12096 research outputs found
Sort by
Diseño de una planta fotovoltaica en el polideportivos de Orvina, para la comunidad energética de la Txantrea.
No full textEste trabajo de fin de grado plantea el diseño de una instalación fotovoltaica en la cubierta del polideportivo municipal de Orvina, en el barrio de la Txantrea (Pamplona), con el objetivo de integrarse como base de la futura Comunidad Energética del barrio. La ubicación ha sido considerada de gran potencial por la Agencia Energética del Ayuntamiento de Pamplona, tanto por su superficie útil como por su proximidad a consumidores estratégicos.
La planta fotovoltaica diseñada permitirá abastecer energéticamente a viviendas particulares, edificios públicos cercanos como centros educativos, y dos puntos de recarga para vehículos eléctricos, reforzando el modelo de autoconsumo compartido. Este enfoque responde al marco legislativo vigente y a los objetivos del Plan Energético de Navarra (PEN 2030) y la Ley Foral 4/2022.
El proyecto incluye el análisis del marco normativo, el diseño y dimensionado técnico de la instalación, la simulación energética con herramientas como PVGIS y PVSyst, un estudio de viabilidad económico-energética y presupuesto. Con ello, se sientan las bases para la creación de una comunidad energética eficiente, realista y sostenible en el barrio de la Txantrea.Graduado o Graduada en Ingeniería en Tecnologías Industriales por la Universidad Pública de NavarraIndustria Teknologietako Ingeniaritzan Graduatua Nafarroako Unibertsitate Publikoa
Las emociones en momentos de desastres naturales
No full textVivimos momentos de cambios, y entre ellos en el planeta cada vez están más presentes las catástrofes naturales, con lo que conlleva a la afectación de la salud física, psicológica y social. Al hilo de lo anterior, voy a reflejar mis vivencias como persona y profesional en los terremotos de El Salvador de 2001 y la repercusión que tiene en la actualidad en las clases de salud mental que imparto a estudiantes de Grado en Enfermería de la Universidad Pública de Navarra
Análisis y agregación de modelos de predicción de series temporales
No full textEste trabajo investiga el uso de modelos univariados frente a los modelos multivariados en la predicción de series temporales, donde las predicciones de los modelos univariados dependen únicamente de una variable,
como puede ser la temperatura de una ciudad, mientras que los modelos
multivariados son capaces de capturar dependencias entre múltiples variables, como la temperatura y la humedad, para realizar la predicción. Las
series temporales son conjuntos de datos observados en función del tiempo,
como la temperatura diaria de una ciudad durante un año. También se
abordan las ventajas que puede suponer la agregación de predicciones, una
técnica que combina las predicciones de diferentes modelos para obtener
una estimación más robusta y precisa. Se detalla la metodología empleada
para la generación de predicciones, comenzando con el preprocesamiento
de los datos temporales, la modelización, la predicción y la agregación de
predicciones, para finalmente evaluar los resultados obtenidos.This study covers the use of univariate models versus multivariate
models in time series forecasting, where the predictions of univariate
models depend solely on a single variable, such as the temperature of
a city, while multivariate models can capture dependencies between multiple variables, such as temperature and humidity, to make predictions.
