39936 research outputs found
Sort by
Determine creep and rheological parameters for hot isostatic pressing (HIP) Model of 316L. Master thesis
Full text linkIn this thesis, experiments which are needed to obtain initial and secondary creep parameters of full dense 316L and rheological parameters of porous 316L are presented.
After completing these experiments, parameters needed for HIP simulation model will be determined. Subsequently densification process of different geometries during HIP will be simulated with the determined parameters in order to check the numerical problems and verify the quality of HIP simulation results.En esta tesis se presentan los experimentos que se necesitan para obtener los parámetros de fluencia inicial y secundaria de los aceros inoxidables 316L y 304 completamente densos y los parámetros reológicos de 316L con porosidades.
Después de completar estos experimentos, los parámetros necesarios para el modelo de simulación HIP serán determinados. Posteriormente, el proceso de densificación de diferentes geometrías durante HIP se simulará con los parámetros determinados con el fin de comprobar los problemas numéricos y verificar la calidad de los resultados de la simulación HIP. HIP son las siglas en inglés de Hot Isostatic Pressing (Prensado Isostático en caliente).Ingeniería Industria
Multi-Disciplinary Optimization of Space Mobility Maneuvers for Commercial Applications of Micro Propulsion
No full textThis thesis aims to develop a comprehensive framework for designing and optimizing space mobility solutions, with a particular emphasis on commercial applications. Focusing on small satellites equipped with electric propulsion in Low Earth Orbit, the research jointly explores various aspects of mission design optimization, including propulsion system design, maneuver optimization, and constellation deployment strategies. Chapter 2 introduces a multi-objective concurrent trajectory/system design methodology. The proposed heuristics-based architecture facilitates rapid exploration of a wide design space while considering hybrid variables, complex operational constraints, and subsystem interactions. The framework is applied to modular propulsion optimization, demonstrating its advantages for mission design/planning over Commercial-Off-The-Shelf propulsion solutions. In Chapter 3, a novel methodology for optimizing variable-performance electric
propulsion transfers is proposed. By relating maneuvering efficiency to specific impulse through a parametric function, significant improvements in propellant consumption and maneuver duration over the constant case are achieved, enhancing trade-offs and enlarging the possible set of solutions for small satellite mobility. Chapter 4 delves into the use of propulsion for constellation deployment strategies. The work proposes a numerical optimization method based on the developed framework, which is applicable to both electric and chemical propulsion systems. Two deployment methods are analyzed, highlighting the importance of launcher selection, propulsion technology, and atmospheric drag effects on deployment optimality. Expanding on the constellation deployment in Chapter 5, logistics related to selecting suitable launch opportunities are explored. A hybrid multiobjective method is proposed to select launch opportunities and maneuvering strategies
concurrently. Case studies underscore the methodology’s effectiveness in identifying optimal deployment strategies for various constellation sizes and launch scenarios. Overall, the thesis establishes a comprehensive mathematical framework for the multidisciplinary optimization of propulsion-enabled space missions. The research offers valuable insights into optimizing space missions, providing practical solutions for propulsion system design, maneuver optimization, and constellation deployment strategies.Questa tesi mira a sviluppare un framework per la progettazione e l’ottimizzazione
di soluzioni di mobilità spaziale, con particolare enfasi sulle applicazioni commerciali.
