Portail HAL ONERA
Not a member yet
12842 research outputs found
Sort by
Increased Hydrogen escape from Mars atmosphere during periods of high obliquity
No full textInternational audienceIt is still unknown how much water has escaped from Mars during its history. Hydrogen escape from Mars’s atmosphere probably played a major role in drying the planet, but present-day H loss rates (~3 × 1026 atoms per second on average) cannot explain the geological evidence for the large volumes of liquid water on ancient Mars. Here we used the three-dimensional Mars-Planetary Climate Model to show that H loss rates could have increased by more than one order of magnitude (6 × 1027 atoms per second) during higher spin axis obliquity periods, notably in the last few million years when Mars’s obliquity was about 35° on average. The resulting accumulated H escape over Mars’s history translates into an ~80 m global equivalent layer, which is close to the lower limit of geological estimates, assessing the major role of atmospheric escape in drying Mars
Performance Evaluation of the ITU-R Prediction Model of the Rain Attenuation in Earth-Satellite Links
No full textInternational audienceHigh throughput satellite communication systems, need to utilize less congested high-frequency bands, such as the Ka-and Q-bands and above with a large available radio channel bandwidth. However, radio wave propagation at these frequencies is hindered by increased attenuation and impairment caused by tropospheric phenomena, particularly rainfall. It is crucial for such systems to characterize rain attenuation in the slant path for the design and setup of the satellite communication system. The statistical prediction model of rain attenuation in Earth-space paths, provided in Section 2.2.1.1 of Recommendation ITU-R P.618-14, depends on several parameters, including the rainfall rate and the rain height. This paper investigates the performance of predicting rainfall rate with ITU-R P.837-7 and predicting rain height with ITU-R P.839-4 based on the digital maps from the ECMWF ERA5 database and examines their influence on the rain attenuation prediction model in ITU-R P.618-14
Model Checking de LTL sur Traces Finies et Infinies avec Domaines Concrets
No full textInternational audienceIl existe différentes sémantiques pour la logique temporelle linéaire (LTL) concernant la prise en compte de traces finies. Bien que plusieurs d’entre elles puissent s’avérer utiles en fonction du contexte de vérification, aucun cadre de vérification ne permet de gérer leur diversité de manière simple. Une autre limitation des outils de vérification de LTL actuels est leur incapacité à traiter les domaines concrets (entiers bornés et infinis, nombres réels, etc.). Nous présentons une approche permettant de faire du model checking de LTL sur des traces finies et infinies avec des domaines concrets. Notre méthode repose sur un solveur SMT et sur le model checking borné (BMC). Nous présentons également quelques expérimentations et comparons notre outil avec NuSMV et nuXmv
Implementation of airborne ML models with semantics preservation
No full textInternational audienceMachine Learning (ML) may offer new capabilities in airborne systems. However, as any piece of airborne systems, ML-based systems will be required to guarantee their safe operation. Thus, their development will have to be demonstrated to be compliant with the adequate guidance. So far, the European Union Aviation Safety Agency (EASA) has published a concept paper and an EUROCAE/SAE group is preparing ED-324. Both approaches delineate high-levelobjectives to confirm the ML model achieves its intended function and maintains training performance in the target environment.The paper aims to clarify the difference between an ML model and its corresponding unambiguous description, referred to as the Machine Learning Model Description (MLMD). It then refines the essential notion of semantics preservation to ensure the accurate replication of the model. We apply our contributions to several industrial use cases to build and compare several target models
When Prediction Meets Perception: The Effect of Action-Based Expectations on Visual Perception
No full textInternational audienceEngaging in voluntary actions enables us to anticipate and influence changes in the external world. The ability to predict outcomes play a significant role in shaping sensory processing and perception. For instance, research showed improved perceptual performance and also sensory attenuation for predicted outcomes compared to unpredicted ones. However, the neuronal mechanisms by which prediction -particularly those derived from voluntary actions- shape perception remains unclear despite the development of several theoretical frameworks. This study aims at contributing to our understanding of how action-based predictions shape sensory experience. Participants complete a 2-interval-forced-choice task in which they are presented two consecutive gratings and have to indicate whether the second grating is tilted clockwise or counterclockwise compared to the first. In the active condition, the orientation of the first stimulus is predictable from participants’ action, whereas in the passive condition, participants cannot predict the upcoming stimulus. We expect higher perceptual discrimination performance in the active condition compared to the passive condition. The results will be interpreted within the frameworks of the cancellation and sharpening models of sensory prediction. Data collection for the experiment is currently underway
