What the study found
An updated finite element model of the laboratory footbridge matched the measured vibration behaviour more closely than the initial model. The final model reduced natural-frequency discrepancies to less than 8% and showed high agreement in mode shapes.
Why the authors say this matters
The authors say a clearer understanding of the footbridge's fundamental dynamic properties was needed to support later research on structural dynamics. The study suggests that model updating can improve how well the numerical model represents the measured behaviour of the structure.
What the researchers tested
The researchers developed a reconfigurable laboratory-scale footbridge made of two steel girders spliced near midspan and composite deck panels using a sandwich plate system, meaning two steel faceplates bonded by a polyurethane core. They built a preliminary finite element model, carried out experimental modal analysis, tested the spliced beams and deck panels at component level, and then used an optimisation-based model-updating process that adjusted boundary restraints, material properties, and section parameters.
What worked and what didn't
The initial finite element model did not reproduce some vibration modes seen in the experiments. Component testing indicated that the primary beams needed reduced effective stiffness because of the splice connection, while the deck stiffness needed to be increased to reflect the sandwich composite action. After updating, the model achieved strong correlation with the experimental results, including natural frequencies within less than 8% discrepancy and high modal assurance criterion (MAC) agreement for mode shapes.
What to keep in mind
The abstract describes a laboratory-scale footbridge and a specific modelling workflow, so the findings are limited to that test-bed and its measured behaviour. Limitations beyond the noted modelling assumptions, material representation, and boundary conditions are not described in the available summary.
Key points
- A finite element model of the footbridge was updated to better match experimental vibration data.
- The final model reduced natural-frequency discrepancies to less than 8%.
- Mode shapes from the updated model showed high agreement with the test results.
- Component testing suggested lower effective stiffness for the spliced beams and higher stiffness for the deck panels.
- The initial model missed some vibration modes observed in the experiments.
Disclosure
- Research title:
- Updated finite element model matched footbridge vibrations more closely
- Authors:
- Wai Kei Ao, Aleksandar Pavić, James Brownjohn
- Institutions:
- Hong Kong Polytechnic University, University of Exeter
- Publication date:
- 2026-03-09
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.