IDENTIFICATION OF AERODYNAMIC MODELS OF BRIDGE DECKS FROM NUMERICAL SIMULATION
A new efficient method based on CFD discrete-time aerodynamic models is presented,which employs 3211 multi-step inputs of velocity to force a CFD model to oscillate over a wide frequency range of interest. With the displacements integrated from 3211 multi-step inputs of velocity and aerodynamic forces from CFD computing, discrete-time aerodynamic models can be developed through system identification technique. The aerodynamic models are then adopted to simulate aerodynamic forces to harmonic displacements. Finally, the displacement inputs and aerodynamic forces are used to dentify flutter derivatives corresponding to reduced wind speeds. The present method needs only one run of CFD modeling for heaving or pitching motion resulting in significant saving on CPU time to extract flutter derivatives. The identified flutter derivatives of a thin plate are compared to those from wind tunnel tests, all of these flutter derivatives share the same trends.
Flutter CFD System identification 3211 multi-step Model simulation
Zhi-Wen Zhu You-Xiang Li Zhao-Xiang Wang Zhen-Qing Liu
College of Civil Engineering, Hunan University, Changsha 410082, P.R. China
国际会议
第十届国际结构工程青年学者研讨会(The Tenth International Symposium on Structural Engineering for Young Experts)
长沙
英文
1673-1678
2008-10-19（万方平台首次上网日期，不代表论文的发表时间）