Time series are datasets observed over time, such as the daily temperature
of a city over the course of a year. The study also examines the advantages of prediction aggregation, a technique that combines the predictions
of different models to obtain a more robust and accurate estimate. The
methodology for generating predictions is detailed, starting with the preprocessing of time-dependent data, followed by modeling, prediction and
aggregation of predictions, and finally evaluating the results obtained.Graduado o Graduada en Ingeniería Informática por la Universidad Pública de NavarraInformatika Ingeniaritzako Graduatua Nafarroako Unibertsitate Publikoa
Optimal charging station deployment for drone-assisted delivery
No full textLast-mile delivery of goods made by drones is considered to be in its experimental phase. Nevertheless, international enterprises such as Amazon, Google, UPS or DHL are expanding new unmanned aerial vehicle technologies related to delivery issues. Flight range of drones is compromised due to the limited battery capacity and the payload of delivered parcels. This challenge is addressed through the placement of charging stations where drone batteries are recharged. As assignment issues have not yet received much attention in the literature, this study will focus on designing drone assignment strategies through optimization. The optimization aims at minimizing charging station installation costs, drone energy consumption, and operational costs. The aim of this work is to design a model to determine the optimal number of the drone hubs, along with their configuration. Moreover, we will determine their location and size, allocating the customer demands to stations and dimensioning the drones¿ fleet in each station to deliver packages efficiently.This work has been partially supported by the Spanish Ministry of Science, Innovation, and Universities (PID2022-140278NB-I00 project and RED2022-134703-T network). Additionally, we acknowledge the support from the Public University of Navarre for Young Researchers Projects Program (PJUPNA26-2022) and the UNED Pamplona (UNEDPAM/PI/ PR24/04P project)
Neural controller based on reinforcement learning for takeoff and landing of drones in environments with variable wind
No full textEl control autónomo de drones en condiciones de viento variable representa un desafío crucial en los campos de la aeronáutica y la robótica. En esta tesis, se presenta el diseño y la evaluación de un controlador neuronal basado en aprendizaje por refuerzo (RL), orientado a optimizar la maniobrabilidad autónoma durante las fases críticas de despegue y aterrizaje en entornos complejos. El objetivo principal es superar las limitaciones de los controladores PID tradicionales, mejorando la estabilidad y la precisión del vuelo. Para validar estos avances, se realizarán pruebas exhaustivas mediante simulaciones Hardware-in-the-Loop (HIL), estableciendo comparaciones detalladas con el desempeño de los controladores PID.
El aprendizaje por refuerzo (RL - Reinforcement Learning) ha emergido como una solución innovadora para sistemas complejos, permitiendo a los agentes desarrollar políticas óptimas de control a partir de la interacción directa con su entorno, sin requerir modelos precisos del sistema. Este enfoque se destaca por su adaptabilidad y su capacidad para gestionar no linealidades en la dinámica de vuelo de los drones, superando así limitaciones de los métodos convencionales. En este trabajo, el RL se implementa progresivamente en controladores neuronales profundos: desde algoritmos en espacios de acción discretos como Deep Q-Network (DQN) hasta soluciones definitivas en entornos continuos mediante Deep Deterministic Policy Gradient (DDPG) y Proximal Policy Optimization (PPO), integrando simulaciones de entornos realistas que modelan dispositivos y fuerzas externas, incluyendo efectos de viento.
Una de las contribuciones clave es el desarrollo de una arquitectura de red neuronal con un Módulo de Adaptación y un Módulo de Conversión, que transforman las fuerzas y momentos en velocidades de motor. Esta innovación permite al controlador neuronal responder a ráfagas de viento de hasta 10 m/s, optimizando a su vez la previsibilidad y confiabilidad mediante una discretización de acciones, lo cual reduce tanto la cantidad de acciones necesarias como el error de posición durante maniobras.
Los resultados de las pruebas muestran mejoras notables en estabilidad y precisión de trayectoria, así como en la capacidad de respuesta ante variaciones de viento abruptas. Durante las pruebas, el controlador permitió al modelo 3DR Iris+ mantener la estabilidad en situaciones de viento de hasta 10 m/s (91%de su velocidad máxima) en maniobras de despegue y aterrizaje, obteniendo un rendimiento competitivo con drones avanzados en resistencia relativa. Las pruebas con Hardware-in-the-Loop (HIL) también validaron la eficacia del controlador en entornos físicos, comparándolo contra sistemas PID.
