Identification of rotor-bearing systems in the frequency domain part II: Estimation of modal parameters

This paper discusses the parametric identification of multiple-input-multip le-output (MIMO) rotor-bearing systems in the frequency domain on the basis of the maximum likelihood estimator (MLE). An identification procedure, co nsidering noise on both inputs and outputs (errors-in variables model), is developed for real as well as for complex modal testing. Quantitative measu res for the uncertainties (covariances) of the estimated parameters were to be obtained. It is shown that the MLE solution of real and complex modal a nalysis is fully equivalent. It follows that, in case of complex modal test ing, reduction of the system is not allowed in general when noise informati on is taken into account. A comparison between random and broadband periodi c (multisine) excitations is made. Measurements on a rotor test rig demonst rated that multisines yield more accurate estimates of the modal parameters (as compared to random excitation). (C) 2001 Academic Press.