Normalized Curvature Ratio for Damage Detection in Beam-Like Structures

Kliewer, Kaitlyn; Glišić, Branko

doi:10.3389/fbuil.2017.00050

Cited by 10 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(a) High-pass bridge, (b) sensor installation, and (c) sensor location. 31 that are expected to fail was limited to one of the three sensors shown in Figure 13(c). The training dataset generation method and the CNN architecture described in the ''Data recovery method'' section were applied in the same manner.…”

Section: Description Of the Bridgementioning

confidence: 99%

Convolutional neural network–based data recovery method for structural health monitoring

Glišić²,

Kim

et al. 2020

Structural Health Monitoring

Self Cite

View full text Add to dashboard Cite

In this study, a structural response recovery method using a convolutional neural network is proposed. The aim of this study is to restore missing strain structural responses when they cannot be collected due to a sensor fault, data loss, or communication errors. To this end, a convolutional neural network model for data recovery is constructed using the strain monitoring data stably measured before the occurrence of data loss. Under the assumption that specific sensors fail among the multiple sensors installed on a structure, the structural responses of these specific sensors are intentionally excluded and the remaining structural responses are set as the input data of the convolutional neural network. In addition, the intentionally excluded structural responses are set as the output data of the convolutional neural network. In case of a sensor fault, the trained convolutional neural network is used to recover the missing strain responses using functional sensors alone. The applicability of the proposed method is verified by a numerical study on a beam structure and an experimental study on a frame structure. The data recovery performance of the proposed convolutional neural network is discussed according to the number of failed sensors and the types of structural members with the failed sensors. Finally, the field applicability of the proposed method is examined using strain monitoring data measured from an overpass bridge in use over a long period of time.

show abstract

Section: Description Of the Bridgementioning

confidence: 99%

Convolutional neural network–based data recovery method for structural health monitoring

Glišić²,

Kim

et al. 2020

Structural Health Monitoring

Self Cite

View full text Add to dashboard Cite

show abstract

“…The advantage of using a network of strain sensors to determine the movement of a structure is that the same sensor network can also be used to assess other monitoring parameters. These parameters include the neutral axis [66], prestressing forces [67], thermal identity verification [68], and curvature [69]. Therefore, the present study focuses on a strain-based method.…”

Section: A Methodologymentioning

confidence: 99%

A Machine Learning-Based Surrogate Finite Element Model for Estimating Dynamic Response of Mechanical Systems

2023

View full text Add to dashboard Cite

An efficient approach for improving the predictive understanding of dynamic mechanical system variability is developed in this work. The approach requires low model assessment time through the fitting of surrogate models. ML-based surrogate algorithms for finite element analysis (FEA) are developed in this study to accelerate FEA and prevent rerunning complex simulations. The research begins with an overview of the recent novelties in ML algorithms applied to finite element (FE) and other physics-based computational schemes. To predict the time-varying response variables, that is, the displacement of a twodimensional truss structure, a surrogate FE technique based on ML algorithms is developed. In this work, several ML regression algorithms, including decision trees (DTs) and deep neural networks, are developed, and their efficacies are compared. In this study, the ML-based surrogate FE models are able to effectively predict the response of the truss structure in two dimensions over the entire structure. Extreme gradientboosting DTs provide more precise outcomes and outperform other ML algorithms.INDEX TERMS Surrogate modeling, finite element analysis, mechanical system analysis, machine learning, artificial neural networks, random forest trees, gradient boosting regression trees, adaptive boosting trees.

show abstract

“…This is where measuring displacement indirectly can be beneficial, with methods using sensors such as accelerometers (Park et al, 2005), tilt sensors/inclinometers (Hou et al, 2005), or strain sensors (Foss and Haugse, 1995;Davis et al, 1996;Sigurdardottir et al, 2017). The benefit of using network of strain sensors to determine the displacement of a structure, is that the same sensor network can also be used to calculate other monitoring parameters such as the neutral axis (Sigurdardottir and Glisic, 2014), pre-stressing forces (Abdel-Jaber and , thermal signatures (Reilly and Glisic, 2018), and curvature (Kliewer and Glisic, 2017), and that is the reason why this research focuses on strain-based methods.…”

Section: Introductionmentioning

confidence: 99%

A Comparison of Strain-Based Methods for the Evaluation of the Relative Displacement of Beam-Like Structures

Kliewer

Glišić

2019

Front. Built Environ.

Self Cite

View full text Add to dashboard Cite

This research presents an evaluation and comparison of the various strain to displacement transformation methods for a beam-like structure. Displacements can provide useful information for the monitoring and assessment of structural performance, health, and safety. The displacement of a structure is correlated with the curvature of a structure, so any unusual behavior of the structure that alters the curvature will also affect the displacement of the structure. Additionally, monitoring the displacement of a structure is useful for evaluating service limits as excessive displacement are uncomfortable to users and can cause damage to surround structures. Direct displacement monitoring of a real-life structure can be challenging, especially for long-term measurements. Because of this, the focus of this research is on in-direct displacement monitoring based on strain sensors. Several methods to determine displacements from strain measurements have been presented in the literature; this paper provides a quantitative comparison of selected methods using both a static and dynamic analysis, providing the errors of the methods. The methods were applied in a small-scale laboratory test to a beam with fiber Bragg-grating strain sensors, with both static and dynamic loading cases. The experimental displacement results are compared to displacement results from LVDT displacement sensors. Finally, the methods are applied to an existing structure equipped with long-gauge fiber Bragg-grating strain sensors, an in-service highway overpass subjected to vehicle loading. The displacements for the overpass were obtained and compared with the service requirements.

show abstract

Normalized Curvature Ratio for Damage Detection in Beam-Like Structures

Cited by 10 publications

References 24 publications

Convolutional neural network–based data recovery method for structural health monitoring

Convolutional neural network–based data recovery method for structural health monitoring

A Machine Learning-Based Surrogate Finite Element Model for Estimating Dynamic Response of Mechanical Systems

A Comparison of Strain-Based Methods for the Evaluation of the Relative Displacement of Beam-Like Structures

Contact Info

Product

Resources

About