysis
Wind turbines
Turbomachine blades
autocorrelation
turbine blades
ranking
wind turbines
structural health monitoring
sensors
coefficients

Keywords

  • damage detection
  • damage localisation
  • Fukunaga-Koontz transformation
  • neuro-fuzzy networks
  • optimal feature selection
  • statistical hypothesis testing
  • wind turbines

Cite this

@article{5d1a350ecccb452fb46ba1433cc1409c,
title = "Fukunaga-Koontz feature transformation for statistical structural damage detection and hierarchical neuro-fuzzy damage localisation",
abstract = "Considering jointly damage sensitive features (DSFs) of signals recorded by multiple sensors, applying advanced transformations to these DSFs and assessing systematically their contribution to damage detectability and localisation can significantly enhance the performance of structural health monitoring systems. This philosophy is explored here for partial autocorrelation coefficients (PACCs) of acceleration responses. They are interrogated with the help of the linear discriminant analysis based on the Fukunaga-Koontz transformation using datasets of the healthy and selected reference damage states. Then, a simple but efficient fast forward selection procedure is applied to rank the DSF components with respect to statistical distance measures specialised for either damage detection or localisation. For the damage detection task, the optimal feature subsets are identified based on the statistical hypothesis testing. For damage localisation, a hierarchical neuro-fuzzy tool is developed that uses the DSF ranking to establish its own optimal architecture. The proposed approaches are evaluated experimentally on data from non-destructively simulated damage in a laboratory scale wind turbine blade. The results support our claim of being able to enhance damage detectability and localisation performance by transforming and optimally selecting DSFs. It is demonstrated that the optimally selected PACCs from multiple sensors or their Fukunaga-Koontz transformed versions can not only improve the detectability of damage via statistical hypothesis testing but also increase the accuracy of damage localisation when used as inputs into a hierarchical neuro-fuzzy network. Furthermore, the computational effort of employing these advanced soft computing models for damage localisation can be significantly reduced by using transformed DSFs.",
keywords = "damage detection, damage localisation, Fukunaga-Koontz transformation, neuro-fuzzy networks, optimal feature selection, statistical hypothesis testing, wind turbines",
author = "Simon Hoell and Piotr Omenzetter",
note = "Piotr Omenzetter and Simon Hoell’s work on this paper within the Lloyd’s Register Foundation Centre for Safety and Reliability Engineering at the University of Aberdeen was supported by Lloyd’s Register Foundation. The Foundation helps to protect life and property by supporting engineering-related education, public engagement and the application of research.",
year = "2017",
month = "7",
day = "21",
doi = "10.1016/j.jsv.2017.03.048",
language = "English",
volume = "400",
pages = "329--353",
journal = "Journal of Sound and Vibration",
issn = "0022-460X",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Fukunaga-Koontz feature transformation for statistical structural damage detection and hierarchical neuro-fuzzy damage localisation

AU - Hoell, Simon

AU - Omenzetter, Piotr

N1 - Piotr Omenzetter and Simon Hoell’s work on this paper within the Lloyd’s Register Foundation Centre for Safety and Reliability Engineering at the University of Aberdeen was supported by Lloyd’s Register Foundation. The Foundation helps to protect life and property by supporting engineering-related education, public engagement and the application of research.

PY - 2017/7/21

Y1 - 2017/7/21

N2 - Considering jointly damage sensitive features (DSFs) of signals recorded by multiple sensors, applying advanced transformations to these DSFs and assessing systematically their contribution to damage detectability and localisation can significantly enhance the performance of structural health monitoring systems. This philosophy is explored here for partial autocorrelation coefficients (PACCs) of acceleration responses. They are interrogated with the help of the linear discriminant analysis based on the Fukunaga-Koontz transformation using datasets of the healthy and selected reference damage states. Then, a simple but efficient fast forward selection procedure is applied to rank the DSF components with respect to statistical distance measures specialised for either damage detection or localisation. For the damage detection task, the optimal feature subsets are identified based on the statistical hypothesis testing. For damage localisation, a hierarchical neuro-fuzzy tool is developed that uses the DSF ranking to establish its own optimal architecture. The proposed approaches are evaluated experimentally on data from non-destructively simulated damage in a laboratory scale wind turbine blade. The results support our claim of being able to enhance damage detectability and localisation performance by transforming and optimally selecting DSFs. It is demonstrated that the optimally selected PACCs from multiple sensors or their Fukunaga-Koontz transformed versions can not only improve the detectability of damage via statistical hypothesis testing but also increase the accuracy of damage localisation when used as inputs into a hierarchical neuro-fuzzy network. Furthermore, the computational effort of employing these advanced soft computing models for damage localisation can be significantly reduced by using transformed DSFs.

AB - Considering jointly damage sensitive features (DSFs) of signals recorded by multiple sensors, applying advanced transformations to these DSFs and assessing systematically their contribution to damage detectability and localisation can significantly enhance the performance of structural health monitoring systems. This philosophy is explored here for partial autocorrelation coefficients (PACCs) of acceleration responses. They are interrogated with the help of the linear discriminant analysis based on the Fukunaga-Koontz transformation using datasets of the healthy and selected reference damage states. Then, a simple but efficient fast forward selection procedure is applied to rank the DSF components with respect to statistical distance measures specialised for either damage detection or localisation. For the damage detection task, the optimal feature subsets are identified based on the statistical hypothesis testing. For damage localisation, a hierarchical neuro-fuzzy tool is developed that uses the DSF ranking to establish its own optimal architecture. The proposed approaches are evaluated experimentally on data from non-destructively simulated damage in a laboratory scale wind turbine blade. The results support our claim of being able to enhance damage detectability and localisation performance by transforming and optimally selecting DSFs. It is demonstrated that the optimally selected PACCs from multiple sensors or their Fukunaga-Koontz transformed versions can not only improve the detectability of damage via statistical hypothesis testing but also increase the accuracy of damage localisation when used as inputs into a hierarchical neuro-fuzzy network. Furthermore, the computational effort of employing these advanced soft computing models for damage localisation can be significantly reduced by using transformed DSFs.

KW - damage detection

KW - damage localisation

KW - Fukunaga-Koontz transformation

KW - neuro-fuzzy networks

KW - optimal feature selection

KW - statistical hypothesis testing

KW - wind turbines

U2 - 10.1016/j.jsv.2017.03.048

DO - 10.1016/j.jsv.2017.03.048

M3 - Article

VL - 400

SP - 329

EP - 353

JO - Journal of Sound and Vibration

JF - Journal of Sound and Vibration

SN - 0022-460X

ER -