Identifiability of helicopter models incorporating higher-order dynamics

Abstract

This paper examines the identification of a linearized mathematical model of a Puma helicopter from experimental data gathered during flight tests in hover. The objective has been to study the sensitivities of the model parameters to the choice of approach available. The model of the helicopter represents the vertical response of the aircraft to collective pitch, but has been extended to incorporate higher-order rotor dynamics associated with blade flapping and induced velocity degrees of freedom. The approaches used in the identification of the model are frequency-domain based: output error and transfer function matching of frequency responses identified using time series methods. Examples of the identification of helicopter models incorporating higher-order rotor dynamics using rotor measurements from flight data are new. It is concluded that models of helicopter behavior that include higher-order dynamics can be identified successfully from flight data, but engineering judgment is the key to successful application of the methods and interpretation of the results.