[en] The investigation of nonlinear aeroelastic phenomena is becoming increasingly important to the
aerospace community. The existence of structural and aerodynamic nonlinearities
in aircraft has always been acknowledged but, it is only mainly with the advent of modern
digital computers that their investigation has become possible. Additionally, aircraft
control systems are becoming increasingly nonlinear with the introduction of Active
Control Technology. The effects of these nonlinearities on the dynamic response of aircraft
have created the need for further research into the modelling, identification and
prediction nonlinear aeroelastic systems.
This thesis deals with four aspects of nonlinear aeroelasticity. Firstly, the effect of the
common industrial approach to nonlinearity, i.e. that of linearisation, is investigated.
Six flutter prediction methods for linear aircraft are tested and compared on linear and nonlinear
mathematical models of aeroelastic systems. The performances of the methods on linear
systems are evaluated and compared. Subsequently, their predictions predictions when
applied to nonlinear systems are assessed.
Secondly, the dynamic response of nonlinear aircraft is investigated by means of the Harmonic
Balance method and the
direct integration of the nonlinear mathematical model. Emphasis is given to
the explanation of the appearance of Limit Cycle Oscillations as Hopf bifurcations and on the
control and suppression of these oscillations by means of a feedback control system.
The chaotic vibration of nonlinear aeroelastic systems is also investigated by means of
Poincare diagrams and Lyapunoff exponents.
Thirdly, the identification of nonlinear aeroelastic systems is considered.
Identification of aeroelastic systems is important since, especially in the case
of structural nonlinearities, it is often not known whether an aircraft is linear
or not and what nonlinearities it may contain until it is tested, either on the
ground (Ground Vibration Testing) or in the air (Flight Flutter Testing).
An existing nonlinear system identification method is compared to an approach developed
during the course of the present project. The two techniques are applied to
a nonlinear mathematical aeroelastic system and to a set of nonlinear input-output
data obtained from an experimental system. Both methods were found to be able
to deal with both systems with varying degrees of success.
Finally, the gust response of nonlinear aircraft is investigated with particular
emphasis on the calculation of gust design loads. Turbulent gust clearance is a very
important part of any airworthiness testing procedure. Until recently, the linear
assumption was considered adequate by the requirements however, there is a current
shift towards setting new requirements that take into account nonlinear phenomena.
Eight gust load prediction methods for nonlinear aircraft(both stochastic and deterministic) are
applied to a simple and a more complex nonlinear mathematical aircraft model. The
performance of the methods is assessed with respect to both accuracy and computational
efficiency.
Disciplines :
Aerospace & aeronautics engineering
Author, co-author :
Dimitriadis, Grigorios ; Université de Liège - ULiège > Département d'aérospatiale et mécanique > Intéractions fluide structure et aérodynamique expérimentale