21549 research outputs found
Sort by
Aeroelastic scaling of flying demonstrator: flutter matching
Get PDFThe traditional approach for the design of aeroelastically scaled models assumes that either there exists flow similarity between the full-size aircraft and the model, or that flow non-similarities have a negligible effect. However, when trying to reproduce the behavior of an airliner that flies at transonic conditions using a scaled model that flies at nearly-incompressible flow conditions, this assumption is no longer valid and both flutter speed and static aerodynamic loading are subject to large discrepancies. To address this issue, we present an optimization-based approach for wing planform design that matches the scaled flutter speeds and modes of the reference aircraft when the Mach number cannot be matched. This is achieved by minimizing the squared error between the full-size and scaled aerodynamic models. This method is validated using the Common Research Model wing at the reference aircraft Mach number. The error in flutter speed is computed using the same wing at model conditions, which are in the nearly-incompressible regime. Starting from the baseline wing, its planform is optimized to match the reference response despite different conditions, achieving a reduction of the error in the predicted flutter speed from 7.79% to 2.13%
Integrated optimal design of a hybrid-electric aircraft powertrain
Get PDFIn 2019, transportation was the fastest growing sector, contributing to environmental degradation. Finding sustainable solutions that pollute less is a key element in solving this problem, particularly for the aviation sector, which accounts for around 2% of global CO2 emissions. With the advent of Covid-19, air traffic seems to have come to a fairly permanent halt, but this pandemic reinforces the need to move towards a "cleaner sky" and respect for the environment, which is the objective
of the Clean Sky2 program (H2020 EU), the context in which the HASTECS project and our thesis are set. The main objective of HASTECS (Hybrid Aircraft Academic reSearch on Thermal and Electrical Components and Systems) is to couple thermal and electrical studies within the hybrid electric propulsion chain of a regional aircraft, by integrating the environmental constraints (in particular partial discharges) specific to the aviation sector. The aim is to identify the most promising technologies and breakthroughs and to develop the tools that will significantly increase the compactness and efficiency of the electrical processes within the hybrid propulsion chain. In our case, only series hybrid electric architecture was studied in this project, as it leads to a high dimensioning power maximizing the technological constraints on the chain. The technological targets set in HASTECS, considered under two horizons (2025 then 2035), are the following:
Electric motor + cooling system
Specific power : 5kW/kg - 10kW/kg
Cruise efficiency : 96% - 98.5%
Maximal design point efficiency : 94.5% - 97%
Power electronics + cooling system
Specific power : 15kW/kg - 25kW/kg
Cruise efficiency : 98% - 99.5%
Maximal design point efficiency : 96.5% - 99%
In the framework of this project, our thesis aims at the design by optimization of the complete propulsion chain integrating in particular the models resulting from the technological developments of the major components (power electronics, wiring and distribution architecture, actuation) while considering, from a simplified energy management, the hybridization of a main (thermal) and auxiliary (electrical) source. A first objective of our thesis concerned the development of an environment model. Once these conditions are set, the system integration consists in building a suite of scale models whose granularity allows the global (systemic) evaluation of the energy yields and masses of each component up to the complete propulsion chain. The propulsion system is designed via an iterative process estimating, according to the
design choices, the mas
Environmental and economic assessments of selection for feed efficiency in pigs
