Aeroelasticity of very flexible aircraft at low altitudes
File(s)
Author(s)
del Carre de la Portilla, Alfonso
Type
Thesis
Abstract
This work deals with the nonlinear aeroelastic response of very flexible aircraft at low altitude. In order to study it, an efficient nonlinear aeroelasticity framework, SHARPy, has been written and validated.
With it, time domain nonlinear aeroelastic simulations of a representative High Altitude, Long Endurance (HALE) very flexible aircraft immersed in 3-D low altitude boundary layer velocity fields have been conducted. Obtained loads have shown a very strong dependency with the turbulence model used in the velocity field generation. The features of the flow affecting loads the most are the span-varying character of the velocity distribution and the coherent structures that can only be captured with physics-based turbulence modelling.
Another relevant, challenging to tackle numerically, part of the low altitude segment of typical HALE missions is their reliable deployment. In order to study it, a complex simulation of a catapult-assisted take off has been set up. This highly unsteady manoeuvre is characterised by large velocity and structural deflection ranges in a short period of time. This problem requires highly robust structural dynamics capabilities, as well as an accurate unsteady aerodynamics solver.
A parametric study of the manoeuvre is included. Thanks to the improved efficiency of the code, the catapult take off problem can be optimised without the need for further simplifications. A novel hybrid optimisation algorithm based on Bayesian Optimisation and local gradient-less methods suitable for constrained expensive cost functions with low dimensionality is presented and applied to the launch problem. When optimising for a combination of ground clearance and wing root loads, the resulting manoeuvres show a strong dependency with the maximum ramp length constraint. This introduces a link between release velocity, acceleration and ramp length that affects the cost of the manoeuvre in a complex manner.
The code used in all these simulations has been released under an Open Source license.
With it, time domain nonlinear aeroelastic simulations of a representative High Altitude, Long Endurance (HALE) very flexible aircraft immersed in 3-D low altitude boundary layer velocity fields have been conducted. Obtained loads have shown a very strong dependency with the turbulence model used in the velocity field generation. The features of the flow affecting loads the most are the span-varying character of the velocity distribution and the coherent structures that can only be captured with physics-based turbulence modelling.
Another relevant, challenging to tackle numerically, part of the low altitude segment of typical HALE missions is their reliable deployment. In order to study it, a complex simulation of a catapult-assisted take off has been set up. This highly unsteady manoeuvre is characterised by large velocity and structural deflection ranges in a short period of time. This problem requires highly robust structural dynamics capabilities, as well as an accurate unsteady aerodynamics solver.
A parametric study of the manoeuvre is included. Thanks to the improved efficiency of the code, the catapult take off problem can be optimised without the need for further simplifications. A novel hybrid optimisation algorithm based on Bayesian Optimisation and local gradient-less methods suitable for constrained expensive cost functions with low dimensionality is presented and applied to the launch problem. When optimising for a combination of ground clearance and wing root loads, the resulting manoeuvres show a strong dependency with the maximum ramp length constraint. This introduces a link between release velocity, acceleration and ramp length that affects the cost of the manoeuvre in a complex manner.
The code used in all these simulations has been released under an Open Source license.
Version
Open Access
Date Issued
2020-09
Date Awarded
2021-03
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Palacios Neito, Rafael
Sponsor
Airbus (Firm)
Publisher Department
Aeronautics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)