New modeling and control design techniques for aircraft structural dynamics using smart materials

In this research, new modeling and control techniques for aircraft structural systems are presented. The sensors and actuators used in this study are smart materials. The smart structural system, which is under external aerodynamic load, is modeled utilizing an integrated finite element method. The resulting model, presented in generalized coordinates, has a mass matrix, a non-symmetric aerodynamic damping matrix and a non-symmetric stiffness matrix (due to aerodynamic stiffness). Next the system is then transformed to a real but non-orthogonal modal coordinates, and a reduced order model is developed. The control design consists of a flight mechanic control, a vibration control, and a linear matrix second order observer design. The flight mechanic control problem is to achieve a roll maneuver with a desired roll rate using a laminated flexible wing actuated by piezoelectric actuators. The equation of angular roll motion with nodal voltage actuation is derived from the finite element model. A new control design algorithm based on the ‘Reciprocal State Space’ framework is employed to achieve the desired roll rate. The deformed structure that obtains the specific roll rate is made up of new mass and stiffness matrices which are functions of the steady-state input voltage of the roll maneuver. For implementation, a laminated active twisting plate and a cantilevered beam with piezoelectric actuators and sensor are chosen to achieve the active twisting motion and the active damping effect, respectively.; Another component of this study involves vibration damping of large scale aircraft components such as landing gear. The damping is achieved using the combination of active and passive controls. A steel tube, which is structurally equivalent to a Boeing 747 landing gear break rod, is selected as a test specimen. The expected goal is to dissipate the fundamental vibration mode of the tube. In order to accomplish this task, beam type dynamic absorber and a constrained layer damping method are used for passive vibration controls. In this case, both Matlab simulations and experimental results are provided for the dynamic absorber. A simulation result for the ‘Reciprocal State Space’ based optimal control scheme is also provided. Because of hardware limitations, real time experiments can not be performed for the optimal control problem. A fuzzy logic based controller employing acceleration measurement using piezoelectric actuators (PZT) is implemented in dSPACE for the active vibration control of the system. The integrated controller with passive and active components can absorb the fundamental mode of the system well according to both the simulation and experimental results. A new matrix second order observer is employed to obtain more accurate estimate values for the second order system at hand in contrast to the traditional observer design based on the state space framework. In this portion of research, the issue of designing observers for linear matrix second order systems in the matrix second order framework before they are transformed to the state space framework is presented. Several important reasons are presented to justify the need for this approach, and the advantages of the second order observer are explained and contrasted with those of the typical first order observer. Finally, the conditions of existence of the observer gains and the design methodologies are presented.