Use this URL to cite or link to this record in EThOS:
Title: Model reduction for flight dynamics using computational fluid dynamics
Author: Pagliuca, Giampaolo
ISNI:       0000 0004 7659 1820
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
The coupling of computational fluid dynamics and rigid body dynamics promises enhanced multidisciplinary simulation capability for aircraft design and certification. Industrial application of such coupled simulations is limited however by computational cost. In this context, model reduction can retain the fidelity of the underlying model while decreasing the overall computational effort. Thus, investigation of such coupled model reduction is presented in this thesis. The technique described herein relies on an expansion of the full order non-linear residual function in a truncated Taylor series and subsequent projection onto a small modal basis. Two procedures are outlined to obtain modes for the projection. First, flight dynamics eigenmodes are obtained with an operator-based identification procedure which is capable of calculating the global modes of the coupled Jacobian matrix related to flight dynamics without computing all the modes of the system. Secondly, proper orthogonal decomposition is used as a data-based method to obtain modes representing the coupled system subject to external disturbances such as gusts. Benefits and limitations of the two methods are investigated by analysing results for both initial and external disturbance simulations. Three test cases of increasing complexity are presented. First, an aerofoil, free to translate vertically and rotate, is investigated with aerodynamics based on the Euler equations. Secondly, a two-dimensional wing-tail configuration is studied for longitudinal dynamics. Aerodynamics is modelled with Reynolds-averaged Navier-Stokes equations and Spalart-Allmaras turbulence model. Thirdly, a three-dimensional industrial use case, which concerns a large civil aircraft, is investigated and longitudinal as well as lateral dynamics are both taken into account. Overall, reduced order models relying on both operator-based and data-based identifications are able to retain the accuracy of the high-fidelity tools to predict accurately flight dynamics responses and loads while reducing the computational cost by up to two orders of magnitude. If adopted, these techniques are expected to speed-up aircraft design and lowering certification costs with the final aim of reduced expense for airlines and, as a consequence, for flying passengers.
Supervisor: Timme, Sebastian Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral