Flutter simulation and prediction with CFD-based reduced-order model

Abstract

The paper presents a numerical aeroelastic simulation and prediction environment for applications to complex configurations. The simulation environment embeds a high-order computational fluid dynamic (CFD) code and a reduced-order modeling (ROM) technique. To set up the simulation system, the CFD is applied first to the full-order model to obtain the non-aerodynamically loaded aeroelastic behavior in accordance with the Filter-Impulse Method (FIM). A system identification and reduction scheme is then used to construct the ROMs from the FIM responses. The adoption of ROM permits a black box evaluation of the aeroelastic system in such a way that after a single full order aeroelastic simulation for one flow condition (at a given Mach number) the system state for other flow conditions (at that Mach number) might be estimated. The ROMs are in state space form and so can easily be coupled to a structural model for aeroelastic calculations. Pilot studies?,? have shown that the ROMs are capable of predicting accurately the flutter boundary in a tiny fraction of the time required by full-order CFD code. The crucial first step, and also one major difficulty, in reduced-order modeling is the generation of ROMs that retain the accuracy of the full nonlinear CFD methods. Following previous work,? we present in the paper improved method of model reduction for complex configurations. Applications to fighter-type aircraft, the NORTHROP F-5E TIGER II, will be presented. The reduced-order model is compared to the full-order model, demonstrating the practical suitability of reduced-order model for complex configurations. Computations of flutter boundary using ROMs will also be presented with the AGARD 445.6 test cases.