The fundamental axioms of structural health monitoring

Worden, Keith; Farrar, Charles R.; Manson, G.; Park, Gyuhae

doi:10.1098/rspa.2007.1834

Cited by 436 publications

(298 citation statements)

References 22 publications

Supporting

Mentioning

292

Contrasting

Unclassified

Order By: Relevance

“…Methodologies for structural identification have been studied extensively (Worden et al, 2007;Beck, 2010;Cross et al, 2013;Moon and Catbas, 2013). However, every civil structure is unique due to its form, function, and utility and this requires explicit consideration of uncertainties in decision making.…”

Section: Introductionmentioning

confidence: 99%

Comparing Structural Identification Methodologies for Fatigue Life Prediction of a Highway Bridge

Pai

Nussbaumer

Smith

2018

Front. Built Environ.

View full text Add to dashboard Cite

Accurate measurement-data interpretation leads to increased understanding of structural behavior and enhanced asset-management decision making. In this paper, four datainterpretation methodologies, residual minimization, traditional Bayesian model updating, modified Bayesian model updating (with an L -norm-based Gaussian likelihood func-∞ tion), and error-domain model falsification (EDMF), a method that rejects models that have unlikely differences between predictions and measurements, are compared. In the modified Bayesian model updating methodology, a correction is used in the likelihood function to account for the effect of a finite number of measurements on posterior probability-density functions. The application of these data-interpretation methodologies for condition assessment and fatigue life prediction is illustrated on a highway steel-concrete composite bridge having four spans with a total length of 219 m. A detailed 3D finite-element plate and beam model of the bridge and weigh-in-motion data are used to obtain the time-stress response at a fatigue critical location along the bridge span. The time-stress response, presented as a histogram, is compared to measured strain responses either to update prior knowledge of model parameters using residual minimization and Bayesian methodologies or to obtain candidate model instances using the EDMF methodology. It is concluded that the EDMF and modified Bayesian model updating methodologies provide robust prediction of fatigue life compared with residual minimization and traditional Bayesian model updating in the presence of correlated nonGaussian uncertainty. EDMF has additional advantages due to ease of understanding and applicability for practicing engineers, thus enabling incremental asset-management decision making over long service lives. Finally, parallel implementations of EDMF using grid sampling have lower computations times than implementations using adaptive sampling.

show abstract

Section: Introductionmentioning

confidence: 99%

Comparing Structural Identification Methodologies for Fatigue Life Prediction of a Highway Bridge

Pai

Nussbaumer

Smith

2018

Front. Built Environ.

View full text Add to dashboard Cite

show abstract

“…Following Worden and his colleagues [2], we use unsupervised learning for identifying the existence and location of damage but supervised learning for identifying the type and severity of damage. Our objective at this stage is to explore various machine-learning techniques and note their characteristics so that various combinations of them may be used appropriately in various circumstances The supervised learning techniques investigated are support vector machines (SVMs), naive Bayes classifiers (NBs), and feed-forward neural networks (FNNs).…”

Section: Discussionmentioning

confidence: 99%

“…Worden and his colleagues [2] suggest seven axioms for SHM. The following are those most relevant to this paper.…”

Section: Structural Health Monitoringmentioning

confidence: 99%

“…Worden and his colleagues [2] have formulated seven axioms for SHM that capture general aspects that have emerged in several decades of experience. Of particular interest is their Axiom III, which states that unsupervised learning can be used for identifying the existence and location of damage but identifying the type and severity of damage can only be done with supervised learning.…”

mentioning

confidence: 99%

See 1 more Smart Citation

A Study of Machine Learning Techniques for Detecting and Classifying Structural Damage

Nick¹,

Asamene²,

Bullock³

et al. 2015

IJMLC

View full text Add to dashboard Cite

Abstract-We report on work that is part of the development of an agent-based structural health monitoring system. The data used are acoustic emission signals, and we classify these signals according to source mechanisms, those associated with crack growth being particularly significant. The agents are proxies for communication-and computation-intensive techniques and respond to the situation at hand by determining an appropriate constellation of techniques. It is critical that the system have a repertoire of classifiers with different characteristics so that a combination appropriate for the situation at hand can generally be found. We use unsupervised learning for identifying the existence and location of damage but supervised learning for identifying the type and severity of damage. The supervised learning techniques investigated are support vector machines (SVM), naive Bayes classifiers, and feed-forward neural networks (FNN). The unsupervised learning techniques investigated are k-means (with k equal to 3, 4, 5, and 6) and self-organizing maps (SOM, with 3, 4, 5, and 6 output neurons). For each technique except SOM, we tested versions with and without principal component analysis (PCA) to reduce the dimensionality of the data. We found significant differences in the characteristics of these machine learning techniques, with trade-offs between accuracy and fast classification runtime that can be exploited by the agents in finding appropriate combinations of classification techniques. The approach followed here can be generalized for exploring the characteristics of machine-learning techniques for monitoring various kinds of structures.

show abstract

“…The authors defined SHM procedure in four steps: (i) operation evaluation, (ii) data acquisition, (iii) feature extraction and (iv) decision making and classification. Each of these steps can devote a separate research to find useful and reliable techniques for developing successful SHM methods and address some of the fundamental axioms presented in [4].…”

Section: Introductionmentioning

confidence: 99%

Study on Singular Spectrum Analysis as a data-driven technique for damage diagnosis. Comparison between time and frequency domain.

Garcia

Trendafilova

2018

MATEC Web Conf.

View full text Add to dashboard Cite

Abstract. Vibration-based Structural Health Monitoring methodologies have been developed in many different applications with the aim of damage diagnosis. Recently, purely data-driven methods have been gained popularity because these methods do not assume any linearity or model in their analysis. Data-driven methods use the measured vibration signals as data-input to extract features that can conclude obtain useful information for the damage diagnosis. In this work a methodology based on Singular Spectrum Analysis (SSA) technique is presented which decomposes the measured vibration responses in a certain number of principal components which reveal the rotational patterns at any frequency in the motion. One of the steps of the methodology is to create a reference state where the observations can be compared for damage assessment. The data used to create the reference state determines how the information is represented in the reference state and therefore how meaningful and informative are the feature vectors for damage assessment. This study presents of the effect of the data representation considered on the creation of the reference state when the data is introduced in the time or frequency domain. The results obtained are different depending on the signal representation and hence they should have different interpretation when the state is created based on vibratory signals represented in the time or frequency domain.

show abstract

The fundamental axioms of structural health monitoring

Cited by 436 publications

References 22 publications

Comparing Structural Identification Methodologies for Fatigue Life Prediction of a Highway Bridge

Comparing Structural Identification Methodologies for Fatigue Life Prediction of a Highway Bridge

A Study of Machine Learning Techniques for Detecting and Classifying Structural Damage

Study on Singular Spectrum Analysis as a data-driven technique for damage diagnosis. Comparison between time and frequency domain.

Contact Info

Product

Resources

About