Bearing assessment tool for longitudinal bridge performance

Abstract

This work provides an unsupervised learning approach based on a single-valued performance indicator to monitor the global behavior of critical components in a viaduct, such as bearings. We propose an outlier detection method for longitudinal displacements to assess the behavior of a singular asymmetric prestressed concrete structure with a 120 m high central pier acting as a fixed point. We first show that the available long-term horizontal displacement measurements recorded during the undamaged state exhibit strong correlations at the different locations of the bearings. Thus, we combine measurements from four sensors to design a robust performance indicator that is only weakly affected by temperature variations after the application of principal component analysis. We validate the method and show its efficiency against false positives and negatives using several metrics: accuracy, precision, recall, and F1 score. Due to its unsupervised learning scope, the proposed technique is intended to serve as a real-time supervision tool that complements maintenance inspections. It aims to provide support for the prioritization and postponement of maintenance actions in bridge management.

Deep learning enhanced principal component analysis for structural health monitoring

“…But if these devices lose their sliding properties, large displacements and subsequent cracking can occur at the pier caps. Garcia-Sanchez et al 16 provide further details. Figure 2 shows the profile of the bridge.

Figure 2.Beltran Bridge profile with sensor location.

…”

“…For each level of damage, we duplicate the test data set and apply the reduction factor to the second half of the measurements. 16 Figure 6

Figure 6.Beltran Bridge testing control charts.

compares the control charts for the three damage levels. The shadowed regions in Figure 6 cover the part of the testing affected by damage.…”

“…[14][15][16][17] Mujica et al 18 use the residual between a sample and its corresponding projection onto the model of principal components as a novelty indicator for structural condition assessment. Garcia-Sanchez et al 16 proposed an outlier detection method to assess the behavior against horizontal loads of a bridge in Mexico. The goal was to detect malfunctions in the sliding bearings that allow the relative displacement between the deck and the piers.…”

“…We apply this methodology to two real structures under service. The first one employs the same dataset as Garcia-Sanchez et al 16 for the Beltran bridge in Mexico. The second case of study includes multivariate sensor data from a continuous monitoring campaign of five years of service life of bridge Infante D. Henrique in Porto, Portugal.…”

“…13 There are some works that employ this methodology for outlier detection purposes. [14][15][16][17] Mujica et al 18 use the residual between a sample and its corresponding projection onto the model of principal components as a novelty indicator for structural condition assessment. Garcia-Sanchez et al 16 proposed an outlier detection method to assess the behavior against horizontal loads of a bridge in Mexico.…”