21549 research outputs found
Sort by
On bio-inspired induced drag reduction techniques
Get PDFThe current study was aimed at identifying wing planform
geometries which favours low induced drag. The
geometries were defined by mathematical functions and
following a subsequent parametric study, with a focus
on sweep angle and taper ratio, inspired by birds wings
and aquatic animals flukes or caudal fins planforms. The
parametric studies were carried out using a vortex lattice
method (VLM) which was validated with experimental
data from literature. The main finding is a confirmation
of the fact that the maximum Oswald efficiency factor, e
is not exclusive to an elliptical wing planform. Despite
its apparent simplicity, in fact a swept-back-tapered wing
with a linear chord distribution can reach e = 1, the limit
or minimum induced drag established from elliptical lift
distribution. Moreover, the crescent wing geometry, such
as the one displayed by the swift, can even surpass this
limit
Massive GPU Parallelisation for Cislunar Debris Mitigation Analyses
No full textNowadays, the issue of space debris is becoming central to any operation in space. Future lunar exploration missions won’t be exempted from this issue, with the aggravating factor of operating in a gravitational
multi-body chaotic environment. Debris-related analyses often involve propagating numerous trajectories
independent from each other. These allow, for example, estimating probabilities of further interaction of
space debris with the Earth and other protected regions of space; additionally, initial uncertainties on the
object’s state or the magnitude and direction of manoeuvres can be introduced in the analyses. This can
translate into massive grid searches or Monte-Carlo runs, to map the totality of the solution space. Op-
erators and mission designers can then use this cartography to support the design of normal mission and
disposal phases. A similar approach can also help identify the areas most susceptible to chaotic behaviours,
or the risk posed by a spacecraft fragmentation in such locations. All these efforts are often paired with
the necessity to propagate for very long times with high precision, which increases the load on the com-
putational infrastructure. This paper discusses the application of Graphics Processing Units (GPUs) to
implement massive parallelisation of the propagation of motion in the Sun-Earth-Moon gravitational envi-
ronment. GPUs, when compared to Central Processing Units (CPUs), are capable of performing simpler
independent tasks in parallel in a much faster fashion. CUDAjectory, a CUDA-based software developed
by the European Space Operations Centre, has been employed: parallel computations are used to obtain, in
reasonable amounts of computational time, dynamical maps starting from hundreds of thousands of starting
states. Some example cases are discussed, in an effort to show the type of analyses that can be performed,
showing the logic followed in the samples creation, propagation and post-processing. The inner workings of
the propagating tools are also introduced, showing how the advantage with respect to CPU parallelisation is
achieved. The results of this work are the creation of a set of pictorial maps to visualise cislunar dynamical
behaviours and patterns. The paper focuses on some example starting locations, considered as of interest
for future exploration, such as Near Rectilinear Halo Orbits, or Distant Retrograde Orbits. These maps can,
for example, link initial states and manoeuvres magnitudes to possible future disposal locations, such as
lunar impacts or escapes in solar orbit
Singular perturbation for a two-class Processor-Sharing queue with impatience
Get PDFA two-class Processor-Sharing queue with one impatient class is studied. Local exponential decay rates for its stationary distribution are established in the heavy traffic regime where the arrival rate of impatient customers grows proportionally to a large factor A. This regime is characterized by two time-scales, so that no general Large Deviations result is applicable
The ESA Hera Mission: Detailed Characterization of the DART Impact Outcome and of the Binary Asteroid (65803) Didymos
Get PDFHera is a planetary defense mission under development in the Space Safety and Security Program of the European Space Agency for launch in 2024 October. It will rendezvous in late 2026 December with the binary asteroid (65803) Didymos and in particular its moon, Dimorphos, which will be impacted by NASA's DART spacecraft on 2022 September 26 as the first asteroid deflection test. The main goals of Hera are the detailed characterization of the physical properties of Didymos and Dimorphos and of the crater made by the DART mission, as well as measurement of the momentum transfer efficiency resulting from DART's impact. The data from the Hera spacecraft and its two CubeSats will also provide significant insights into asteroid science and the evolutionary history of our solar system. Hera will perform the first rendezvous with a binary asteroid and provide new measurements, such as radar sounding of an asteroid interior, which will allow models in planetary science to be tested. Hera will thus provide a crucial element in the global effort to avert future asteroid impacts at the same time as providing world-leading science
Hybrid CFD simulations of two-phase flows in inline flow splitters
