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

Title: Ensembles of novelty detection classifiers for structural health monitoring using guided waves

Abstract

Guided wave structural health monitoring uses sparse sensor networks embedded in sophisticated structures for defect detection and characterization. The biggest challenge of those sensor networks is developing robust techniques for reliable damage detection under changing environmental and operating conditions. To address this challenge, we develop a novelty classifier for damage detection based on one class support vector machines. We identify appropriate features for damage detection and introduce a feature aggregation method which quadratically increases the number of available training observations.We adopt a two-level voting scheme by using an ensemble of classifiers and predictions. Each classifier is trained on a different segment of the guided wave signal, and each classifier makes an ensemble of predictions based on a single observation. Using this approach, the classifier can be trained using a small number of baseline signals. We study the performance using monte-carlo simulations of an analytical model and data from impact damage experiments on a glass fiber composite plate.We also demonstrate the classifier performance using two types of baseline signals: fixed and rolling baseline training set. The former requires prior knowledge of baseline signals from all environmental and operating conditions, while the latter does not and leverages the fact that environmental andmore » operating conditions vary slowly over time and can be modeled as a Gaussian process.« less

Authors:
ORCiD logo; ; ; ; ;
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1422274
Report Number(s):
PNNL-SA-129076
Journal ID: ISSN 0964-1726
DOE Contract Number:  
AC05-76RL01830
Resource Type:
Journal Article
Resource Relation:
Journal Name: Smart Materials and Structures; Journal Volume: 27; Journal Issue: 1
Country of Publication:
United States
Language:
English

Citation Formats

Dib, Gerges, Karpenko, Oleksii, Koricho, Ermias, Khomenko, Anton, Haq, Mahmoodul, and Udpa, Lalita. Ensembles of novelty detection classifiers for structural health monitoring using guided waves. United States: N. p., 2017. Web. doi:10.1088/1361-665X/aa973f.
Dib, Gerges, Karpenko, Oleksii, Koricho, Ermias, Khomenko, Anton, Haq, Mahmoodul, & Udpa, Lalita. Ensembles of novelty detection classifiers for structural health monitoring using guided waves. United States. doi:10.1088/1361-665X/aa973f.
Dib, Gerges, Karpenko, Oleksii, Koricho, Ermias, Khomenko, Anton, Haq, Mahmoodul, and Udpa, Lalita. Fri . "Ensembles of novelty detection classifiers for structural health monitoring using guided waves". United States. doi:10.1088/1361-665X/aa973f.
@article{osti_1422274,
title = {Ensembles of novelty detection classifiers for structural health monitoring using guided waves},
author = {Dib, Gerges and Karpenko, Oleksii and Koricho, Ermias and Khomenko, Anton and Haq, Mahmoodul and Udpa, Lalita},
abstractNote = {Guided wave structural health monitoring uses sparse sensor networks embedded in sophisticated structures for defect detection and characterization. The biggest challenge of those sensor networks is developing robust techniques for reliable damage detection under changing environmental and operating conditions. To address this challenge, we develop a novelty classifier for damage detection based on one class support vector machines. We identify appropriate features for damage detection and introduce a feature aggregation method which quadratically increases the number of available training observations.We adopt a two-level voting scheme by using an ensemble of classifiers and predictions. Each classifier is trained on a different segment of the guided wave signal, and each classifier makes an ensemble of predictions based on a single observation. Using this approach, the classifier can be trained using a small number of baseline signals. We study the performance using monte-carlo simulations of an analytical model and data from impact damage experiments on a glass fiber composite plate.We also demonstrate the classifier performance using two types of baseline signals: fixed and rolling baseline training set. The former requires prior knowledge of baseline signals from all environmental and operating conditions, while the latter does not and leverages the fact that environmental and operating conditions vary slowly over time and can be modeled as a Gaussian process.},
doi = {10.1088/1361-665X/aa973f},
journal = {Smart Materials and Structures},
number = 1,
volume = 27,
place = {United States},
year = {Fri Nov 17 00:00:00 EST 2017},
month = {Fri Nov 17 00:00:00 EST 2017}
}