Los hallazgos indican que el enfoque propuesto no solo ofrece una solución robusta y eficiente para el control autónomo de drones, sino que además abre nuevas oportunidades para aplicaciones seguras en áreas críticas como vigilancia, rescate y logística. Esta tesis aporta significativamente al campo del control autónomo de UAVs, estableciendo una base sólida para futuros desarrollos en controladores adaptativos inteligentes y en el estudio del vuelo en condiciones ambientales adversas.The autonomous control of drones under variable wind conditions represents a critical challenge in the fields of aeronautics and robotics. This thesis presents the design and evaluation of a neural controller based on reinforcement learning (RL), aimed at optimizing autonomous maneuverability during critical phases of takeoff and landing in complex environments. The main goal is to overcome the limitations of traditional PID controllers, improving flight stability and precision. To validate these advances, exhaustive testing will be conducted through Hardware-in-the-Loop (HIL) simulations, establishing detailed comparisons with the performance of PID controllers.
Reinforcement learning has emerged as an innovative solution for complex systems, enabling agents to develop optimal control policies from direct interaction with their environment without requiring precise system models. This approach stands out for its adaptability and ability to handle nonlinearities in drone flight dynamics, overcoming the limitations of conventional methods. In this work, RL is progressively implemented in Deep neural controllers: from discrete-action space algorithms like Deep Q-Network (DQN) to definitive solutions in continuous environments using Deep Deterministic Policy Gradient (DDPG) and Proximal Policy Optimization (PPO), integrating realistic environment simulations that model devices and external forces, including wind effects.
One of the key contributions is the development of a neural network architecture with an Adaptation Module and a Conversion Module, which transform forces and moments into motor speeds. This innovation allows the neural controller to respond to wind gusts of up to 10 m/s, optimizing predictability and reliability through action discretization, reducing both the number of required actions and position errors during maneuvers.
The test results show notable improvements in stability and trajectory accuracy, as well as responsiveness to abrupt wind variations. During the tests, the controller enabled the 3DR Iris+ model to maintain stability in winds of up to 10 m/s (91% of its maximum speed) during takeoff and landing maneuvers, achieving competitive performance compared to advanced drones in relative resilience. Hardware-in-the-Loop (HIL) tests also validated the controller’s effectiveness in physical environments, comparing it to PID systems.
The findings indicate that the proposed approach not only provides a robust and efficient solution for autonomous drone control but also opens new opportunities for safe applications in critical areas such as surveillance, rescue, and logistics. This thesis makes a significant contribution to the field of UAV autonomous control, establishing a solid foundation for future developments in intelligent adaptive controllers and the study of flight under adverse environmental conditions.Programa de Doctorado en Ciencias y Tecnologías Industriales (RD 99/2011)Industria Zientzietako eta Teknologietako Doktoretza Programa (ED 99/2011
Immigrant self-selection into incentive-based compensation schemes