Get PDFThe aim of this thesis was to develop an approach to evaluate breeding strategies for feed efficiency in pigs to identify mitigating solutions for environment and economy. A life cycle environmental assessment model (LCA) and a bio-economic model were developed at the individual level. It enabled to assess varieties of feed efficiency scenarios on an individual basis. The models were applied to assess the impacts of selection for feed efficiency alone or combined with diet optimisation as a strategy to achieve overall farm feed efficiency. Data from two pig lines selected for feed efficiency based on residual feed intake (RFI), showing genetically contrasted feed efficiency, were used. A multiple objective diet optimisation was developed using genetic line nutritional requirements as constraints, and minimising a score combining environmental impacts and/or cost. The consistency of the models was obtained from considering net energy as the core linkage between line requirements, ingredients dietary composition, price, and environmental impacts. The individual performance traits in response to new optimised diets for each line were simulated with the InraPorc® pig growth simulation software. For a least cost, a least environmental impacts, and a joint economic-environment diets, these performances were used as inputs to the LCA and bio-economic models. The individual economic and environment assessments showed that selection for improved feed efficiency in pigs increases the profitability by 23.4% and reduces the environmental footprint by 7%. Selection for feed efficiency combined with diet optimisation even enhanced these economic and environmental improvements through restoring part of the advantages of selection that did not emerge due to feeding the lines the same diet. Thus, for increased pig sustainability, a selection for feed efficiency should be combined to diet optimisation including environmental constraints. However, feeding less efficient pigs an optimised diet strongly reduced the genetic differences and alleviated most of the economic and environmental burdens. Finally, the assessment at the individual level gave access to the covariances between the performance traits and the environmental impacts and profit. High correlations of feed conversion ratio with environmental impacts and profit in both lines confirmed the importance of feed efficiency as a lever for the sustainability of pig production, and the moderate correlations with RFI pointed this trait as a potential lever to improve environmental impacts with limited correlated effects on other production traits. These results and tools will contribute to move from breeding goals essentially based on economic objectives to more holistic breeding goals, to contribute to increased sustainability in pig production
Loss assessment of the NASA SDT configuration using URANS with phase-lagged assumption
Get PDFThis paper presents the study of the Source Diagnostic Test
fan rig of the NASA Glenn (NASA SDT). Numerical simulations
are performed for the three different Outlet Guide
Vane (OGV) geometries (baseline, low-count and low-noise)
and three rotational speeds corresponding to approach, cutback
and sideline operating conditions respectively. Unsteady
Reynolds Averaged Navier-Stokes (URANS) approach
is used. The in- and out-duct flow including the nacelle are
considered in the numerical simulations and results are compared
against available measurements. Due to the blade
count of the fan and OGVs (22 fan blades and either 54
or 26 blades for the OGVs), the simulation can only be reduced
to half the full annulus simulation domain using periodic
boundary conditions that still represents a significant
cost. To alleviate this issue, a URANS with phase-lagged assumption
is used. This method allows to perform unsteady
simulations on multistage turbomachinery configurations including
multiple frequency flows with a reduced computational
domain composed of one single blade passage for each
row. The large data storage required by the phase-lagged
approach is handled by a compression method based on a
Proper Orthogonal Decomposition (POD) replacing the traditional
Fourier Series Decomposition (FSD). This compression
method improves the signal spectral content especially
at high frequency. Based on the numerical simulations, the
flow field is described and used to assess the losses generated
in the turbofan architecture based on an entropy approach.
