Online early damage detection and localisation using multivariate data analysis: Application to a cable‐stayed bridge

Abstract

An online data-based methodology for early damage detection and localisation under the effects of environmental and operational variations (EOVs) is proposed. The methodology is described in detail and implemented in a large prestressed concrete cable-stayed bridge of which 3.5 years of data are available. The effects of EOVs are suppressed by the combined application of two well-established multivariate data analysis methods: multiple linear regression and principal component analysis. Criteria for the systematic choice of the predictor variables and the number of principal components to retain are proposed. Because the bridge is new and sound, the experimental time series are corrupted with numerically simulated damage scenarios in order to evaluate the damage detection ability. It is demonstrated that the sensitivity to damage is increased when daily, 2-day, or 3-day averaged data are used instead of hourly data. The effectiveness of the proposed methodology is also demonstrated with the detection of a real, small, and temporary sensor anomaly.The implemented methodology has revealed to be robust and efficient, presenting a contribution to the transition of structural health monitoring from academia to industry. HIGHLIGHTS• Early damage detection and localisation in a cable-stayed bridge.• Environmental, operational, and long-term variations suppression.• Three years and half of continuous experimental data.• Transition of structural health monitoring technology from academia to industry.KEYWORDS cable-stayed bridges, damage detection, damage localisation, environmental and operational variations, long-term effects, structural health monitoring

Measure point arrangement strategy for in-service continuous girder bridge SHM with consideration of structural robustness (Special Issue of ICAST 2019)

“…Structural health monitoring (SHM) has rapid developed in the last few decades. Several bridge SHM systems have been designed and established, such as the Corgo Bridge (pre-stressed concrete box girder bridge) in Portugal (Sousa Tom et al, 2019), the Jintang Bridge (cable-stayed bridge) in China (Li and Ou, 2015), the Akashi-Kaikyo Bridge (suspension bridge) in Japan (Ab and Fujino, 2017), and the Tamar Suspension Bridge in United Kingdom (Webb et al, 2014). Moreover, dozens of SHM codes or specifications have also been put forward all over the world, such as the guidelines for SHM (Canada, 2001; Germany, 2006) and the design standard for SHM system (China, 2012)(Gatti, 2019).…”

Ensemble learning‐based structural health monitoring by Mahalanobis distance metrics

“…On this basis, the model learned in the training phase is a nonparametric or parametric novelty detector. Parametric novelty detectors are those that need to estimate some unknown parameters such as the number of clusters for a clustering‐based approach, 20 the number of principal components associated with the principal component analysis (PCA)‐based technique, 21 and the numbers of layers and neurons required for an artificial neural network (ANN)‐based method 22 . On the contrary, nonparametric novelty detectors do not require estimating an unknown parameter, in which case those are more beneficial for SHM compared to the parametric novelty detectors.…”

Long‐term performance monitoring and assessment of concrete beam bridges using neutral axis indicator

“…Dynamic damage indicators, such as frequency, mode shape, modal strain energy, etc., are more widely used in SHM. [26][27][28] However, these indicators are not sufficient to precisely detect local damage.…”

“…Static measurement‐based damage indicators include the traditional threshold of strain and acceleration. Dynamic damage indicators, such as frequency, mode shape, modal strain energy, etc., are more widely used in SHM 26–28 . However, these indicators are not sufficient to precisely detect local damage.…”