Due to an increasing number of bridges approaching the end of their design lives, there is a growing interest by the infrastructure managers on monitoring systems able to provide information on the service capability of structures. Focusing, in particular, on railway bridges, serviceability is related to both structural serviceability and runnability, where the first is related to the structural capability of the bridge, while the second is related to the constraints on rail geometry. In this work the authors present an investigation of the capability of a permanent monitoring system to detect the presence of damages that would lead to an exceedance of the structural threshold values imposed by runnability norms. The target structure is a double-span Warren truss railway bridge, located in Italy, designed in 1946 and currently instrumented with a permanent monitoring system. The latter performs dynamic, quasi-static and static measurements, with sensors ranging from velocimeters to end displacement transducers and clinometers. Thanks to a FE model of the actual structure, a set of damage scenarios can be simulated, defining the damage entities that lead to the exceedance of thresholds imposed by norms in terms of runability assessment. As a result, the installed sensors mesh was found to distinguish such critical structural conditions as well as lower entity damage scenarios. The paper discusses the capability of different sensor configurations and monitoring strategies (“dynamic”, “static”, etc) to detect such damage scenarios, in terms both of “early warning” capability and of minimum time required to detect the anomaly. The same approach can be extended to the analysis with respect to structural serviceability limits.
Read full abstract