Robust
flutter margins and uncertainty operators
Contact:
Context:
The SISTHEM team is working on identification and damage detection in
civil structures as well as monitoring aero-elastic phenomena in
aeronautical structures under natural or controlled excitation. This
work has been conducted under both national and European Eureka
projects (Sinopsys, FliTE, FliTE2).
Subject:
The SISTHEM team is currently
investigating how to detect and prevent flutter to happen in mechanical
structures. Applications include aircrafts and bridges. Typically, such
structures are modelled using a linear discretization of the (M,C,K)
continuous equation. In such modelling it has been shown at SISTHEM
that evolution of modes can be tracked to prevent flutter scenario to
happen (for example, crossing of frequencies or fluctuation of
dampings).
Within NASA Flight Research Center (Lind et al.), extended non linear
models have been proposed to include aeroelasticity models into the
linear system driving the dynamics of the structure. From a practical
point of view, for aircraft manufacturers, defining flutter relates to
determine the flight envelope devoid of unstabilities for any new
aircraft or new configurations of current aircrafts. Critical flutter
conditions are defined by the region closest to the flight envelope
where unstability starts. The end problem is to define the maximum
deviation allowed with respect to a nominal condition (usually defined
on ground).
An approach for computing worst case flutter margin has been
formulated in a robust stability framework within NASA Flight Research
Center. Uncertainty operators are used to describe errors and flutter
variations. It consists in small perturbation of the system operator.
Then, the mu-margins are robust margins that indicate worst case
stability estimates with respect to the defined uncertainty. This
theory relates to the small gain theorem to define upper bounds on the
norm of the operator and the methods end up with finding the smallest
maximum eigen value over all set of perturbations Delta statisfying:
det (I - P Delta) = 0.
The objective of this work is to pave the road for a flutter monitoring
method combining the flutter margin approach of NASA with our local
approach to the design of detection algorithms.
INRIA strategic priority:
This subject takes place within the INRIA scientific and technological
challenge no 4: Coupling models and data to simulate and control
complex systems.
Desired profile of the candidate:
This postdoc position involves understanding the relation between
adding aeroelasticity terms in the linear system, the mu-method and the
classical realization theory.
