Finite Elements Model Updating: The case study of a bridge subjected to scouring

Finite Elements Model Updating (FEMU) is an advanced technique that has been applied for decades in Civil Engineering and used to improve the accuracy of Finite Elements Models (FEMs) by updating certain model parameters on the basis of experimental data. The application of this methodology to a FEM is crucial for ensuring that it represents, as accurately as possible, the true behavior of the structure and its materials under analysis. If these conditions are met, the updating process is concluded and the numerical model can be defined calibrated. Something to always keep in mind is that no matter how calibrated the model is, it will never behave exactly like the existing structure but its response and analysis can surely be useful to understand and predict the behavior of the structure. Since a good level of precision is needed in order to analyze the response of a FEM, a critical aspect of the calibration process is the choice of the optimization algorithm to use. Various algorithms, based on the integration of statistical and machine learning methods, are available for this purpose: they allow the adjustment of selected material properties and boundary conditions in order to achieve the most accurate calibration possible. Therefore, selecting the appropriate algorithm can significantly impact the efficiency and accuracy of the FEMU process. In this Thesis various algorithms are used for the calibration of a FEM of an existing bridge in Italy by using experimental data collected on the structure before and after a retrofit intervention aimed at fixing the condition of one of its piers that had been subjected to scouring. Given the little presence rather than the absence of comparable contributions in the existing scientific literature about similar FEMU processes in which the scouring effect is considered and simulated in a FEM, this Thesis work will have a pioneering character, giving a new impetus to new research perspectives on this topic.