The results show different flow topologies for the fan
depending on the rotational speed with a leading edge shock
at high rotational speed. The fan boundary layer contributes
strongly to losses with the majority of the losses being generated
close to the leading edge
Potentiel de l'assimilation des données radar à double polarisation pour la prévision numérique du temps à l'échelle convective
Get PDFLes conditions initiales du modèle de prévision numérique du temps à échelle convective AROME sont fournies par un système d’assimilation de données 3D-Var dans lequel les observations issues des radars au sol sont prépondérantes lors de situations précipitantes (vitesse radiale et réflectivité horizontale). La réflectivité horizontale n’est pas exploitée de manière optimale car elle est transformée au préalable en profil d’humidité relative avant son utilisation dans le 3D-Var. Les contenus en hydrométéores dont elle dépend directement ne sont donc pas modifiés par l’assimilation. Le réseau radar national permet également d’accéder à des variables polarimétriques grâce à l’émission d’ondes polarisées horizontalement et verticalement. Ces observables fournissent des informations complémentaires sur la microphysique (phase, forme et orientation des hydrométéores). Cette thèse propose d’étudier le potentiel de ces variables pour l’initialisation des hydrométéores dans le modèle AROME grâce à une méthode d’assimilation variationnelle directe. Pour cela, un opérateur d’observation radar polarimétrique s’appuyant sur la méthode des T-matrices pour décrire la diffusion a été utilisé afin de simuler les réflectivités horizontale et différentielle, la phase différentielle spécifique et le coefficient de corrélation à partir des contenus en hydrométéores du modèle AROME. Sa capacité à décrire précisément ces variables a été étudiée sur un ensemble varié de situations météorologiques et les statistiques d’innovations ont été calculées. Par ailleurs, les sensibilités des variables polarimétriques aux contenus en hydrométéores ont été mises en évidence en déterminant les éléments de la matrice Jacobienne en différences finies. Les réflectivités horizontale et différentielle sont apparues comme étant les variables les plus informatives. Du fait des limitations de l’opérateur et de la microphysique du modèle, le coefficient de corrélation n’a pas pu être exploité. Après avoir linéarisé l’opérateur d’observation et estimé les matrices de covariances d’erreurs pour l’ébauche et les observations, le contenu en information des variables polarimétrique a été quantifié. Enfin, des expériences d’assimilation 1D-EnVar ont été menées pour des situations météorologiques contrastées. Elles ont montré la capacité de l’algorithme à extraire une information pertinente sur les hydrométéores précipitants (principalement en phase liquide) à partir de la réflectivité horizontale, et dans une moindre mesure de la réflectivité différentielle. Les autres variables polarimétriques peinent à proposer une valeur ajoutée, notamment dans la phase glacée
Control of rotation-floating space robots with flexible appendages for on-orbit servicing
Get PDFOn-orbit operations are facing a growing need for autonomous robotic systems. Debris
removal, on-orbit servicing and in-space deployment/assembly are examples of applications
considering the use of robot manipulators. This paper addresses design and control
problems related to autonomous space manipulator systems when using kinetic moment
exchange devices in presence of flexible appendages. The paper introduces a method to
develop a common control of the spacecraft base and manipulator. An extended stat
Robust estimation of high-order phase dynamics using Variational Bayes inference
Get PDFCycle slips strongly impact the performance of phase tracking algorithm leading, in the worst case, to a permanent loss of lock of the signal. In this paper, we propose a new nonlinear phase estimator to obtain more robust tracks. The latter stems from a Variational Bayes (VB) approximation used within the optimal Bayesian filtering formulation in case of high-order phase dynamics. A comparison with a more conventional technique, namely a Kalman filter based PLL (Phase Lock Loop), is performed in terms of mean square error of the phase estimate and mean time to first slip. Results show that the proposed
method outperforms the conventional linear filter with respect to both metrics, especially at low signal-to-noise ratio
Assessment of the Industrial Potentials for an Innovative High-tech Separation Method
