Identification of flexible structures dynamics.

The pursuit of aerodynamic efficiency and the advances in materials technology, particularly in composite material, has contributed to shifting the paradigm of wing design to high aspect ratio wings. Increasing the span, for decreasing drag, and using composite lightweight materials make the new wing very flexible and prone to nonlinear dynamic behaviour. With nonlinearities, increasing challenges arise for the identification and modelling of the wing. These challenges cannot be overlooked for flexible structures as these models are critical for the prediction of aeroelastic phenomena. Hence, it is fundamental to expand the knowledge of the behaviour of these structures through the identification and modelling of sample flexible wing models. In this work, a series of methods and approaches are proposed and employed for the identification and modelling of a flexible wing. First, a system identification technique in the frequency domain, the Loewner Framework, is applied for modal parameters extraction in mechanical systems for structural health monitoring. This new technique, with a linear reduced order model, is used to characterise the flutter behaviour of a flexible wing. The results are compared to similar techniques. A thorough experimental campaign is run on a flexible wing model to characterise its nonlinear behaviour and the underlying linear system. In particular, nonlinearities are detected, identified and quantified. Then, a meta-model technique based on Kriging, the refined Efficient Global Optimisation, is proposed for finite element model updating. First, the technique is used for damage detection in benchmark structures, then, it is employed for the validation of component-based strategies for model updating of a flexible wing.