Date of Award

Spring 2024

Access Type

Dissertation - Open Access

Degree Name

Doctor of Philosophy in Aerospace Engineering

Department

Aerospace Engineering

Committee Chair

Richard Prazenica

First Committee Member

K. Merve Dogan

Second Committee Member

Marc Compere

Third Committee Member

Hever Moncayo

College Dean

James W. Gregory

Abstract

The future of aircraft design strives for lighter weight, more aerodynamically efficient structures. These improvements may come with the drawback of increased structural flexibility and elevated aeroelastic effects, often resulting in a lower flutter speed. This motivates the implementation of advanced control methods to control aeroelastic systems over a range of flight conditions, suppress and delay the onset of flutter, and compensate for disturbances, actuator dynamics, and unmodeled nonlinear dynamics.

This dissertation first develops a novel method for constructing time-domain simulation models of two and three-dimensional aeroelastic systems, resulting in models that are suitable for the implementation of state-space control algorithms. Then, model reference adaptive control strategies are implemented on the aeroelastic systems with the objective of controlling these systems over a range of pre- and post-flutter flight conditions subject to disturbances, nonlinear dynamics, and actuator dynamics. To model the aeroelastic systems, the structural dynamics equations of motion are first developed for a two-dimensional pitch-plunge-flap airfoil section with nonlinear torsional stiffness, first- and second-order actuator dynamics, and actuator freeplay. This system utilizes quasi-steady aerodynamic forcing to develop an open-loop flutter model upon which direct model reference adaptive control is applied. Next, the structural dynamics equations of motion for a three-dimensional wing are developed using energy methods and modal solutions to the forced vibration problem. For the three-dimensional system, an unsteady vortex-lattice solver is implemented to calculate real-time unsteady aerodynamic forces and aerodynamic control force inputs. This is coupled with the flexible equations of motion to form the full aeroelastic equations of motion, which are then validated against a Nastran model. Finally, the elastic wing is coupled to a longitudinal -rigid body aircraft dynamics model to understand the relationship between rigid-body dynamics and flexible motion in the presence of a controller.

Direct model reference adaptive control (MRAC) strategies are then implemented on the two- and three-dimensional aeroelastic systems, and the performance is compared to that obtained using a standard linear quadratic regulator (LQR). The simulation studies demonstrate that MRAC provides more effective aeroelastic control and flutter or limit-cycle oscillation suppression over a wider range of flight conditions compared to the standard LQR controller. In addition, the MRAC is shown to provide robustness to nonlinear stiffness, first-order and second-order actuator dynamics, and actuator freeplay. The effect of varying actuator bandwidth on the MRAC performance is studied for several cases, with the objective of determining minimum bandwidth requirements.

Share

COinS