Flutter modal identification with frequency-based least squares estimator using vector orthogonal polynomials

Frequency-based least squares (LS) estimators are widely used tools for flutter modal identification of aircraft. These methods are known to suffer from a poor numerical condition when the frequency spans or model order becomes large. A new approach using vector orthogonal polynomial basis is employed in this paper to solve the problem, The key idea of this method is to represent the frequency response function (FRF) matrix in a left matrix fraction description (LMFD) model, and expand the numerator and denominator polynomial matrices in a vector orthogonal basis. As a result, a perfect numerical condition (numerical condition=1) can be obtained for linear least squares estimator. Finally, the efficiency of the approach is illustrated by means of real flight test data.