Abstract

Structural health monitoring is an important problem of interest in many civil infrastructure and aerospace applications. In the last few decades, many techniques have been investigated to address the detection, estimation, and classification of damage in structural components. One of the key challenges in the development of real-world damage identification systems, however, is variability due to changing environmental and operational conditions. Conventional statistical methods based on static modeling frameworks can prove to be inadequate in a dynamic and fast changing environment, especially when a sufficient amount of data is not available. In this paper, a novel adaptive learning structural damage estimation method is proposed in which the stochastic models are allowed to perpetually change with the time-varying conditions. The adaptive learning framework is based on the use of Dirichlet process (DP) mixture models, which provide the capability of automatically adjusting to structure within the data. Specifically, time-frequency features are extracted from periodically collected structural data (measured sensor signals), that are responses to ultrasonic excitation of the material. These are then modeled using a DP mixture model that allows for a growing, possibly infinite, number of mixture components or latent clusters. Combined with input from physically based damage growth models, the adaptively identified clusters are used in a state-space setting to effectively estimate damage states within the structure under varying external conditions. Additionally, a data selection methodology is implemented to enable judicious selection of informative measurements for maximum performance. The utility of the proposed algorithm is demonstrated by application to the estimation of fatigue-induced damage in an aluminum compact tension sample subjected to variable-amplitude cyclic loading.

Original languageEnglish (US)
Pages (from-to)125-143
Number of pages19
JournalJournal of Intelligent Material Systems and Structures
Volume26
Issue number2
DOIs
StatePublished - Jan 30 2015

Fingerprint

Structural health monitoring
Aerospace applications
Stochastic models
Aluminum
Statistical methods
Identification (control systems)
Ultrasonics
Fatigue of materials
Sensors

Keywords

  • active data selection
  • adaptive learning
  • Bayesian filtering
  • damage estimation
  • Dirichlet process mixture
  • Kullback-Leibler divergence
  • Markov chain Monte Carlo methods
  • matching pursuit decomposition
  • probability density function
  • Structural health monitoring
  • time-frequency analysis

ASJC Scopus subject areas

  • Materials Science(all)
  • Mechanical Engineering

Cite this

An adaptive learning damage estimation method for structural health monitoring. / Chakraborty, Debejyo; Kovvali, Narayan; Papandreou-Suppappola, Antonia; Chattopadhyay, Aditi.

In: Journal of Intelligent Material Systems and Structures, Vol. 26, No. 2, 30.01.2015, p. 125-143.

Research output: Contribution to journalArticle

@article{fac79569f2314e73845981788524b46f,
title = "An adaptive learning damage estimation method for structural health monitoring",
abstract = "Structural health monitoring is an important problem of interest in many civil infrastructure and aerospace applications. In the last few decades, many techniques have been investigated to address the detection, estimation, and classification of damage in structural components. One of the key challenges in the development of real-world damage identification systems, however, is variability due to changing environmental and operational conditions. Conventional statistical methods based on static modeling frameworks can prove to be inadequate in a dynamic and fast changing environment, especially when a sufficient amount of data is not available. In this paper, a novel adaptive learning structural damage estimation method is proposed in which the stochastic models are allowed to perpetually change with the time-varying conditions. The adaptive learning framework is based on the use of Dirichlet process (DP) mixture models, which provide the capability of automatically adjusting to structure within the data. Specifically, time-frequency features are extracted from periodically collected structural data (measured sensor signals), that are responses to ultrasonic excitation of the material. These are then modeled using a DP mixture model that allows for a growing, possibly infinite, number of mixture components or latent clusters. Combined with input from physically based damage growth models, the adaptively identified clusters are used in a state-space setting to effectively estimate damage states within the structure under varying external conditions. Additionally, a data selection methodology is implemented to enable judicious selection of informative measurements for maximum performance. The utility of the proposed algorithm is demonstrated by application to the estimation of fatigue-induced damage in an aluminum compact tension sample subjected to variable-amplitude cyclic loading.",
keywords = "active data selection, adaptive learning, Bayesian filtering, damage estimation, Dirichlet process mixture, Kullback-Leibler divergence, Markov chain Monte Carlo methods, matching pursuit decomposition, probability density function, Structural health monitoring, time-frequency analysis",
author = "Debejyo Chakraborty and Narayan Kovvali and Antonia Papandreou-Suppappola and Aditi Chattopadhyay",
year = "2015",
month = "1",
day = "30",
doi = "10.1177/1045389X14522531",
language = "English (US)",
volume = "26",
pages = "125--143",
journal = "Journal of Intelligent Material Systems and Structures",
issn = "1045-389X",
publisher = "SAGE Publications Ltd",
number = "2",