Concentrandosi su satelliti di piccole dimensioni provvisti di propulsione elettrica in orbita
terrestre bassa, la ricerca esplora congiuntamente vari aspetti dell’ottimizzazione
del progetto di missioni spaziali, inclusa la progettazione del sistema di propulsione,
l’ottimizzazione delle manovre e le strategie di dispiegamento delle costellazioni. Nel
Capitolo 2 viene introdotta una metodologia di progettazione del sistema/traettoria concorrente
multi-obiettivo. L’architettura proposta, basata su euristiche, facilita l’esplorazione
rapida di un ampio spazio di progettazione, considerando variabili ibride, vincoli operativi
complessi e interazioni tra i sottosistemi. Il framework viene applicato all’ottimizzazione
di sistemi propulsivi modulari, dimostrandone i vantaggi per la progettazione/pianificazione
delle missioni rispetto alle soluzioni di propulsione disponibili in commercio. Nel Capitolo
3 viene proposta una nuova metodologia per ottimizzare i trasferimenti di propulsione
elettrica a prestazioni variabili. Relazionando l’efficienza della manovra a impulso specifico
attraverso una funzione parametrica, si ottengono miglioramenti significativi nel consumo
di propellente e nella durata della manovra rispetto al caso costante, incrementando
le possibilità di trade-off e ampliando il possibile insieme di soluzioni per la mobilità
dei piccoli satelliti. Il Capitolo 4 approfondisce l’uso della propulsione per le strategie
di dispiegamento delle costellazioni. Il lavoro propone un metodo di ottimizzazione numerica
basato sul framework sviluppato, che è applicabile sia ai sistemi di propulsione
elettrica che a quelli chimici. Vengono analizzati due metodi di dispiegamento, evidenziando
l’importanza della selezione del lanciatore, della tecnologia di propulsione e degli
effetti dell’attrito atmosferico sull’ottimalità del dispiegamento. Espandendo sul dispiegamento
della costellazione nel Capitolo 5, viene esplorato il problema logistico legato
alla selezione delle opportunità di lancio idonee. Viene proposto un metodo ibrido multiobiettivo
per selezionare le opportunità di lancio e le strategie di manovra contemporaneamente.
Gli esempi proposti evidenziano l’efficacia della metodologia nell’identificare
strategie di dispiegamento ottimali per varie dimensioni di costellazioni e scenari di lancio.
In generale, la tesi stabilisce un framework matematico completo per l’ottimizzazione
multidisciplinare delle missioni spaziali provviste di propulsione. La ricerca offre una
preziosa comprensione sull’ottimizzazione delle missioni spaziali, fornendo soluzioni
pratiche per la progettazione del sistema di propulsione, l’ottimizzazione delle manovre e
le strategie di dispiegamento delle costellazioni.Esta tesis tiene como objetivo desarrollar un framework integral para diseñar y optimizar
soluciones de movilidad espacial, con especial énfasis en las aplicaciones comerciales.
Centrándose en satélites pequeños equipados con propulsión eléctrica en órbita
terrestre baja, la investigación explora conjuntamente varios aspectos de la optimización
del diseño de la misión, incluyendo el diseño del sistema de propulsión, la optimización de
maniobras y las estrategias de despliegue de constelaciones. En el Capítulo 2 se presenta
una metodología de trayectoria/diseño de sistemas simultáneos multiobjetivo. La arquitectura
heurística propuesta facilita la exploración rápida de un amplio espacio de diseño
mientras se consideran variables híbridas, restricciones operativas complejas e interacciones
de subsistemas. El framework se aplica a la optimización de la propulsión modular,
lo que demuestra sus ventajas para el diseño/planificación de misiones sobre las soluciones
de propulsión comercial. En el Capítulo 3 se propone una nueva metodología para
optimizar las transferencias de propulsión eléctrica de rendimiento variable. Al relacionar
la eficiencia de maniobra con un impulso específico a través de una función paramétrica,
se logran mejoras significativas en el consumo de propelente y la duración de maniobra
en el caso constante, mejorar las compensaciones y ampliar el posible conjunto de soluciones
para la movilidad de los pequeños satélites. El Capítulo 4 profundiza en el uso de
la propulsión para las estrategias de despliegue de constelaciones. El trabajo propone un
método de optimización numérica basado en el framework desarrollado, que es aplicable
tanto a sistemas de propulsión eléctrica como química. Se analizan dos métodos de despliegue,
destacando la importancia de la selección del lanzador, la tecnología de propulsión
y los efectos de arrastre atmosférico en la optimalidad del despliegue. Ampliando
el despliegue de constelaciones en el Capítulo 5, se explora la logística relacionada con
la selección de oportunidades de lanzamiento adecuadas. Se propone un método multiobjetivo
híbrido para seleccionar oportunidades de lanzamiento y estrategias de maniobra
al mismo tiempo. Los estudios de casos subrayan la eficacia de la metodología para
identificar estrategias óptimas de despliegue para diversos tamaños de constelaciones y
escenarios de lanzamiento. En general, la tesis establece un framework matemático integral
para la optimización multidisciplinaria de las misiones espaciales con propulsión.
