Structural Health Monitoring by Means of Strain Field Pattern Recognition on the basis of PCA and Automatic Clustering Techniques Based on SOM

Julián Sierra-Pérez; Miguel A. Torres-Arredondo; Guénaël Cabanes; Alfredo Güeme; Luis E. Mujica

doi:10.1016/j.ifacol.2015.12.259

Structural Health Monitoring by Means of Strain Field Pattern Recognition on the basis of PCA and Automatic Clustering Techniques Based on SOM

Julián Sierra-Pérez, Miguel A. Torres-Arredondo, Guénaël Cabanes, Alfredo Güeme, Luis E. Mujica

Research output: Contribution to journal › Article › Research › peer-review

10 Scopus citations

Abstract

A new methodology to perform Structural Health Monitoring (SHM) in complex structures which is based on Data Driven Models (DDM) by means of strain measurements from Fiber Optic Sensors (FOS), in particular Fiber Bragg Gratings (FBGs), was developed by using Principal Component Analysis (PCA) and automatic clustering techniques based on Self-Organizing Maps (SOM) and density methods. The methodology includes techniques to uncoupling the changes in the strain field caused by the damage occurrence and the change in the operational conditions. Those techniques can be classified as Optimal Baseline Selection (OBS) techniques. Several experiments where performed to develop the methodology and demonstrate the whole concept. Some representative results are presented and discussed.

Original language	English
Pages (from-to)	987-992
Number of pages	6
Journal	IFAC-PapersOnLine
Volume	48
Issue number	28
DOIs	https://doi.org/10.1016/j.ifacol.2015.12.259
State	Published - 2015

Bibliographical note

Funding Information:
⋆ ⋆ The research included in this document was partially supported The research included in this document was partially supported coyortdhiena“tMedinipstreorjieoctdse DCPieIn2c0i1a1-e28I0n3n3o-vCa0ci3ó-n0”2 iannSdpaDinPIt2h0r1o1u-g2h80t3h3e-by the “Ministerio de Ciencia e Innovación” in Spain through the co0o3r-d0i3n;ataendd pbryojetchtes EDuPrIo2p0e1a1n-28C0o3m3-mCi0s3s-i0o2n athnrdouDghPI2th01e1-p2r8o0j3e3ct-coordinated projects DPI2011-28033-C03-02 and DPI2011-28033-SA03R-I0S3T; Uand by the European Commission through the project C03-03;SARISTUand by the European Commission through the project SASARRIISTUSTU

Funding Information:
In this paper results from several experiments are presented. Such results were obtained thanks to the support of several projects. The authors would like to thank the support from the “Ministerio de Ciencia e Innovación” in Spain through the coordinated projects DPI2011-28033-C03-02 and DPI2011-28033-C03-03; and the European Commission through the project SARISTU.

Publisher Copyright:
© 2015

Keywords

Classification
Clustering
Dimensional Reduction
Strain Field, Patterns
Structural Health Monitoring

Access to Document

10.1016/j.ifacol.2015.12.259

Cite this

@article{1a20dca544c44c7c8a8e55b76926cf40,

title = "Structural Health Monitoring by Means of Strain Field Pattern Recognition on the basis of PCA and Automatic Clustering Techniques Based on SOM",

abstract = "A new methodology to perform Structural Health Monitoring (SHM) in complex structures which is based on Data Driven Models (DDM) by means of strain measurements from Fiber Optic Sensors (FOS), in particular Fiber Bragg Gratings (FBGs), was developed by using Principal Component Analysis (PCA) and automatic clustering techniques based on Self-Organizing Maps (SOM) and density methods. The methodology includes techniques to uncoupling the changes in the strain field caused by the damage occurrence and the change in the operational conditions. Those techniques can be classified as Optimal Baseline Selection (OBS) techniques. Several experiments where performed to develop the methodology and demonstrate the whole concept. Some representative results are presented and discussed.",

keywords = "Classification, Clustering, Dimensional Reduction, Strain Field, Patterns, Structural Health Monitoring",

author = "Juli{\'a}n Sierra-P{\'e}rez and Torres-Arredondo, {Miguel A.} and Gu{\'e}na{\"e}l Cabanes and Alfredo G{\"u}eme and Mujica, {Luis E.}",

note = "Publisher Copyright: {\textcopyright} 2015",

year = "2015",

doi = "10.1016/j.ifacol.2015.12.259",

language = "Ingl{\'e}s",

volume = "48",

pages = "987--992",

journal = "IFAC-PapersOnLine",

issn = "1474-6670",

number = "28",

}

TY - JOUR

T1 - Structural Health Monitoring by Means of Strain Field Pattern Recognition on the basis of PCA and Automatic Clustering Techniques Based on SOM

AU - Sierra-Pérez, Julián

AU - Torres-Arredondo, Miguel A.

AU - Cabanes, Guénaël

AU - Güeme, Alfredo

AU - Mujica, Luis E.

PY - 2015

Y1 - 2015

N2 - A new methodology to perform Structural Health Monitoring (SHM) in complex structures which is based on Data Driven Models (DDM) by means of strain measurements from Fiber Optic Sensors (FOS), in particular Fiber Bragg Gratings (FBGs), was developed by using Principal Component Analysis (PCA) and automatic clustering techniques based on Self-Organizing Maps (SOM) and density methods. The methodology includes techniques to uncoupling the changes in the strain field caused by the damage occurrence and the change in the operational conditions. Those techniques can be classified as Optimal Baseline Selection (OBS) techniques. Several experiments where performed to develop the methodology and demonstrate the whole concept. Some representative results are presented and discussed.

AB - A new methodology to perform Structural Health Monitoring (SHM) in complex structures which is based on Data Driven Models (DDM) by means of strain measurements from Fiber Optic Sensors (FOS), in particular Fiber Bragg Gratings (FBGs), was developed by using Principal Component Analysis (PCA) and automatic clustering techniques based on Self-Organizing Maps (SOM) and density methods. The methodology includes techniques to uncoupling the changes in the strain field caused by the damage occurrence and the change in the operational conditions. Those techniques can be classified as Optimal Baseline Selection (OBS) techniques. Several experiments where performed to develop the methodology and demonstrate the whole concept. Some representative results are presented and discussed.

KW - Classification

KW - Clustering

KW - Dimensional Reduction

KW - Strain Field, Patterns

KW - Structural Health Monitoring

UR - http://www.scopus.com/inward/record.url?scp=84988504894&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2015.12.259

DO - 10.1016/j.ifacol.2015.12.259

M3 - Artículo

AN - SCOPUS:84988504894

SN - 1474-6670

VL - 48

SP - 987

EP - 992

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

IS - 28

ER -

Structural Health Monitoring by Means of Strain Field Pattern Recognition on the basis of PCA and Automatic Clustering Techniques Based on SOM

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this