M.Sc Thesis

M.Sc StudentDavid Azoulay
SubjectCharacterization of Methods for Dynamic Response Analysis
of Aeroelastic Systems to Gust Excitation
DepartmentDepartment of Aerospace Engineering
Supervisor Professor Emeritus Karpel Mordechay
Full Thesis textFull thesis text - English Version


The design of modern flight vehicles requires the evaluation of structural dynamic loads in response to discrete (deterministic) and continuous (stochastic) gust excitations. The process of determining the dynamic response and dynamic loads of an aeroelastic system subject to gust excitations is composed of several phases. The first phase consists of modeling of the flying vehicle. The basic elements to be modeled in the analysis are the elastic, inertial, and aerodynamic characteristics of the complete aircraft. The modal approach is usually used, where the motion of the structure is described as a combination of rigid-body and elastic modes of vibration. The aerodynamic model, used to compute the aerodynamic forces, is based on the selected unsteady aerodynamics theory: Strip Theory or Panel method, two- or three-dimensional, for incompressible or compressible flow. In the second phase, frequency-domain or time-domain approaches can be used to solve the equations of motion of the system. In two-dimensional theories like the Strip Theory, determination of the unsteady aerodynamic forces is accomplished through the use of oscillatory lift functions in the frequency domain or indicial lift functions in the time domain. Three-dimensional panel methods preferentially use the frequency-domain approach through generation of aerodynamic influence coefficients matrices. The third phase of the analysis consists of determining critical loads on structural components in order to verify that the structure can sustain the designed gust intensities. Structural loads are extracted from the modal equations of motion governing the gust response, according to two possible methods: the Mode Displacement method or the Summation-Of-Forces method.

The main objective of this study is to characterize and compare the different existing techniques for analyzing the response of aeroelastic systems to gust excitations. For supporting this purpose, three models are used in this work, a rectangular wing, a swept wing, and a complete aircraft model, on which comparative analyses are performed. The performances of the methods are measured in terms of accuracy, stability, efficiency and convergence. Computational effort is also considered. An improved version of the Strip Theory is introduced. In this new method, some steady aerodynamics data is imported from a high-fidelity aerodynamic code, and aerodynamic coupling between strips is restored. It is shown that this method, applied to flutter and gust response analyses yields results comparable to more complex approaches, and facilitates direct time-domain response analyses via convolution.