A hybrid experimental-numerical framework for prestressed concrete bridge model validation and sensor placement optimization: a case study

Conducting numerical modal validations of operational bridges are vital for their structural health monitoring (SHM). This type of validation fully depends on the accurate assessment of the dynamic characteristics of bridges. Accordingly, the arrangement and placement of sensors to measure vibration data also play a crucial role in ensuring the effective capturing of both global and localised dynamic behaviours. Hence, a proper sensor layout is required to accurately identify the bridge behaviour. This study investigates the effectiveness of different sensor configurations in identifying modal parameters of a case study bridge, located in New South Wales, Australia using three sensor layouts: full-width coverage and two partial-width configurations, each covering one side of the bridge. The performance of these configurations was evaluated through field testing and numerical validation. The results showed up to 97% similarity for the first mode shape under partial-width layouts. The partial-width configurations demonstrated high accuracy in capturing both global and localized modal responses and were effective in monitoring localized stress variations as they are aligned with traffic flow zones. However, the full-width layout is preferable for global structural assessments and showed limitations in detecting higher-order modes. This study underscores the critical importance of sensor placement in achieving accurate modal analysis and robust numerical validation while also highlighting the comparative advantage of integrating dynamic response metrics into sensor placement strategies.