No description available.

No description available.

Aeroelastic phenomena such as flutter can have a detrimental effect on aircraft performance and can lead to severe damage or destruction. Buffet leads to a re- duced fatigue life and therefore higher operating costs and a limited performance envelope. As such the simulation of these aeroelastic phenomena is of utmost importance. Computational aeroelasticity couples computational fluid dynamics and computational structural dynamics solvers through the use of a transforma- tion method. There have been interesting developments over the years towards more efficient methods for predicting the flutter boundaries based upon the sta- bility of the system of equations. This thesis investigates the influence of transformation methods on the flutter boundary predition and considers the simulation of shock-induced buffet of a transport wing. This involves testing a number of transformation methods for their effect on flutter boundaries for two test cases and verifying the flow solver for shock-induced buffet over an aerofoil. This will be followed by static aeroelastic calculations of an aeroelastic wing. It is shown that the transformation methods have a significant effect on the predicted flutter boundary. Multiple transformation methods should be used to build confidence in the results obtained, and extrapolation should be avoided. CFD predictions are verified for buffet calculations and the mechanism behind shock-oscillation of the BGK No. 1 aerofoil is investigated. The use of steady calculations to assess if a case may be unsteady is considered. Finally the static aeroelastic response of the ARW-2 wing is calculated and compared against ex- perimental results.

No description available.

An experimental and computational study has been carried out to enhance current understanding of three dimensional (3D) cascade aeroelastic mechanisms. 3D unsteady pressure data produced during executing this project is the first-of-its-kind, which can be directly used for validation of advanced 3D numerical methods for the prediction of aeroelastic problems in turbomachines. A new, low speed flutter test rig with a linear compressor cascade consisting of seven Controlled-Diffusion Blades has been commissioned. The unsteady aerodynamics of the oscillating cascade is investigated using the Influence Coefficient Method, by which the middle blade is mechanically driven to oscillate in a 3D bending mode. Off-board pressure transducers are utilized to allow detailed measurement of the unsteady blade surface pressures in conjunction with a Tubing Transfer Function (TTF) method to correct tubing distortion errors. The linearity of the unsteady aerodynamic response is confirmed by tests with different oscillation amplitudes, which enables unsteady results of a tuned cascade to be constructed by using the Influence Coefficient Method at various inter-blade phase angles. An examination of the techniques adopted and experimental errors indicates a good level of accuracy and repeatability to be attained in the measurement of unsteady pressure. A detailed set of steady flow is obtained from the middle three blades, which demonstrates a reasonable blade-to-blade periodicity. At a nominal steady flow condition unsteady pressure measurements were performed at six spanwise sections between 20% and 98% span for three different reduced frequencies. The 2D laminar bubble-type separation around middle chord on the suction surface is identified to have a local effect on the unsteady flow. The measured results illustrate the fully 3D unsteady flow

The exploitation of computational fluid dynamics for non linear aeroelastic simulations is mainly based on time domain simulations of the Euler and Navier-Stokes equations coupled with structural models. Current industrial practice relies heavily on linear methods which can lead to conservative design and flight envelope restrictions. The significant aeroelastic effects caused by nonlinear aerodynamics include the transonic flutter dip and limit cycle oscillations. An intensive research effort is underway to account for aerodynamic nonlinearity at a practical computational cost.To achieve this a large reduction in the numbers of degrees of freedoms is required and leads to the construction of reduced order models which provide compared with CFD simulations an accurate description of the dynamical system at much lower cost. In this thesis we consider limit cycle oscillations as local bifurcations of equilibria which are associated with degenerate behaviour of a system of linearised aeroelastic equations. This extra information can be used to formulate a method for the augmented solve of the onset point of instability - the flutter point. This method contains all the fidelity of the original aeroelastic equations at much lower cost as the stability calculation has been reduced from multiple unsteady computations to a single steady state one. Once the flutter point has been found, the centre manifold theory is used to reduce the full order system to two degrees of freedom. The thesis describes three methods for finding stability boundaries, the calculation of a reduced order models for damping and for limit cycle oscillations predictions. Results are shown for aerofoils, and the AGARD, Goland, and a supercritical transport wing. It is shown that the methods presented allow results comparable to the full order system predictions to be obtained with CPU time reductions of between one and three orders of magnitude.

La vitesse critique de flottement est un facteur essentiel à la conception aéronautique car elle caractérise le régime de vol au-delà duquel l’aéronef risque de subir un mécanisme de ruine. L’objectif de cette thèse est d’étudier l’impact des incertitudes d’origines aléatoires et épistémiques sur la limite de stabilité linéaire pour des configurations aéroélastiques idéalisées. Dans un premier temps, un problème de propagation directe d’incertitudes aléatoires relatives à des paramètres de fabrication d’une aile en forme de plaque en matériau composite stratifié a été considéré. La représentation du matériau par la méthode polaire lève la contrainte de grande dimensionnalité du problème stochastique initial et permet l’utilisation du Chaos Polynômial. Cependant, la corrélation introduite par cette paramétrisation nécessite une adaptation de la base polynômiale. Enfin, un algorithme d’apprentissage automatique a été employé pour traiter des discontinuités dans le comportement modal des instabilités aéroélastiques. Le second volet de la thèse concerne la quantification d’incertitudes de modélisation de caractère épistémique qui sont introduites au niveau de l’opérateur aérodynamique. Ces travaux, menés à partir d’un formalisme Bayésien, permettent non seulement d’établir des probabilités de modèle, mais aussi de calibrer les coefficients des modèles dans un contexte stochastique afin d’obtenir des prédictions robustes pour la vitesse critique. Enfin, une étude combinée des deux types d’incertitude permet d’améliorer le processus de calibration. / The critical flutter velocity is an essential factor in aeronautic design because it caracterises the flight envelope outside which the aircraft risks to be destroyed. The goal of this thesis is the study of the impact of uncertainties of aleatory and epistemic origin on the linear stability limit of idealised aeroelastic configurations. First, a direct propagation problem of aleatory uncertainties related to manufacturing parameters of a rectangular plate wing made of a laminated composite material was considered. The representation of the material through the polar method alleviates the constraint of the high number of dimensions of the initial stochastic problem, which allows the use of polynomial chaos. However, the correlation which is introduced by this parametrisation requires an adaption of the polynomial basis. Finally, a machine learning algorithm is employed for the treatment of discontinuities in the modal behaviour of the aeroelastic instabilities. The second part of the thesis is about the quantification of modelling uncertainties of epistemic nature which are introduced in the aerodynamic operator. This work, which is conducted based on a Bayesian formalism, allows not only to establish model probabilities, but also to calibrate the model coefficients in a stochastic context in order to obtain robust predictions for the critical velocity. Finally, a combined study of the two types of uncertainty allows to improve the calibration process.