Exploring coupling effects between airframe and propulsion in parametric models for electric regional aircraft with range extender
No full textInternational audienceRecently, the Hybrid Aircraft: academic reSearch on Thermal and Electrical Components and Systems (HASTECs) project developed an advanced serial hybrid electric propulsion system, for regional transport aircraft, with a parametric model in each propulsion component. This approach integrated main physical phenomena covering several fields (electrical, mechanical, thermal) for sizing and evaluation of device losses and weights. The study highlighted the necessity for propulsive system optimization rather than component optimization, intending to achieve an optimal balance between efficiency and mass. However, the project decoupled the airframe from the propulsion system, resulting in the inability to ascertain if this compromise was reflected in the overall aircraft optimization. Concerning the Overall Aircraft Design (OAD) front, the FAST-OAD software has been used in several studies to analyze the propulsion-airframe integration but it relied on low-fidelity models for the hybrid propulsion system. This approach introduces significant uncertainties in the results, especially due to the lack of consideration for certain electrical phenomena, such as partial discharges, making it difficult to satisfy design constraints.The objectives of this study are twofold: first, to present a new OAD process by coupling HASTECs and FAST-OAD software; second, to structure the oad process with the integration of different topologies of aircraft (turboelectric, serial-hybrid, full electric), the number of motors, and a method to determine the maximum possible range in pure electric mode. Finally, the objective is to ensure the robustness and relevance of the final OAD results despite strong interactions with an acceptable computation cost facing the system complexity for further optimizations. The case study for the new design process is the design exploration of an electric regional transport aircraft with a range extender
Detection of Spruce Budworm with Sentinel-1 Time Series
No full textInternational audienceCan Sentinel-1 time series enhance the detection and management of spruce budworm outbreaks in forest ecosystems
Analyse l'approche à vitesse du son artificielle de Rossow pour le calcul d'écoulements compressibles dans la limite incompressible
No full textInternational audienceWe look at a simple modification of the Roe approximate Riemann solver in the incompressible limit, considering the Rossow's artificial speed of sound approach, for compressible flow simulations with inflow Mach numbers decreased up to 10 -6 . Although published by C.-C. Rossow more than twenty years ago, this approach appears surprisingly to be much less popular than the Roe-Turkel scheme introduced in 1999 by Guillard and Viozat, or than the Rieper's low Mach number fix proposed in 2011. In addition, no asymptotic discrete analysis of this modification of the Roe scheme is known in the literature. A common feature of these modified Roe schemes is that they all aimed at introducing a necessary rescaling of the original Roe's matrix-valued dissipation for the computation of low Mach number flows. In this paper, the analysis conducted gives a new insight into the artificial speed of sound approach, with emphasis on properties of the corresponding rescaled dissipation matrix for the solution accuracy, the Von Neumann stability, and the asymptotic behavior of the discrete solutions in the incompressible limit. It is especially shown that, first, the discrete pressure field recovers the proper order of pressure disturbances in space, and second, in opposite to the Rieper's fix, the Rossow's artificial speed of sound is not prone to pressure checkerboard modes. This modification of the Roe scheme has also almost identical discrete asymptotic properties, and the same stringent Von Neumann stability condition, than the Roe-Turkel scheme, although being potentially slightly less dissipative.Nous regardons une simple modification du solveur de Riemann approché de Roe dans la limite incompressible, considérant l'approche à vitesse du son artificielle de Rossow, pour des simulations d'écoulement compressible avec des nombres de Mach d'entrée abaissés jusqu'à . Bien que publiée par Rossow il y a plus de 20 ans, cette approche apparait de manière surprenante être beaucoup moins populaire que le schéma de Roe-Turkel scheme introduit en 1999 par Guillard et Viozat ou que le correctif bas Mach de Rieper proposé en 2011. De plus, aucune analyse discrète de cette modification du schéma de Roe n'est connue dans la littérature. Un aspect commun à ces schémas de Roe modifiés est qu'ils visent tous à introduire un rééquilibrage de la matrice de dissipation originale de Roe pour le calcul d'écoulements à faible Mach. Dans ce papier, l'analyse conduite donne un nouvel aperçu de l'approche à vitesse du son artificielle, avec l'accent mis sur les propriétés de la matrice de dissipation rééquilibrée correspondante pour la précision de la solution, la stabilité de Von Neumann et l'analyse asymptotique de solutions discrètes dans la limite incompressible. Il est particulièrement montré que, premièrement, la pression discrète retrouve le bon ordre des fluctuations spatiales et, deuxièmement, à l'opposé du correctif de Rieper, la vitesse du son artificielle de Rossow n'est pas enclin à des modes de pression en damier. Cette modification du schéma de Roe a également presque les mêmes propriétés asymptotiques, et la même contrainte rigoureusement de stabilité que le schéma de Roe-Turkel, bien qu'étant potentiellement moins dissipatif
