Damage detection of a highway bridge under severe temperature changes using extended Kalman filter trained neural network

Jin, Chenhao; Jang, Shinae; Sun, Xiaojun; Li, Jingcheng; Christenson, Richard

doi:10.1007/s13349-016-0173-8

Cited by 77 publications

(37 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This is indeed one of the principal challenges to transit SHM technology from research to practice. Jin et al [4] proposed an extended Kalmar filter-based artificial neural network for damage detection in a highway bridge under severe temperature changes. The time-lagged natural frequencies, time-lagged temperature and season index are selected as the inputs for the neural network, which predicts the natural frequency at the next time step.…”

Section: Literature Reviewmentioning

confidence: 99%

Structural health monitoring of bridges: a model-free ANN-based approach to damage detection

Neves

González

Leander

et al. 2017

J Civil Struct Health Monit

179

View full text Add to dashboard Cite

As civil engineering structures are growing in dimension and longevity, there is an associated increase in concern regarding the maintenance of such structures. Bridges, in particular, are critical links in today's transportation networks and hence fundamental for the development of society. In this context, the demand for novel damage detection techniques and reliable structural health monitoring systems is currently high. This paper presents a model-free damage detection approach based on machine learning techniques. The method is applied to data on the structural condition of a fictitious railway bridge gathered in a numerical experiment using a three-dimensional finite element model. Data are collected from the dynamic response of the structure, which is simulated in the course of the passage of a train, considering the bridge in healthy and two different damaged scenarios. In the first stage of the proposed method, artificial neural networks are trained with an unsupervised learning approach with input data composed of accelerations gathered on the healthy bridge. Based on the acceleration values at previous instants in time, the networks are able to predict future accelerations. In the second stage, the prediction errors of each network are statistically characterized by a Gaussian process that supports the choice of a damage detection threshold. Subsequent to this, by comparing damage indices with said threshold, it is possible to discriminate between different structural conditions, namely between healthy and damaged. From here and for each damage case scenario, receiver operating characteristic curves that illustrate the trade-off between true and false positives can be obtained. Lastly, based on the Bayes' Theorem, a simplified method for the calculation of the expected total cost of the proposed strategy, as a function of the chosen threshold, is suggested.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Structural health monitoring of bridges: a model-free ANN-based approach to damage detection

Neves

González

Leander

et al. 2017

J Civil Struct Health Monit

179

View full text Add to dashboard Cite

show abstract

“…While the dynamic properties of the lab specimen showed a very slightly dependence on temperature, real monitored bridges showed clear increasing or decreasing trends of the natural frequencies, depending on modes, with the temperature increase. Jin et al [30] proposed a new damage detection method, using artificial neural network and an extended Kalman filter, for damage identification in a composite steel girder bridge under severe temperature changes, also considering freezing effects. They found that the natural frequencies decrease when temperature increases, and vice versa.…”

Section: Introductionmentioning

confidence: 99%

Effect of Environmental Conditions on the Modal Response of a 10‐Story Reinforced Concrete Tower

Regni

Arezzo

Carbonari

et al. 2018

Shock and Vibration

View full text Add to dashboard Cite

We analyse the effect of temperature and wind velocity on the natural frequencies and modal damping ratios of the Faculty of Engineering Tower at the Università Politecnica delle Marche, a 10-story reinforced concrete frame building, permanently monitored with low-noise accelerometers. The data recorded over the first 5 months of monitoring demonstrate that temperature variations and wind intensity have a clear effect on the first three natural frequencies and the corresponding damping ratios. Temperature is positively correlated to the first and second frequencies, corresponding to shear displacement modes and negatively correlated to the third frequency, corresponding to a torsional mode. All frequencies are positively correlated to wind velocity and changes in damping ratios are inversely correlated to any change in frequency. A mechanical explanation of these phenomena is offered, based on a critical review of literature case studies. These results suggest that using changes in modal parameters for damage detection always requires accurate knowledge of the correlation between modal parameters and environmental quantities (temperature, humidity, and wind velocity), an information which is only available through long-term continuous monitoring of the structural response.

show abstract

“…Jin et al [121] conducted further work on the Meriden Bridge in a study that aimed to model the frequency-temperature relationship using an extended Kalman filter for NN learning, which attains a superior convergence speed and success than other common NN training algorithms, such as back propagation. The extended Kalman filter works as a second-order algorithm for recursive state estimation in nonlinear dynamic systems, and as NN training can be considered as a nonlinear estimation problem, the two work together in synergy.…”

Section: Pattern Recognition Methodsmentioning

confidence: 99%

A State of the Art Review of Modal-Based Damage Detection in Bridges: Development, Challenges, and Solutions

2017

View full text Add to dashboard Cite

Abstract:Traditionally, damage identification techniques in bridges have focused on monitoring changes to modal-based Damage Sensitive Features (DSFs) due to their direct relationship with structural stiffness and their spatial information content. However, their progression to real-world applications has not been without its challenges and shortcomings, mainly stemming from: (1) environmental and operational variations; (2) inefficient utilization of machine learning algorithms for damage detection; and (3) a general over-reliance on modal-based DSFs alone. The present paper provides an in-depth review of the development of modal-based DSFs and a synopsis of the challenges they face. The paper then sets out to addresses the highlighted challenges in terms of published advancements and alternatives from recent literature.

show abstract

Damage detection of a highway bridge under severe temperature changes using extended Kalman filter trained neural network

Cited by 77 publications

References 22 publications

Structural health monitoring of bridges: a model-free ANN-based approach to damage detection

Structural health monitoring of bridges: a model-free ANN-based approach to damage detection

Effect of Environmental Conditions on the Modal Response of a 10‐Story Reinforced Concrete Tower

A State of the Art Review of Modal-Based Damage Detection in Bridges: Development, Challenges, and Solutions

Contact Info

Product

Resources

About