Closed -loop model validation with application to an aeroservoelastic system

This thesis considers frequency domain and time domain model validation techniques for closed-loop models using norm-bounded uncertainties. The uncertainty models presented in this thesis are performed for uncertainty models structured using coprime factorizations and use knowledge of a stabilizing feedback controller to structure and formulate the uncertainty within an uncertainty model. Subsequently, the controller dependent coprime factor uncertainty model can be used to formulate model validation tests on the basis of closed-loop data.;The closed-loop model validation techniques presented in this thesis determine whether an observed set of input and output measurements could have been created by an uncertainty model. The uncertainty model is developed using knowledge of the nominal behavior and of the stabilizing controller and is addressed for both the frequency and time domains. Specifically, the frequency domain techniques involve calculating the maximum singular value of a linear expression and the time domain techniques involve solving a linear matrix inequality. The techniques developed in this thesis are applicable to numerous applications where a plant necessarily operates in a closed-loop fashion. Applications of the techniques developed in this thesis are shown for simulation studies and data from flexible structures and aeroservoelastic systems.