No full textEste estudio analiza si los trabajadores inmigrantes tienen una mayor probabilidad de estar empleados bajo esquemas de compensación objetiva como una forma de evitar la posible discriminación. La hipótesis es que la remuneración por incentivos, cuando se basa en criterios objetivos y medibles, reduce las oportunidades de evaluaciones sesgadas y se vuelve más atractiva para los trabajadores que podrían anticipar un trato injusto. El análisis utiliza microdatos de la edición de 2015 de la Encuesta Europea de Condiciones de Trabajo (EWCS), que incluye información detallada sobre estructuras salariales, características del empleo y antecedentes demográficos en los países europeos. Se estima un modelo probit con efectos fijos por país, sector y ocupación para probar la relación entre la condición de inmigrante y la probabilidad de estar bajo esquemas de compensación objetiva. Los resultados muestran una asociación positiva y estadísticamente significativa. El efecto marginal promedio indica que ser inmigrante aumenta la probabilidad de recibir este tipo de compensación en 4.4 puntos porcentuales, lo que respalda la hipótesis de autoselección hacia sistemas de pago más transparentes.This study examines whether immigrant workers are more likely to be employed under
objective compensation schemes as a way to avoid potential discrimination. The
hypothesis is that incentive pay, when based on objective and measurable criteria, reduces
opportunities for biased evaluations and becomes more attractive to workers who may
anticipate unfair treatment. The analysis uses microdata from the 2015 wave of the
European Working Conditions Survey (EWCS), which includes detailed information on
pay structures, employment characteristics, and demographic background across
European countries. A probit model with fixed effects for country, sector and occupation
is estimated to test the relationship between immigrant status and the likelihood of being
under objective compensation schemes. The results show a positive and statistically
significant association. The average marginal effect indicates that being an immigrant
increases the probability of receiving such compensations by 4.4 percentage points,
supporting the hypothesis of self-selection into more transparent pay systems.Graduado o Graduada Internacional en Administración y Dirección de Empresas/Graduado o Graduada Internacional en Economía por la Universidad Pública de NavarraEnpresen Administrazio eta Zuzendaritzan Nazioarteko Graduatua / Ekonomian Nazioarteko Graduatua Nafarroako Unibertsitate Publikoa
Mediation of obesity-related variables in the association between physical fitness and cardiometabolic risk in children and adolescents: a systematic review and meta-analysis
No full textObjective. To examine the mediation of obesity-related variables in the association between physical fitness and cardiometabolic risk in children and adolescents.
Design. Systematic review and meta-analysis.
Data sources. Studies from electronic databases from inception to 31 December 2023.
Eligibility criteria for selecting studies. Included were 123 observational studies (cross-sectional and longitudinal) that assessed risk by constructing a continuous score incorporating cardiometabolic parameters. Studies were considered if they evaluated at least one fitness component as an exposure in children and adolescents (5–19 years). Thirty-one were included in the main meta-analyses.
Results. Cross-sectional findings indicate that cardiorespiratory fitness is modestly but beneficially associated with cardiometabolic risk, either indirectly via obesity-related variables (indirect standardized beta coefficient [βIndirect]=−0.17; 95% confidence interval [CI] −0.23; −0.11; inconsistency index [I2]=94.4%) or directly and independently from obesity-related variables (r=−0.11; 95% CI −0.15; −0.07; I2=87.4%), whereas muscular fitness seems to be associated with risk only via obesity-related variables (βIndirect=−0.34; 95% CI −0.47; −0.20; I2=85.1%). There was no cross-sectional difference between biological sexes (p≥0.199). Longitudinal findings indicate no total (r=−0.12; 95% CI −0.24; 0.01; I2=23.1%) and direct (r=−0.03; 95% CI −0.08; 0.03; I2=0%) associations.
Conclusion. The association between fitness and risk appears to take place either indirectly through the reduction of obesity-related levels or directly by influencing risk. The latter underscores that the inverse association extends beyond a mere reduction in obesity-related variables, encompassing specific enhancements linked to exercise training, including increased metabolic efficiency, and cardiovascular capacity.
