Modern aircraft engine design faces the challenge of coping with conflicting aerodynamic and
structural needs. In order to reduce air transport costs and pollution, aero-engine weight is being
decreased, by reducing the number of mechanical parts and by designing thinner, lighter, and more
loaded blades. This strategy allows for reduction of fuel consumption and maintenance costs, but also
causes the blades to be more prone to aerodynamically induced vibrations, such as flutter. Indeed,
aeroelastic phenomena are nowadays acknowledged by manufacturers as a key inhibiting factor for
the provision of better, more ecological and competitive aircraft engines.
Aeroelasticity has therefore become one of the critical aspects of turbomachine blade design:
assessing whether a blade row will experience flutter or forced vibrations under given operating conditions
is an important part of the design activity.
Computational Aeroelasticity (CA) studies the development of numerical simulation tools used
to investigate the aeroelastic response of structures. Due to the complexity of aeroelastic simulations
in turbomachines, a large number of numerical approaches have been proposed, which have become
more and more accurate, as computing power has increased.
The first methods, developed at the beginning of the 1970s, were based on linear approaches where
every equation is linearized [1] [2]. Since the 1990s, new time-linearized methods have been widely
developed. According to these approaches, the flow is decomposed into a non-linear steady flow plus
a small-perturbation harmonic unsteady flow [3] [4] [5] [6] [7]. In recent years, non-linear methods
have become a viable alternative and several non-linear (uncoupled and coupled with structural
solver) approaches have been implemented [8] [9] [10] [5] [11] [12] [13]. Although computationally
expensive, these methods are able to include non-linear effects into aeroelastic analysis.
The aim of this paper is to present the application of two numerical methods (a time-linearized
method and a non-linear one) to the flutter analysis of a turbine test rig measured by the EPFL (Ecole
Polytechnique Fédérale de Lausanne, Switzerland), within the European research project FUTURE.
The numerical results are compared with each other and with the experimental data.