Computational fluid dynamics-based transonic flutter suppression with control delay

Abstract

This work investigates the effects of control input time delay on closed-loop transonic computational aeroelastic analysis. Control input time delays are becoming critical as the demand for high frequency control actions is increasing. The flow in transonic conditions exhibits strong nonlinearities which require accurate physical modeling techniques, in turn resulting in large dimensional systems that are computationally costly to solve. A unified framework is demonstrated for the robust and efficient generation of reduced order models. Once generated, the reduced order model is employed for the flutter boundary search, and excellent agreement with the large order coupled model is demonstrated. The aero-servo-elastic reduced order model is then exploited to design a feedback control law, which is implemented in the fully coupled computational fluid/structural dynamics solver. As expected, the controller effectiveness is found to degrade for increasing time delay, up to a critical value where the controller fails to suppress flutter. It is shown that a controller for a time-delay system may be designed using the same aero-servo-elastic reduced order model, incurring in no extra costs or complications. The new controller is found to achieve excellent flutter suppression characteristics. The aero-servo-elastic reduced order model may also be used to identify, for a given feedback controller, the critical value of control input time delay at which the closed-loop aero-servo-elastic system loses its stability. The test cases are for a two-dimensional pitch-plunge aerofoil section and the AGARD 445.6 wing modified with a trailing-edge control surface.

More information

Identifiers

ORCID for Andrea Da Ronch: orcid.org/0000-0001-7428-6935

Catalogue record

Export record

Altmetrics