PROSPERO registration number. CRD42022354628
The presence of adipose tissue in aortic valves influences inflammation and extracellular matrix composition in chronic aortic regurgitation
No full textAdipose tissue is present in aortic valves (AVs). Valve interstitial cells (VICs) could differentiate into adipogenic lineages. We here characterize whether the presence of adipose tissue in the AV influences inflammation and extracellular matrix (ECM) composition in patients with aortic regurgitation (AR). A total of 144 AVs were analyzed by histological and molecular techniques. We performed discovery studies using Olink Proteomics® technology in 40 AVs (N = 16 without and N = 24 with adipose tissue). In vitro, human white adipocytes (HWAs) or VICs were cultured with adipogenic media and co-cultured with control VICs. Of Avs, 67% presented white-like adipocytes within the spongiosa. Discovery studies revealed increased levels of inflammatory and ECM molecules in AVs containing adipocytes. Interestingly, the presence of adipocytes was associated with greater AV thickness, higher inflammation, and ECM remodeling, which was characterized by increased proinflammatory molecules, collagen, fibronectin, proteoglycans, and metalloproteinases. AV thickness positively correlated with markers of adipose tissue, inflammation, and ECM. In vitro, adipocyte-like VICs expressed higher levels of adipocyte markers, increased cytokines, fibronectin, decorin, and MMP-13. Analyses of supernatants from co-cultured control VICs with HWA or adipocyte-like VICs showed higher expression of inflammatory mediators, collagen type I, proteoglycans, and metalloproteinases. AVs presenting adipocytes were thicker and exhibited changes characterized by increased inflammation accompanied by aberrant expression of collagen, proteoglycans, and metalloproteinases. VICs could differentiate into adipogenic pathway, affect neighbor VICs, and contribute to inflammation, collagen and proteoglycan accumulation, as well as to metalloproteinases secretion. In summary, the presence of adipose tissue in AV could modify its composition, favoring inflammation and remodeling with an impact on AV thickness.This work was supported by Fondo de Investigaciones Sanitarias [PI18/01875; PI21/00280] from Instituto de Salud Carlos III-FEDER. M.G. is supported by a Miguel Servet Foundation PhD studentship, and E.M.-N. is supported by a Margarita Salas
Multi antenna structure assisted by metasurface concept providing circular polarization for 5G millimeter wave applications
No full textThis paper presents a circularly polarized multi-antenna structure designed for 5G millimeter-wave applications. The structure is based on circular patch radiators, each enhanced with metasurface (MTS) characteristics through the integration of multi-split ring slots. Each radiating element is enclosed within a decoupling wall constructed from a microstrip transmission line, which features both wide (capacitive) and thin (inductive) impedance profiles. The antennas are excited from below using metallic pins, which connect to the radiators through via-holes stemming from coplanar waveguide ports on the ground plane. Experimental results demonstrate a wide bandwidth from 25.6 to 29.7 GHz, corresponding to a fractional bandwidth of 14.82%. Additionally, the antenna exhibits stable radiation patterns, with an average gain of 2.7 dBi and a radiation efficiency of 57%. Using a single radiator configuration, a 3 × 3 antenna array was implemented. In this design, electromagnetic coupling between adjacent radiators is significantly reduced. The resulting array, measuring 20 × 20 × 0.32 mm3, achieves excellent performance across a wide frequency range from 24 to 31 GHz, corresponding to a bandwidth of 25.45%. Key metrics include an average isolation between radiating elements exceeding 17 dB and an average gain and radiation efficiency of 9.0 dBi and 91.5%, respectively.This work was funded by the Deanship of Scientific Research at Jouf University under Grant Number (DSR2022-RG-0110)
Resonance-based optical gas sensors
No full textGas sensors play a critical role in numerous human activities. Their necessity continues to grow across diverse fields as technological advancements drive demand for precision agriculture and bioengineering among other applications. Among existing sensor technologies, optical gas sensors stand out due to their ability to operate remotely in high-risk environments while remaining unaffected by electromagnetic interference. Resonance-based optical sensors offer targeted gas detection through the functionalization of their sensitive surfaces. This work focuses on reviewing the state of the art in resonance-based optical gas sensors (ROGSs), addressing their fundamental principles, recent advances in fabrication processes, waveguide designs, and materials employed both for resonance generation and as sensitive coatings. In addition, the review examines achieved sensitivity, emerging applications, and key developments in the field, including those efforts on improving ROGS performances by means of artificial intelligence techniques. The study encompasses optical sensors leveraging surface plasmon resonance, lossy mode resonance, and hyperbolic mode resonance¿the latter representing a notable breakthrough in recent years as a particular case of Bloch surface waves.This work was supported in part by the Agencia Estatal de Investigación under Grant PID2022-137437OB-I00 and Grant PDC2023-145831-I00, in part by the Institute Smart Cities and Public University of Navarra Ph.D. student grant, and in part by the International Mobility Scholarship of the Navarra Government