Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
755669 | Chinese Journal of Aeronautics | 2006 | 8 Pages |
The aim of this paper is to present a new method for flight flutter modal parameter identification in noisy environment. This method employs a time frequency (TF) filter to reduce the noise before identification, which depends on the localization property of sweep excitation in TF domain. Then, a generalized total least square (GTLS) identification algorithm based on stochastic framework is applied to the enhanced data. System identification with noisy data is transformed into a generalized total least square problem, and the solution is carried out by the generalized singular value decomposition (GSVD) to avoid the intensive nonlinear optimization computation. A nearly maximum likelihood property can be achieved by ‘optimally’ weighted generalized total least square. Finally, the efficiency of the method is illustrated by means of flight test data.