TABLE OF CONTENTS The effort required for model setup, model structure determination, and parameter identification increases, as expected, with the increasing degree of complexity in the model structures of Sec. 13.1. The simpler three-DOF fixed-wing models can involve several man-hours to a couple of days of effort, at least for the first complete experience. Coupled six-DOF models of helicopters can involve a week or more the first time. Formulating the hybrid models might require background reading on rotorcraft flight dynamics and are more complex to set up. Finally, the physics-based models of the SBMR92 and the UH-60103 each involved several months for model structure derivation from first-principles and parameter identification.
Given the potential complexity of the task, the best approach is to build up knowledge about the appropriate model structure in stages. The frequency-response method is ideally suited for this process. The first step is to study the frequency responses generated from flight data in detail, looking for clues about the model structure. Important aspects include locations of key break frequencies and modes; presence of unstable dynamics, as seen by a phase rise where the magnitude rolls off; and the importance of coupling, as detected by the coherence in the off-axis responses.
In the next step, lower-order SISO transfer-function models should be identified for the key on-axis responses to further establish the appropriate model structure (e.g., quasi-steady vs coupled rotor-body model) and to determine the location of the basic modes and dominant stability...