Multifidelity Bayesian Sequential Optimization and Reliability Assessment for Aerospace Systems Design
No full textInternational audienceThe design of aerospace systems requires the use of computationally intensive numerical solvers. Most often, the designer has access to simulation codes with different levels of fidelity, characterized by different accuracy and computational cost. In addition, certain phenomena manifest a stochastic nature that needs to be accounted for in the design process. Incorporating multiple sources of uncertainty through reliability-based design optimization (RBDO) then becomes a challenging task. Among the potential strategies, decoupled approaches such as SORA iterate between deterministic optimization and reliability analysis. However, when employing high-fidelity solvers, such methods still induce significant computational burdens. A way to further alleviate the computational cost issues is to involve models of fidelity throughout the RBDO problem. In this paper, a Bayesian approach with multifidelity surrogate models is proposed to tackle the optimization problems within SORA. This approach facilitates the incorporation of additional sources of information provided by lower fidelity models. Furthermore, the surrogate models are built in an augmented space allowing to reuse information along the RBDO iterations. The efficiency of the proposed framework is compared to reference approaches on four test cases with increasing complexity, and two aerospace realistic cases concerning the optimization of a solid-propellant rocket booster and a sounding rocket
Conception d'antenne pour l'analyse modale du bruit large bande se propageant en conduit
No full textInternational audienceThe azimuthal and radial content of the acoustic field propagating in a duct can be experimentally characterized using an array of phased, wall-mounted microphones. Depending on the number of modes to reconstruct, the number of available microphones and their positions, such an inverse problem may be ill-conditioned, leading to an inaccurate estimation of the modal content of the acoustic signal. One solution involves introducing prior information about the unknown modal content, as achieved through regularization or compressive sensing methods, often based on empirical evidence. Another solution is to better choose the microphone positions. Following this latter approach, an automatic design procedure to identify the microphone array configuration that yields a well-posed inverse problem across the widest possible frequency range is proposed. The microphone array design problem is formulated as the minimization of the condition number of a Gram matrix, derived from the propagation functions between modes and microphones. Three other design experiment criteria are also investigated and compared. The proposed antenna design algorithm is both automatic and computationally efficient, thanks to a gradient-based solution method. The number of microphones is fixed and the frequency band within which the antenna is theoretically adequate is investigated. The present methodology is applied to the design of ONERA's test bench. Synthetic observations are considered to assess the antenna's reconstruction performances relative to a baseline configuration. Robustness with respect to frequency, flow presence and small microphone position shifts are investigated. The selected design is subsequently validated through an experimental campaign.Le contenu azimutal et radial du champ acoustique se propageant dans un conduit peut être caractérisé expérimentalement à l'aide d'un réseau de microphones phasés montés en paroi du conduit. En fonction du nombre de modes à déterminer, du nombre de microphones disponibles et de leurs positions, le problème à résoudre peut être mal conditionné, conduisant à une estimation erronée du contenu modal du signal acoustique. Une solution consiste à réduire la dimension du problème en imposant des informations a priori sur les inconnues, via des méthodes de régularisation ou de compressive sensing, souvent basées sur des preuves empiriques. Une autre approche est de sélectionner au mieux la répartition des microphones de l'antenne. Dans cet article, une procédure de conception automatique est proposée pour identifier la position des microphones conduisant à un problème mathématique inverse bien posé sur la gamme de fréquences la plus large possible. Le problème de conception d'antenne repose sur la minimisation du conditionnement de la matrice de Gram associée aux fonctions de propagation entre les modes et les microphones. Trois autres critères de conception expérimentale sont également étudiés et comparés. L'algorithme de conception proposé, basé sur une méthode de descente de gradient, est automatique et particulièrement efficace en terme de temps de calculs. Le nombre de microphones est fixe et la bande de fréquence dans laquelle l'antenne est théoriquement adéquate est étudiée. La méthodologie présentée est appliquée à la configuration du banc modes tournants de l'ONERA. Une évaluation sur simulations permet d'évaluer les performances de reconstruction obtenues en environnement parfaitement maîtrisé. La robustesse de l'approche par rapport à la fréquence, à la présence d'écoulement ainsi qu'à de légères variations de position des microphones est étudiée. L'antenne sélectionnée est ensuite évaluée lors d'une campagne expérimentale dédiée