Numerical Model Calibration of a Bridge by Using Inverse Engineering: A Case Study

In recent years, existing transport infrastructures have experienced significant safety criticalities implying expensive repairs, significant business downtime and occasionally demolition with a relevant economic and social impact. Vibration Based Monitoring (VBM) represents a strategic tool for real time assessment of potential damages suffered by civil/strategic structures and infrastructures due to aging and/or after a critical event. As a part of VBM, experimental dynamic identificationmethods allowto characterize modal properties of existing structures both in stationary and non-stationary conditions. The process used to calibrate the numerical model (digital twin) minimizing the errors between considered parameters of real and numerical structure can be considered an inverse engineering application. This paper focuses on the dynamic calibration of a Finite Element (FE) model related to the “Ponte della Musica – Armando Trovajoli” bridge locate in Rome (Italy), built by means design documentation and visual inspection. The model calibration has been performed in terms of eigenfrequencies and equivalent viscous damping factor evaluated by means of accelerometric recordings. Several numerical FEM models were built using different boundary conditions. Uncertainties derived from external constrains and executive details have been accounted by varying stiffness and other mechanical parameters within several parametric analyses with the aim to fit experimental dynamic properties with simulated ones. The present study has been performed within the TaskWP6 “Monitoring and Satellite Data” of the DPC-Reluis 2022–2024 agreement, with the aim to further analyse the influence of temperature on the vertical displacement of monitored bridge and compare them with satellite data information.