Get PDFInline fluid separation using a swirl element is a recent technology for oil/gas extraction. The static swirl element, installed inside the pipeline, is able to transform part of the arriving axial momentum into an azimuthal momentum when the flow goes through its blades. This produces centrifugal forces up to 100 times the gravitational acceleration separating then the phases based on their densities, the heavy phase is pushed next to the wall and the light one to the centre to be recovered at the outlet of the separator by a pick-up tube. The current study is part of a European project TOMOCON aims at developing CFD methods in the IMFT inhouse code JADIM to simulate the two phase flow separation. Since the scales are ranging from 1 m the length of the device (pipe, swirl element) to 1 mm which is the size of the smallest bubbles and drops, the numerical strategy needs to combine Eulerian and Lagrangian schemes. First, taking into account the complexity of the geometry, the pipe, the swirl element and the pick-up tube are simulated using Immersed Boundary Method (IBM) for solid/fluid interaction on a regular Cartesian mesh. The flow being highly turbulent that Direct Numerical Simulation (DNS) is not possible, Large Eddy Simulation (LES) is considered and the turbulence is modeled using mixed dynamic Smagorinsky model. Then the Lagrangian solver is used to track the dispersed phase (droplets/bubbles). Once the separation is done and the bubbles/droplets accumulation takes place leading to large volume of gas/oil compared to the mesh size, we switch to the Volume of Fluid (VoF) method to simulate the core inside the heavy phase. Hybrid models to combine LES/IBM, IBM/Lagrangian tracking et Lagrangian tracking/VoF are proposed and validated to enable the simulation of the inline fluid separation process. The numerical results help to fix the physical parameters influencing the separation and controlling the efficiency and validate models with experimental data from TU Delft, HZDR and TUL
Etude de la tolérance au déficit hydrique et au stress thermique chez le carthame (Carthamus tinctorius L.) et leurs effets sur le rendement et la qualité des huiles.
Get PDFLe travail présenté opte pour un objectif d’évaluation des potentialités de tolérance d’une espèce oléagineuse aux stress abiotiques notamment le déficit hydrique et l’hyperthermie. Le carthame est l’une de ces espèces qui constituerait comme un candidat potentiel pour combler le retard enregistré dans la filière oléagineuse en Algérie. L’étude présentée évalue le comportement, la teneur et la composition chimique de l’huile des graines mûres de trois génotypes de carthame d’origine, Française, Syrienne et Algérienne, cultivés durant trois campagnes (2015, 2016, 2017) à Tiaret. Les résultats indiquent que la teneur en huile des graines varie parmi les génotypes conduits et à travers les trois années pour se limiter à des valeurs comprises entre 22.8% et 35.68%. La composition chimique démontre que durant les trois années d’expérimentation, l’huile extraite est constituée essentiellement d’un acide gras polyinsaturé, l’acide linoléique. Les teneurs de ce composé varient entre 74.4 et 79.9%. Ces valeurs limites sont détenues par Halab et la population locale Touggourt respectivement. Il a été constaté que cette variation est conditionnée entre autres par le bilan pluviométrique annuel, où l’accroissement de la précipitation s’accompagne d’une augmentation de la teneur de cet acide. L’acide oléique détient une part importante dans cette composition pour atteindre des valeurs oscillant entre 9.8 et 14.7%. La teneur en acides gras saturés s’annonce faible, atteignant une limite maximale de 9%. Cette huile s’avère aussi riche en d’autres composants qui présentent un intérêt primordial dans la nutrition et la santé humaine comme les phytostérols et les polyphénols. Ce dernier composant se distingue par une activité antioxydante conjuguée élevée. Les stress abiotiques ont grandement amélioré le statut de phytostérols dans la partie insaponifiable de l’huile de carthame. Alors que, l’accumulation des polyphénols s’avère variable en fonction du génotype testés et le régime hydrique adopté. L’importance de l’utilisation médicinale de l’huile de carthame serait davantage concrètable grâce à ses effets anti-âges liés particulièrement à ses activités anti-élastase et anticollagénase élaborées. Cette étude a permis aussi de mettre en évidence la richesse du tourteau de carthame en protéine qui pourrait être utilisé comme aliment de bétail. A la lumière des résultats de ce travail, il a été révélé que l’importance attribuée au carthame et liée grandement à la richesse de ses graines en huile ainsi qu’en composantes chimiques ayant d’intérêt cosmétique et alimentaire remarquable. Par conséquent, l’adaptation de cette espèce oléagineuse dans les aires semi-arides algériennes constituerait un départ réel pour y être exploitée et introduite dans le système de production agricole national
Busy and confused? High risk of missed alerts in the cockpit: An electrophysiological study