}

TY - JOUR

T1 - An adaptive learning damage estimation method for structural health monitoring

AU - Chakraborty, Debejyo

AU - Kovvali, Narayan

AU - Papandreou-Suppappola, Antonia

AU - Chattopadhyay, Aditi

PY - 2015/1/30

Y1 - 2015/1/30

N2 - Structural health monitoring is an important problem of interest in many civil infrastructure and aerospace applications. In the last few decades, many techniques have been investigated to address the detection, estimation, and classification of damage in structural components. One of the key challenges in the development of real-world damage identification systems, however, is variability due to changing environmental and operational conditions. Conventional statistical methods based on static modeling frameworks can prove to be inadequate in a dynamic and fast changing environment, especially when a sufficient amount of data is not available. In this paper, a novel adaptive learning structural damage estimation method is proposed in which the stochastic models are allowed to perpetually change with the time-varying conditions. The adaptive learning framework is based on the use of Dirichlet process (DP) mixture models, which provide the capability of automatically adjusting to structure within the data. Specifically, time-frequency features are extracted from periodically collected structural data (measured sensor signals), that are responses to ultrasonic excitation of the material. These are then modeled using a DP mixture model that allows for a growing, possibly infinite, number of mixture components or latent clusters. Combined with input from physically based damage growth models, the adaptively identified clusters are used in a state-space setting to effectively estimate damage states within the structure under varying external conditions. Additionally, a data selection methodology is implemented to enable judicious selection of informative measurements for maximum performance. The utility of the proposed algorithm is demonstrated by application to the estimation of fatigue-induced damage in an aluminum compact tension sample subjected to variable-amplitude cyclic loading.

AB - Structural health monitoring is an important problem of interest in many civil infrastructure and aerospace applications. In the last few decades, many techniques have been investigated to address the detection, estimation, and classification of damage in structural components. One of the key challenges in the development of real-world damage identification systems, however, is variability due to changing environmental and operational conditions. Conventional statistical methods based on static modeling frameworks can prove to be inadequate in a dynamic and fast changing environment, especially when a sufficient amount of data is not available. In this paper, a novel adaptive learning structural damage estimation method is proposed in which the stochastic models are allowed to perpetually change with the time-varying conditions. The adaptive learning framework is based on the use of Dirichlet process (DP) mixture models, which provide the capability of automatically adjusting to structure within the data. Specifically, time-frequency features are extracted from periodically collected structural data (measured sensor signals), that are responses to ultrasonic excitation of the material. These are then modeled using a DP mixture model that allows for a growing, possibly infinite, number of mixture components or latent clusters. Combined with input from physically based damage growth models, the adaptively identified clusters are used in a state-space setting to effectively estimate damage states within the structure under varying external conditions. Additionally, a data selection methodology is implemented to enable judicious selection of informative measurements for maximum performance. The utility of the proposed algorithm is demonstrated by application to the estimation of fatigue-induced damage in an aluminum compact tension sample subjected to variable-amplitude cyclic loading.

KW - active data selection

KW - adaptive learning

KW - Bayesian filtering

KW - damage estimation

KW - Dirichlet process mixture

KW - Kullback-Leibler divergence

KW - Markov chain Monte Carlo methods

KW - matching pursuit decomposition

KW - probability density function

KW - Structural health monitoring

KW - time-frequency analysis

UR - http://www.scopus.com/inward/record.url?scp=84920097211&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920097211&partnerID=8YFLogxK

U2 - 10.1177/1045389X14522531

DO - 10.1177/1045389X14522531

M3 - Article

VL - 26

SP - 125

EP - 143

JO - Journal of Intelligent Material Systems and Structures

JF - Journal of Intelligent Material Systems and Structures

SN - 1045-389X

IS - 2

ER -