Get PDFCurrently, we are facing a general technological gap for the separation of mixtures containing colloidal particles (metallic nano-particles, macro-molecules, etc.). This project aims to extend the principle of Hollow Fiber Flow Field-Flow Fractionation (HF5) toward continuous size-based separation of colloidal suspensions. HF5 is an analytical technique used for particle fractionation (according to their size) in view of their characterization. The principle of particle separation is based on consecutive focusing of particles of different sizes on different ow streamlines with the aid of a permeate ow applied across the hollow-ber wall and their differential elution by the main ow streamlines at the ber outlet. In this Ph.D., we developed an automated sequential prototype based on the HF5 principles. The accumulation and elution conditions were explored, leading to the prescription of a separation methodology, which applies for stable negatively charged Brownian particles (in order to limit particle aggregation and adsorption on the membrane). A CFD model was developed to guide the initial prototype design and the choice of the operating conditions. The numerical model was extended to include an original description of colloidal dynamics near phase transition and its impact on near-membrane particle accumulation and relaxation processes. We evidenced the existence of a permeate ow-rate (during elution) at which the Brownian diffusion prevails over the drag force, and that allows controlling particle differential elution: this reveals to be a crucial condition for successful separation. The developed prototype allows the processing of higher sample amount than classical HF5, by using the total membrane surface during particle focusing. We performed the separation of latex polystyrene nanoparticles with a size ratio of 5, using 4.74 μg of particles per cm 2 of membrane without compromising the peak resolution (similar to AF4 conguration and superior to HF5). Moreover, the prototype allows running several consecutive separation cycles (up to ve cycles were tested), without increasing membrane fouling. To allow prototype scale-up, further optimization of particle accumulation along the membrane and permeate-ow control during elution are needed. Finally, a system for continuous operation was developed. It is not included in this manuscript for confidentiality reasons
Modélisations numériques et expérimentales des écoulements diphasiques en milieux poreux très perméables : passage de l'échelle du pore à l'échelle de Darcy dans le contexte des sites et sols pollués
Get PDFLe changement d’échelle des écoulements diphasiques immiscibles dans des milieux poreux très perméables a été encore très peu étudié. Cependant, la modélisation numérique correcte de ce type d’écoulement à l’échelle macroscopique est cruciale afin d’optimiser les opérations de dépollution des eaux souterraines contaminées par des DNAPL (Dense Non-Aqueous Phase Liquid). En outre, le problème de milieux très perméables implique que certaines hypothèses valables pour le modèle macroscopique couramment utilisées doivent être réévaluées. Parmi les plus importantes, l’hypothèse de couplage mécanique négligeable entre les fluides est particulièrement discutée dans ce travail. Pour ce faire, nous avons étudié, à partir de simulations numériques, l’impact de la perméabilité d’un micro modèle (cellule de Hele-Shaw) sur les forces de trainées aux interfaces, et en particulier à l’interface entre les deux fluides. Les résultats confirment l’importance de prendre en compte le couplage entre les fluides pour modéliser la perte de pression de l’écoulement à travers la cellule. Ceci a motivé le choix de les calculer sous forme de paramètre effectifs, nécessaires pour fermer les équations macroscopiques. Une méthode basée sur une légère modification de la force volumique de chaque fluide alternativement est proposée et testée sur des écoulements diphasiques caractéristiques de milieux très perméables. La méthode proposée est confrontée aux résultats obtenus par la résolution des problèmes de fermeture, dont une modification est proposée sur la base de cette comparaison. La modification des problèmes de fermeture a permis de mieux reproduire l’écoulement et d’obtenir des résultats identiques à la méthode proposée. Enfin, des déplacements stables d’un DNAPL modèle dans un milieu homogène saturé et très perméable ont été reproduits à l’échelle décimétrique dans un dispositif de laboratoire. Les résultats expérimentaux permettent de discuter les stratégies de modélisation numérique dans ce cas simple (forme des équations, prise en compte des termes couplés, etc), puis de mener une étude paramétrique à partir du modèle numérique validé, afin d’étudier l’impact des effets gravitaires et inertiels sur un déplacement stable simplifié représentant un cas d’intrusion d’un DNAPL dans un aquifère. Les résultats indiquent que les effets inertiels ne sont probablement pas à négliger dans certaines situations particulières rencontrées en dépollution
On the distributions of some statistics related to adaptive filters trained with t-distributed samples
Get PDFIn this paper we analyse the behaviour of adaptive filters or detectors when they are trained with -distributed samples rather than Gaussian distributed samples. More precisely we investigate the impact on the distribution of some relevant statistics including the signal to noise ratio loss and the Gaussian generalized likelihood ratio test. Some properties of partitioned complex distributed matrices are derived which enable to obtain statistical representations in terms of independent chi-square distributed random variables. These representations are compared with their Gaussian counterparts and numerical simulations illustrate and quantify the induced degradation