Eventos Anais de eventos
ENCIT 2022
19th Brazilian Congress of Thermal Sciences and Engineering
Guidelines on the Usage of System Identification Techniques in Aeroelastic Studies
Submission Author:
Marcus Vinicius Gama Muniz , SP , Brazil
Co-Authors:
Marcus Vinicius Gama Muniz, João Luiz F. Azevedo
Presenter: João Luiz F. Azevedo
doi://10.26678/ABCM.ENCIT2022.CIT22-0194
Abstract
The present work employs linear system identification techniques, SISO and MIMO algorithms, to the identification of aerodynamic loads over a NACA 0012 airfoil subject to an unsteady airflow. The aerodynamic response of the aeroelastic system to different inputs in both pitch and plunge degrees of freedom are computed by numerically solving the unsteady Euler equations. Three different types of excitations are used in the study of the SISO methodology: exponential pulse, discrete step and unit sample. The strengths and drawbacks of each of these inputs are exposed alongside a discussion on how the interpretation of the CFD solver as a discrete-time system impacts some of the CFD solver parameter settings. The more efficient MIMO methodology is employed using Walsh functions as the excitations of the airfoil degrees of freedom. The power spectral density (PSD) and cross-power spectral density (CPSD) functions, between the inputs and outputs, are computed using the Welch algorithm to obtain the aerodynamic transfer functions of the responses in lift and moment coefficients due to both pitch and plunge inputs. Additional requirements are set to the shape of Walsh functions and also to the Discrete Fourier Transform windows leading to a more precise identification process. A flutter stability analysis is also performed to compare each of those algorithms.
Keywords
flutter prediction, System Identification, Computational fluid dynamics (CFD)
