Browsing by Author "Kornienko, Andrei"

Now showing 1 - 1 of 1

Open Access
System identification approach for determining flight dynamical characteristics of an airship from flight data
(2006) Kornienko, Andrei; Well, Klaus H. (Prof., Ph. D.)
The knowledge of dynamical characteristics of a flight vehicle is necessary for the control system design and realization of high fidelity flight simulators. The development of a flight mechanical model and determination of its basic components, as for example mass properties and the major aerodynamic terms, addresses a complex process involving various analytical, numerical and experimental techniques. The objective of this dissertation is a determination of the basic dynamical characteristics of a research airship from the flight data. In order to achieve this objective a system identification approach is used. As the modern identification methodology requires a coordinated treatment of tasks from multidisciplinary areas, such as modelling, parameter estimation methods, instrumentation and flight maneuver definition, within this research framework these topics are studied and extended with regard to the airship example. The dynamic modelling consists of two parts. The first part is devoted to derivation of the nonlinear flight mechanical model of the airship. It incorporates the classical laws of newtonian mechanics used for derivation of equations of motion. In addition, the airship specific properties, like buoyancy force and the unsteady aerodynamic effects, are included in the equations of motion. In the second part, the linearized longitudinal and lateral-directional models are introduced. Investigations of stability and controllability under different flight and configuration conditions are performed. The important parameters of the linearized models are then utilized in the parameter estimation. For determining the model parameters statistical estimation methods are considered. Their choice is motivated by airship operational properties and limitations posed by the used instrumentation system. A particular emphasis is made on the algorithm that incorporates the Kalman filter to obtain parameter estimates in the presence of process and measurement noise. The process noise is used to model the uncommanded forcing function, mainly caused by atmospheric disturbances. The measurement noise is caused by the distortions of measured channels. A valuable effort in this research framework is devoted to the design of experiments. This includes a realization of the onboard data acquisition system and definition of flight maneuvers. The conventional methods used in the aircraft field are accommodated for the experiment design. Application of the used methodology to the flight data has shown a very good trajectory matching between the model and the real airship responses. It verifies the adequacy of the selected identification methodology for determining the dynamical characteristics of the airship from the flight data.