Get PDFThe ability to react to unexpected auditory stimuli is critical in complex settings such as aircraft cockpits or air traffic control towers, characterized by high mental load and highly complex auditory environments (i.e., many different auditory alerts). Evidences show that both factors can negatively impact auditory attention and prevent appropriate reactions. In the present study, 60 participants performed a simulated aviation task varying in terms of mental load (no, low, high) concurrently to a tone detection paradigm in which the complexity of the auditory environment (i.e., auditory load) was manipulated (1, 2 or 3 different tones). We measured both detection performance (miss, false alarm, d’) and brain activity (event-related potentials) associated with the target tone. Our results showed that both mental and auditory loads affected target tone detection performance. Importantly, their combined effects had a large impact on the percentage of missed target tones. While, in the no mental load condition, miss rate was very low with 1 (0.53%) and 2 tones (1.11%), it increased drastically with 3 tones (24.44%), and this effect was accentuated as mental load increased, yielding to the higher miss rate in the 3-tone paradigm under high mental load conditions (68.64%). Increased mental and auditory loads and miss rates were associated with disrupted brain responses to the target tone, as shown by reduced P3b amplitude. In sum, our results highlight the importance of balancing mental and auditory loads to maintain efficient reactions to alarms in complex working environment
Anticipated Geological Assessment of the (65803) Didymos–Dimorphos System, Target of the DART–LICIACube Mission
Get PDFOn 2022 September 26, the DART spacecraft will impact the surface of Dimorphos, the ∼160 m size satellite of the binary near-Earth asteroid (NEA) (65803) Didymos. What will be observed on the surfaces of both asteroids and at the DART impact site is largely unknown, beyond the details of Didymos revealed by previous Arecibo and Goldstone radar observations. We present here the expected DART and LICIACube observations of the Didymos system and discuss the planned mapping strategies. By searching similar geological features and processes identified on other NEAs, we constrain the impact conditions that DART might encounter at Dimorphos, assessing both the asteroid's surface and interior structure
Vers une utilisation de l'Intelligence Artificielle dans un modèle numérique de climat
Get PDFDans un modèle de climat, les paramétrisations physiques ont pour rôle de représenter l’effet moyen des processus sous-maille (e.g., convection profonde, turbulence) à la résolution du modèle. Elles sont numériquement efficaces, mais aussi la principale source d’incertitude dans les modèles de climat. Depuis quelques années, l’utilisation des techniques d’Intelligence Artificielle (IA), en particulier celle des réseaux de neurones (Neural Networks, NN), a permis de développer des paramétrisations d’un genre nouveau. Contrairement aux paramétrisations physiques traditionnelles, développées en utilisant des connaissances théoriques et des observations, les paramétrisations apprises sont ajustées à un jeu de données issu d’une simulation haute-résolution directement, dans le but d’apprendre une représentation plus précise des différents processus sous-maille. Les premiers résultats sont prometteurs, mais il reste des problèmes majeurs à résoudre avant leur utilisation dans un modèle de climat, en remplacement des paramétrisations traditionnelles. Le but de cette thèse est l’étude de quelques-uns de ces problèmes à l’aide de modèles jouets et à travers l’apprentissage d’une partie de la paramétrisation physique implémentée dans un modèle de climat. Tout d’abord, l’utilisation d’une physique apprise par des NN peut rendre instable le modèle de climat dans lequel elle est branchée. La première partie de cette thèse est dédiée à l’étude de cette instabilité avec l’aide d’un modèle jouet, spécifiquement développé pour étudier ce problème. Un deuxième problème concerne la performance des physiques apprises, une fois celles-ci branchées dans un modèle de climat, notamment en terme de climatologie du modèle (e.g., biais de longterme). Nous avons développé une méthode permettant d’ajuster les statistiques de long-terme d’une physique apprise. La méthode a été illustrée à l’aide de modèles jouets. Enfin, nous nous sommes intéressés à l’apprentissage d’un schéma de convection profonde à l’aide de NN. Le schéma appris a été implémenté avec succès dans le modèle de climat global développé au CNRM, ARPEGE-Climat, en remplacement du schéma physique. Nous avons réalisé plusieurs années de simulation sans constater de divergence du modèle. Les principales caractéristiques du climat moyen sont bien représentées. La physique apprise est interprétée en utilisant des outils d’analyse de sensibilité. Cette étude pose quelques bases en vue du développement de paramétrisations apprises à partir de données haute-résolution, dont l’utilisation pourrait améliorer la précision des modèles numériques du climat
Holistic Operability Projection during Early Aircraft Design
Get PDFAircraft operational performance is one of the key drivers to airline profitability and punctuality. Along with safety and technical performance, aircraft operational performance needs to be projected from the early stages of development to design an aircraft that can fully meet the expectations of airlines and passengers. The ability of a system to meet its operational requirements in terms of reliability, availability and costs is termed as ‘Operability’. This paper proposes a method to model the operability of an aircraft during early design and use it to predict its operational performance. Initially, in-service data is used to create a reference baseline for a system of interest. For a new design, the designers evaluate the changes (deltas) in terms of few high-level metrics from an operations point of view called Consolidated Operability Metrics. An operability model is developed using Bayesian networks that is then used to project the changes in operational performance of the new design in comparison to the baseline. This method will help aircraft architects in conducting trade-off studies during early design from an operational point of view