La investigación ofrece información valiosa sobre la optimización de las misiones espaciales,
proporcionando soluciones prácticas para el diseño de sistemas de propulsión,
optimización de maniobras y estrategias de implementación de constelaciones
On the dimension of orbits of matrix pencils under strict equivalence
Full text linkWe prove that, given two matrix pencils L and M, if M belongs to the closure of the orbit of L under strict equivalence, then the dimension of the orbit of M is smaller than or equal to the dimension of the orbit of M, and the equality is only attained when M belongs to the orbit of L. Our proof uses only the majorization involving the eigenstructures of L and M which characterizes the inclusion relationship between orbit closures, together with the formula for the codimension of the orbit of a pencil in terms of its eigenstruture.The authors thank Andrii Dmytryshyn for suggesting the use of Theorem 6 to prove the main result. This work is part of grant PID2023-147366NB-I00 funded by MICIU/AEI/ 10.13039/501100011033 and FEDER/UE. Also funded by RED2022-134176-T and by the program Excellence Initiative - Research University at the Jagiellonian University in Kraków
Leveraging unlabeled data for lung sound classification through self-supervised contrastive learning
No full textThis paper focuses on the classification of lung sounds as can be instrumental in diagnosing respiratory diseases, that are one of the leading causes of death. We address two intrinsic challenges associated with the application of deep learning models to this task. The first one is the need of large databases annotated by expert medical staff. The second one is the presence of a strong class imbalance in the datasets primarily due to the predominance of non-pathological respiratory sounds. For overcoming these issues we propose a deep convolutional network model that is trained using the contrastive self-supervised learning paradigm. This technique is able to generate useful audio data representations from unlabeled data that can be effectively transferred to the target task employing a limited amount of annotated data. We have evaluated the developed systems on the well-known ICBHI dataset that contains respiratory cycles categorized into four different classes. Results show that our approach outperforms the conventional supervised learning model when the size of available labeled data is reduced. With 40% of annotated data, self-supervision achieves a relative improvement with respect to the baseline of 12.1%, 8.6%, 16% and 66.7% in score, accuracy, and sensitivity respectively, while getting a reduction of 1.0% in specificity. Finally, to corroborate our findings, we have also assessed our system on the SPRSound database, confirming the same trends. We believe that the findings in this paper enlightens the path towards the use of unlabeled data in biomedicine alleviating the need of large annotated datasets.The authors acknowledge the support of the Comunidad de Madrid through grants IntCARE-CM and PEJ-2021-AI/TIC-22744 and the Spanish State Research Agency (MICIU/AEI/10.13039/501100011033) and FEDER, UE through projects PID2020-115363RB-I00 and PID2023-146684NB-I00
Lola Rivas Cherif: La mujer fuerte del presidente de la República
No full textA lo largo de la historia, algunos personajes de relevancia han permanecido en un discreto segundo plano. Este es el caso de Lola Rivas Cherif (1904-1993), una mujer que, desde la sombra, encarnó los valores de la República y fue un pilar fundamental en la vida de Manuel Azaña, tanto en lo personal como en lo ideológico. Junto a su marido, vivió acontecimientos históricos trascendentales: la proclamación de la II República, la sublevación militar y la posterior guerra civil, y el peor de los episodios, el exilio a una Francia ocupada por los nazis. Tras la muerte de Azaña, se exilió en México donde falleció cuarenta y dos años después. Su actitud siempre estuvo marcada por la reconciliación. Aunque el dolor fue una constante en su vida, nunca expresó odio ni rencor. Azaña la describió como mujer fuerte y no se equivocaba. Fue la sombra y el soporte de don Manuel. La mujer que un estadista necesitaba a su lado. Además, la historia le hace justicia por su labor en el cuidado y la difusión de las obras de Azaña. Lola Rivas Cherif, requería una biografía que visibilizara a la mujer del presidente de la II República Española
Effect of injection conditions on the non-linear behavior of the ECDI and related turbulent transport
