Quasi-Static Bearing Evaluation and Monitoring—A Case Study

Abstract

Research was conducted on the quasi-static in-service evaluation and long-term monitoring of bridge bearings through a field case study. The challenge with assessment of bearings, in their current state, is that the visual appearance is not a sufficient indicator of performance. In addition, bearings are difficult to monitor long-term in many cases due to their complex non-linearity. The motivation for evaluation and monitoring of these components is that they are critical to the long-term performance of bridges, particularly those with long-spans. At the service level, bearings accommodate thermal movements along with those from live load and wind effects. For extreme events, such as earthquakes, they dissipate energy and reduce the transfer of force into the superstructure. For the presented study, two primary objectives were established. The first was to evaluate the case study bridge bearings in their present state under service loads. Physics-based methods were evaluated using equilibrium and thermoelasticity. The second objective was to identify a long-term monitoring baseline. Artificial neural networks (ANNs) were explored due to the non-linear behavior present at two of the bearings. An ANN was trained with temperature changes to predict longitudinal bearing movement. The overall study illustrates potential techniques (with their limitations) for in-service evaluation and/or long-term monitoring of bridge bearings that have been assessed with structural health monitoring data.