Full-scale monitoring of the wind-induced response of vertical slender structures, with fixed and rotating masses

Nowadays, structural monitoring is gaining more and more attention in the field of wind engineering. On the wake of these developments, the thesis develops and applies a comprehensive structural monitoring procedure tailored for the validation and investigation in full-scale of the wind-induced response of vertical slender structures, with fixed and rotating masses. All the main aspects of the monitoring practice are discussed, regarding the number, location and type of the sensors, the acquisition and the transmission of the full-scale data, as well as the management of the experimental database by following an encoded scheme. In addition, the thesis highlights a number of issues typical of the monitoring activity that are not addressed in literature, providing inspiration to solve them. The defined procedure finds application in two monitoring campaigns launched by the Wind Engineering group at the University of Genoa: one slender structure with fixed masses (a light tower) and one slender structure with rotating masses (a small vertical axis wind turbine). As regards the light tower, a reference calculation model of the wind-induced response of poles and towers is selected from literature and is validated in full-scale. The input parameters needed for the application of the model are identified from experimental surveys, intersecting wind tunnel tests and dynamic identification techniques. The results highlight the goodness of the selected model and the large uncertainties associated to the input parameters. As regards the wind turbine, the full-scale data are used to investigate the contribution of the rotating parts to the dynamic behavior. In addition, the fatigue damage of the supporting tower is calculated under stationary and non-stationary excitation due to wind, turbine rotation, emergency stop and start. The results highlight the importance of the detail modeling, the fundamental role played by the non-stationary conditions and the errors committed when using conventional models of the load.