Full text linkand perform two-dimensional kinetic simulations under several conditions to analyze the non-linear behavior and the induced transport. Fully periodic simulations, with conditions faithful to the linear theory, are analyzed first. In agreement with existing literature, they show the growth of ECDI modes, ion-wave trapping vortexes, and an induced cross field electron current in early simulation times. However, in contrast with similar works, non-linear saturation is observed and the plasma tends, in the long term, to a new equilibrium with mild oscillations and mild anomalous current. This evolution is consistent with what can be expected from energy conservation. The quenching of the oscillations seems to be highly related to the distortion of ion vortexes in phase space after a long-term interaction with the electrostatic wave. This result suggests that sustained oscillations and turbulent current could thrive if ions are renewed by, e.g., removing and injecting particles through axial boundaries instead of applying periodicity. This second type of simulation shows that injection conditions highly impact the late simulation behavior of ECDI oscillations, where we identify several regimes depending on the value of the ion residence time compared to the characteristic saturation time in the fully periodic case. The intermediate regime, where these two times are close, is the only one providing sustained oscillations and electron transport.This work has been supported by the R&D project PID2022-140035OB-I00 (HEEP) funded by MCIN/AEI/10.13039/501100011033 and by “ERDF A way of making Europe.” E. Bello-Benítez acknowledges the financial support by the European Research Council under the European Union's Horizon 2020 research and innovation programme (project ZARATHUSTRA, Grant agreement No. 950466)
Experimentally-informed numerical investigation of a high-power magnetically shielded Hall thruster operating on xenon and krypton
No full textThe operation of a kW magnetically shielded Hall effect thruster, with xenon and krypton as propellants, is numerically studied with a 2D (axial-radial) axisymmetric hybrid (particle-in-cell/fluid) code. Thrust efficiency with Kr is 2% to 6% lower than with Xe. Kr exhibits smaller propellant utilization, voltage utilization and divergence efficiencies. Contrary to what is observed in previous works, the current efficiency is higher with Kr. Apart from the dissimilar mass and electronic properties of Xe and Kr atoms, the widening of the ionization and acceleration zones with Kr, with respect to the Xe discharge, plays an important role in the Xe-Kr differences in performance. The level of plume velocity dispersion with both propellants is found similar due to two opposing effects: on the one hand, the stronger overlap of the ionization and acceleration zones with Kr leads to a larger dispersion on a per species basis; and, on the other hand, the smaller fraction of doubly-charged ions with Kr reduces the overall dispersion efficiency. Magnetic shielding is effective with both Xe and Kr as propellants, with slightly higher ion impact energy on the walls in the Kr discharge
The regulation of high‐risk AI systems in the AI Regulation: in particular, human oversight
No full text[ES] Este artículo analiza la regulación de los sistemas de Inteligencia Artificial de alto riesgo que se introduce a través del Reglamento 2024/1689, de Inteligencia Artificial. El análisis que se presenta a
continuación contextualiza el tratamiento de la cuestión haciendo referencia, en la primera parte del artículo, a la regulación de los sistemas de IA que el reglamento prohíbe y a la regulación de los sistemas de IA que la norma comunitaria somete a meras obligaciones de transparencia, para pasar, en una segunda parte, a analizar los sistemas de IA de alto riesgo. El trabajo se centra en el análisis de los requisitos y obligaciones que se imponen sobre los sistemas de IA de alto riesgo, haciendo especial énfasis en el requisito, y correlativas obligaciones, de supervisión humana. Finalmente, se extraen en la última sección una serie de conclusiones generales, que conectan el análisis de la cuestión de la supervisión humana de los sistemas de IA de alto riesgo con el resto del análisis de los sistemas de IA prohibidos y de aquellos sometidos a meras obligaciones de transparencia. El argumento principal del artículo es que, en lo tocante a la supervisión humana, las esferas respectivas de responsabilidad del proveedor, del «desplegador» y del supervisor humano serán de deslinde más complejo de lo que en teoría permite hacer ver la regulación establecida por el reglamento de IA.[EN] This article analyzes the regulation of high‐risk Artificial Intelligence systems introduced through the Artificial Intelligence Regulation 2024/1689. The analysis presented below contextualizes the issue by first referring to the regulation of AI systems that the regulation prohibits and those that the EU regulation subjects to mere transparency obligations. Then, in the second part, it moves on to analyze high‐risk AI systems. The study focuses on examining the requirements and obligations imposed on high‐risk AI systems, and it makes special emphasis on the requirement‐and corresponding obligations‐of human oversight. Finally, the last section presents a series of general conclusions, linking the analysis of human oversight in high‐risk AI systems with the broader discussion of prohibited AI systems and those subject to mere transparency obligations. The main argument of the article is that, regarding human oversight, the respective spheres of responsibility of the provider, the deployer and the human supervisor will be more complex to delineate than what the AI regulation theoretically suggests
