Interpreting cumulative displacement in a suspension bridge with a physics-based characterisation of environment and roadway/railway loads

In long-span suspension bridges, premature failure of bearings or expansion joints has been a concern during the bridge service life. To implement a reasonable inspection strategy of these critical components, cumulative displacement must be considered for assessing their life consumption and health state. As suspension bridges are subjected to simultaneous multiple loads, it is important to investigate the contributions of different loads to cumulative displacement. This paper develops an approach to interpreting cumulative displacement through the physics-based characterisation of multiple loads in a machine learning framework. The approach correlates wind, temperature, train, and roadway traffic with bridge response using Structural Health Monitoring (SHM) data from a long-span road-rail suspension bridge located in Portugal. The effects of the different loads on the cumulative displacement are quantified both globally and individually with SHapley Additive exPlanations (SHAP). The SHAP interpretation reveals that the trainload is the most influential parameter in the cumulative displacement of bearings and expansion joints, followed by roadway traffic. Besides, cumulative displacement increases with the trainloads and roadway traffic. Wind and temperature proved less important in explaining the cumulative displacement. The contribution of trainloads to the daily cumulative displacement is also calculated to verify its dominating influence. The results can guide the maintenance of relevant bridge components to avoid premature damage.

  • Record URL:
  • Availability:
  • Supplemental Notes:
    • © Springer-Verlag GmbH Germany, part of Springer Nature 2022. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
  • Authors:
  • Publication Date: 2023-3


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01883072
  • Record Type: Publication
  • Files: TRIS
  • Created Date: May 23 2023 10:10AM