Data-Based Prognosis and Monitoring of Civil Infrastructures

The structural health monitoring and assessment tasks of civil infrastructures e.g. bridges, railway tracks are inevitable in order to keep the transportation network active for smooth operation. And due to the complex inherent traits of the aforementioned infrastructures, it requires appropriate strategies to keep them functional. The conventional monitoring approaches (e.g. manual) are becoming unpopular as a result of the advancement of the sensors-based monitoring. Based on the measured data, long-term monitoring is possible regardless the existence of a prior model, though it may not be so straightforward task if a prior model is missing. Therefore, to avail the advantages of the state-of-the-art tools, this study focuses into the possibility of data-based modelling by adopting sub-space method. To do this end, two different experimentally measured sets of data of (i) a laboratory scaled bridge, and (ii) a rail-track’s data have been used to perform analyses. Initially, the performance of the developed models are evaluated and later the models have been validated. The pilot outcome shows the efficacy of the developed model in terms of prediction capability to the measured data. Further, the model has been utilized to forecast future behaviour that will assist the assessment of the unseen future behaviour of the infrastructures.