Introduction to the Optical Networks Design and Modeling 2024 Special Issue
Full text linkIn this special issue, we include eight papers that were extended from the conference papers presented at the 28th International Conference on Optical Networks Design and Modeling 2024 (ONDM 2024), whose motto was "Optical networks at scale: coping with the scalability challenge of future smart networks." The objective of the conference and this special issue is the selection of the latest research on optical network architectures, control, modeling, and planning techniques driving the design of next-generation networks to meet the challenges of ubiquitous universal green AI. The overall research goal of the works is improving the capacity, efficiency, flexibility, availability, autonomy, resiliency, security, and intelligence of optical access and transport networks in a scalable way. The selected works described below meet the challenge, first looking into the design of digital twins for optical transport networks, investigating advanced mechanisms for failure detection and post-disaster data center content evacuation. Second, through the introduction of quantum technologies. Indeed, the amount of selected works including quantum reveals the growing interest on this topic: on the one hand, the application of quantum-computing-based algorithms to solve classic optical networking problems as an alternative to ILP, and, on the other, the focus on new quantum key distribution technologies capable of coexisting with classical channels either using optical fiber or free space optics
Silver or Lead: 40 Years of Narcofictions (1982 -2022)
Full text linkLas narcoficciones surgieron en América Latina (México y Colombia) a comienzos de la década de 1980, en un entorno marcado por la inestabilidad y la corrupción política. No fue hasta el surgimiento de las plataformas de streaming cuando se produjo su globalización. En este artículo, se analiza su desarrollo a lo largo de cuarenta años y se explican las razones de su éxito y difusión. Para sintetizar y catalogar los títulos existentes, se emplea una metodología mixta cuantitativo-cualitativa, sistemática, completa, explícita y reproducible, basada en el framework search, appraisal, synthesis, and analysis (SALSA), con base en una serie de criterios. Se concluye que los formatos más recurrentes son las biografías (biopics), destacando especialmente las de Pablo Escobar y Joaquín “el Chapo” Guzmán. En ellas, y a través de recuerdos que trascienden sus experiencias individuales, se construyen discursos de memoria colectiva. Por otro lado, es común justificar la intervención en la política latinoamericana mediante figuras como el agente de la Drug Enforcement Administration (DEA). A partir de 2005, se observa una evolución en el rol de los personajes femeninos que pasan de secundarios a protagonistas, con el éxito internacional de la producción colombiana Sin senos no hay paraíso. Se detecta que, a pesar de su representación en posiciones de poder, se siguen reproduciendo y perpetuando los roles de género tradicionales, siendo con frecuencia la maternidad un objetivo y, a la vez, un punto débil en estos personajes.Narcofictions emerged in Latin America (specifically Mexico and Colombia) in the early 1980s, in a context marked by political instability and corruption. It wasn’t until the rise of streaming platforms that the genre became a global phenomenon. This article examines the evolution of narcofictions over the past forty years and explores the reasons behind their success and widespread appeal. To compile and categorize existing titles, a mixed-methods approach is used --both quantitative and qualitative --based on a systematic, comprehensive, explicit, and reproducible methodology grounded in the SALSA framework: Search, Appraisal, Synthesis, and Analysis, with clearly defined criteria. The study finds that biopics are the most common format, especially those focused on figures like Pablo Escobar and Joaquín “El Chapo” Guzmán. These portrayals construct narratives of collective memory through personal recollections that transcend individual experiences. The involvement of the U.S. Drug Enforcement Administration (DEA) agent is often used to justify foreign intervention in Latin American politics. Since 2005, there has been a noticeable shift in female character roles, evolving from secondary figures to protagonists, notably with the international success of the Colombian series Sin senos no hay paraíso. However, despite their presence in positions of power, traditional gender roles continue to be reinforced, with motherhood frequently portrayed as both a central goal and a vulnerability for these characters