Data-driven adaptive dynamic programming control for gust response of camber morphing wing

This paper presents a data-driven adaptive dynamic programming (ADP) control law for gust response alleviation in active camber morphing wings. The controller is synthesized using input-output data generated from a reduced-order aeroelastic model, which is validated across a broad operational envelope, including inflow velocities V = 15-25 m/s, sinusoidal gusts with frequency f g = 1-3 Hz, and gust amplitude A g = 2 m/s. Simulation results demonstrate that the ADP controller adaptively tunes its parameters to achieve a peak alleviation efficiency of 89.6% at f g = 2.4 Hz. Moreover, the control law also performs well in V = 20 m/s, a sinusoidal gust of f g = 2 Hz, A g = 2 m/s with 10 dB Gaussian white noise and “1-cos” discrete gust, respectively. Compared to the GPC-based (Generalized Predictive Control, GPC) controller, the proposed ADP controller achieves superior alleviation performance with significantly reduced training data requirements. Furthermore, a fluid-structure-control interaction (FSCI) framework is developed based on the high-fidelity computational fluid dynamics (CFD) simulation embedded with ADP control to investigate the alleviation mechanism under sinusoidal gusts with large gust ratios (GRs). The mechanism reveals that the ADP controller modulates trailing-edge morphing to generate a counter-acting pressure zone and inhibit the formation of the leading-edge vortex (LEV). This dual action directly suppresses the sources of unsteady aerodynamic forces, thereby reducing the generalized forces and effectively attenuating the structural response.