Data-driven modeling, analysis, and control of flow separation over an airfoil

University of Minnesota M.S.E.E. thesis. August 2018. Major: Electrical/Computer Engineering. Advisor: Maziar Hemati. 1 computer file (PDF); vii, 60 pages.Flow separation leads to degraded performance in many engineering systems. Several flow control strategies have been developed to mitigate flow separation and improve aerodynamic performance. Most studies have focused on optimizing the control action for a given actuator configuration, whereas, actuator placement is intimately tied to achievable performance. In this thesis, a systematic approach for determining the optimal actuator location for separation control from numerical and experimental fluids data for both stable and unstable systems is presented. Pulse response data from high-fidelity numerical simulations are used to determine optimality of candidate locations. Spatial structures obtained from velocity field data from the same simulations reveal the nature of flow due to pulse actuation and corroborate findings of the optimality study. Robust closed-loop controllers are then designed on system models obtained from pulse response data, to track the separation angle