Aeroelastic parameter identification in wind tunnel testing via the extended eigensystem realization algorithm

Abstract

Aeroelastic instability may occur in aircraft during flight, therefore their prediction represents an important issue within aerospace engineering. Experimental aeroelasticity is still an important field in providing the tools to validate and understand instability phenomena analysis. As many industrial practices require fast evaluations of critical conditions, e.g. flight flutter testing, there exists a natural demand for on-line aeroelastic identification. A number of different methods have been proposed to characterize systems, but recently those showing most success for on-line identification have been based on subspace approaches. The eigensystem realization algorithm (ERA) represents one of the first subspace methods for identification, with the advantage of dealing with multi-input, multi-output data. However, its need for repeated application of the singular value decomposition and a dependence on impulse response functions implies limitations to on-line identification. Generalization studies of the ERA method have led to recursive forms of that algorithm. A recursive form closely related to ERA has been developed in terms of a modified batch estimation approach, and it is denoted as the extended eigensystem realization algorithm (EERA). This work presents results on the application of extended EERA method viewing on-line aeroelastic parameters identification of an experimental apparatus in the wind tunnel. Designed to reproduce the conditions for typical section aeroelastic behavior, an apparatus has been used to show the EERA capabilities in identifying on-line aeroelastic frequency and damping parameters. Results have shown that the approach is robust and adequate for aeroelastic characterization during experimental activities.