Abstract (EN):
Ageing Metallic Railway Bridges (MRBs) are still widely in use despite being exposed to traffic loads and environmental conditions that differ significantly from their original design assumptions, often incorporating materials that are no longer in use. While these factors tend to make these structures more susceptible to degradation, they continue to deliver essential socioeconomic value as a vital element of railway networks. To ensure their safe operation and extended service life, it is critical to preserve structural integrity by effectively managing important durability risks, with fatigue being a primary concern. Achieving this requires precise characterisation of current traffic volumes and their variation over time, supported by appropriate evaluation and monitoring strategies. Rather than relying solely on normative load models, this study introduces an approach that uses a Bridge Digital Twin (BDT) demonstrator for fatigue assessment and monitoring, while incorporating real traffic data derived from Weigh-In-Motion (WIM) system and supplemented by Machine Learning (ML) techniques. A direct comparison between normative and real traffic inputs revealed significantly different fatigue outcomes which directly affect conclusions regarding fatigue-critical details and decisions that follow. The study illustrates the added value of using real traffic data instead of relying solely on standard fatigue load models in effectively characterising fatigue states of ageing MRBs. Furthermore, the BDT approach allows for a more dynamic and comprehensive fatigue assessment process, raising conventional standards of MRBs evaluation.
Idioma:
Inglês
Tipo (Avaliação Docente):
Científica
Nº de páginas:
22