Monitoring the Condition of Narrow Bridges Using Data from Rotation-Based and Strain-Based Bridge Weigh-in-Motion Systems

Abstract

This study investigates the concept that, when localised bridge superstructure damage is present, the weights of passing vehicles inferred from a rotation-based Bridge Weigh-in-Motion (B-WIM) system will deviate from those predicted by a strain-based B-WIM system. Rotation measurements were gathered from a 5.4 m simply-supported laboratory model bridge crossed repeatedly by a 4-axle model vehicle. Damage was incorporated by bolting stiffening plates onto localised sections of the test bridge. This 'negative damage' concept, consisting of local increases in stiffness, allows several damage scenarios to be investigated on the test bridge. The experimental results show that when negative damage is present, rotation-calculated Gross Vehicle Weights decrease while strain-calculated Gross Vehicle Weights remain unchanged. It is also shown that the approximate location of damage can be identified.

AIOps Architecture in Data Center Site Infrastructure Monitoring

“…When the results of these algorithms are incorrect, one can also correct the results of the data labels. It is also necessary to classify these data by anomaly type [ 14 ]. In general, these types of anomalous data are derived and different strategies are adopted to deal with them, as shown in Table 4 .…”

“…It is also necessary to classify these data by anomaly type [ 14 ]. In general, these types of anomalous data are derived and different strategies are adopted to deal with them, as shown in Table 4 .…”

Application of bridge weigh-in-motion system in bridge health monitoring: a state-of-the-art review

“…Huseynov et al 73 utilized a 2D model similar to Figure 12(b). It is made up of a 1D beam simply supported over 6 m span and five transverse point loads to mimic a 5-axle vehicle.…”

“…11,12 Hence, there is a possibility of obtaining an inaccurate estimation of GVW using strain-based B-WIM system depending on the damage location. To check the effect of damage near strain sensor, Huseynov et al 73 plotted the response captured by strain sensor at 1/4 L as shown in Figure 17(c). It is found that the strain sensor provides much higher response time history than that of undamaged state.…”

“…Hence, combining more number of sensor readings can improve the accuracy of damage monitoring. Huseynov et al 73 further continued the numerical investigation to study the influence of environmental condition. The variation in Young’s modulus is taken into account to replicate the influence of temperature.…”