Conference Paper Parametric model reduction for unsteady flow about airfoil based on compressible CFD solver

Xiang, Zhao  ,  Kim, Taehyoun

In this work, a new Parametric Reduced-Order Model (PROM) with use of the Modally Equivalent Perturbed System (MEPS) and the Proper Orthogonal Decomposition (POD) is applied to unsteady flow about Airfoil with parameter variations in Mach. By applying the method, the original parameter-varied fluid system can be reduced from a very large dimension to a small dimension (e.g. for 2D case from 10,000 states to 100 states). While traditional Reduced-Order Model (ROM) cannot account for parameter variations, the present PROM is efficient to treat parameter changes adaptively. This means the POD modes obtained are valid for a wide range of parameter values. Towards this end, we have developed a frequency-domain CFD solver in which the fluid model is statically nonlinear, dynamically linearized (SNLDL) allowing unsteady simulations with small oscillations in subsonic, transonic, and supersonic flows in state-space format. With the state-space model, one can predict system responses (e.g. unsteady aerodynamic responses due to small amplitude oscillations of the lifting surface), conduct eigen-analysis, and build the PROMs. The ability of the new method to save substantial amount of computation for the varying parameter will be demonstrated using a 2D airfoil configuration.
Meeting Information: First International Symposium on Flutter and its Application (May 15-17, 2016. Mielparque-tokyo), Minato-ku, Tokyo, Japan
Physical characteristics: Original contains color illustrations
形態: カラー図版あり

Number of accesses :  

Other information