Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

CLARA: Semi-automatic Retraining System

  • Campos-Mocholí, Mario|||0009-0009-4721-2207
  • Chacón-Albero, Oriol de la Pau
  • Julian, Vicente|||0000-0002-2743-6037
  • Botti V.|||0000-0002-6507-2756
  • Marco-Detchart, Cédric
  • Rincón-Arango, Jaime Andrés
[EN] Recent advancements in computer vision have led to humanlevel performance in image recognition tasks, but challenges persist in real-world applications due to differences in data distribution. This paper introduces CLARA, a semi-automatic framework designed to continually retrain computer vision models by incorporating expert annotation of new data. The system was tested in a production environment with a waste image classification model and showed improved performance and robustness against evolving data distributions. This approach encourages collaboration among stakeholders in waste management and utilizes human-in-the-loop strategies to enhance model accuracy. The framework is adaptable to various models and tasks and, additionally, supports fusion techniques that combine outputs from multiple models to improve prediction robustness and overall system reliability., This work was partially supported with grant tPID2021- 123673OB-C31 funded by MCIN/AEI/10.13039/501100011033 and by ERDF A way of making Europe , grant CIPROM/2021/077 from the Conselleria de Innovació, Universitats, Ciencias i Societat Digital GVA - PROMETEO, grant PAID-06-23 by the Vice Rectorate Office for Research from UPV and project Reciclaje Inteligente y Cooperativo en Toda la Cadena de Valor Orientado a Una Sociedad Sostenible 360º (INNEST/2022/180) from AVI.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Post-fire Behaviour of Stainless Steel Rebars

  • Hernández-Figueirido, David
  • Piquer, Ana
  • Roig-Flores, Marta
  • Lapuebla-Ferri, Andrés|||0000-0001-9667-236X
  • Albero, V|||0000-0001-7193-9232
  • Pons Aliaga, David|||0000-0002-9034-9550
  • Ibáñez, Carmen|||0000-0002-9354-5637
[EN] Stainless steel rebars (SSR) offer significant advantages for reinforced concrete structures, especially in environments where corrosion is a major concern. Also, SSR exhibit excellent mechanical strength, high ductility, and reliable performance at elevated temperatures. However, SSR are less commonly used than carbon steel rebars (CSR) mainly due to their higher cost. This cost factor likely contributes to the limited research on the mechanical properties of SSR after exposure to high temperatures. In the present study, an experimental campaign was conducted on duplex SSR of various grades and production routes (hot-rolled and cold-formed) to evaluate their response following exposure to elevated temperatures. Specimens were heated to temperatures up to 1000 ºC and posteriorly cooled in air to room temperature. Afterwards, specimens underwent tensile testing to obtain the residual mechanical properties, including yield strength, ultimate strength, Young¿s modulus, ultimate strain and strain at fracture. The residual values were compared with results obtained in similar tests conducted on CSR. Findings indicate that SSR, after cooling down, fully recover their yield and ultimate strengths and exhibit a greater residual ductility compared to CSR, even in cold-formed specimens., The authors gratefully acknowledge Universitat Politècnica de València for the help provided through the Project STAINPOST (financial grant for first research projects, reference PAID-06-23) and the company ACERINOX for providing the test specimens.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Towards a Low-Rank Approach to Compress Deep Neural Networks

  • Liern-Garcia, Marina
  • López-García, Aaron
  • Marco-Detchart, Cédric
  • Carrascosa Casamayor, Carlos|||0000-0003-3649-6530
[EN] In this work, we introduce a low-rank approach based on the truncated Singular Value Decomposition (SVD) technique to make deep neural networks (DNNs) smaller. Our method focuses on reducing the overall physical size of a neural network model without losing much in terms of its accuracy or increasing its error noticeably. Specifically, we applied our technique to a modified MobileNetV2 model, in the context of automatic plant diseases detection. We chose two of the biggest layers of the model and compressed them. Our goal was to see how this compression affects the breakdown of the layers, the error in rebuilding the layers, and how well the modified model performs. The results showed that the smaller model still predicts very accurately, even with the reduced physical size of its layers. By making these layers smaller, our approach offers a practical way to handle the deployment of large neural networks, especially in devices with limited resources, without compromising their effectiveness., This work was partially supported with grants TED2021-131295B-C32, PID2021-123673OB-C31 funded by MCIN/AEI/ 10.13039/ 501100011033 and by ERDF A way of making Europe , PROMETEO grant CIPROM/2021/077 from the Conselleria de Innovación, Universidades, Ciencia y Sociedad Digital - Generalitat Valenciana, Early Research Project grant PAID-06-23 by the Vice Rectorate Office for Research from Universitat Politècnica de València (UPV)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Rammed Earth as an Alternative to Conventional Construction Methods in Valencia: a Comparative Life Cycle Assessment Perspective

  • Quintana-Gallardo, Alberto|||0000-0002-8911-595X
  • Sirera-Cocera, Arturo
  • Zamora-Marina, Eduardo|||0009-0008-8828-4242
  • Sáez, Vanesa
  • Mendiguchia-Fontes, Fernando Aitor
  • Guillén Guillamón, Ignacio Enrique|||0000-0001-5546-1486
[EN] This paper uses a single-family housing model as a case study to evaluate the environmental and thermal performance of rammed earth construction compared to conventional building materials. This study aims to determine which construction method presents a lower environmental impact while maintaining comparable thermal performance. The research begins by defining the physical properties of the earth material used for rammed earth walls through an extensive literature review. Once characterized, construction solutions for the foundation, walls, and roof are proposed. These solutions are validated using thermal transmittance, condensation, and thermal lag analyses, following UNE EN ISO 13786:2000 standards. A single-family housing model is then developed considering local climatic recommendations provided by the Climate Consultant tool. Its thermal performance is assessed by performing a steady-state energy simulation across three distinct climate zones in the Valencian Community: warm and humid (Valencia), cold and dry (Comarca Esport, Utiel-Requena), and warm and dry (Alicante). A comparable model built with conventional materials is also analyzed and adjusted to achieve the same thermal transmittance. The environmental impact is assessed through a Life Cycle Assessment of the A (production) and B (use) phases for both construction methods. This comparative analysis provides insights into the environmental trade-offs between traditional and conventional construction techniques. By quantifying the environmental impacts and thermal performance, this study contributes to the growing discourse on sustainable construction and highlights the potential of rammed earth as an environmentally favorable alternative., The authors gratefully thank the Vicerrectorado de Investigación of the Polytechnic University of Valencia for funding the project “Parametrización de impactos sociales en vivienda: Análisis de Ciclo de Vida Social” (PAID-06-23).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Approximation to the Environmental Impact of the House of Gongora, Cehegín (Murcia, Spain): an LCA Perspective

  • Quintana-Gallardo, Alberto|||0000-0002-8911-595X
  • Sandoval-Gómez, Francisco
  • Celina-Sáez, Vanesa
  • Zamora-Marina, Eduardo|||0009-0008-8828-4242
  • Secondo Maglia, Juan Gabriel|||0009-0005-0904-4311
  • Mendiguchia-Fontes, Fernando Aitor
  • Guillén Guillamón, Ignacio Enrique|||0000-0001-5546-1486
[EN] This paper deals with an approximation to the environmental impacts of constructing the historic House of Góngora in Cehegín (Murcia), originally built in the 16th century using traditional earth architecture. By analyzing historical construction methods, materials, and energy use, the study approximates the environmental footprint of the house during its original construction. It then compares this with the estimated impacts of building an equivalent structure today, which would imply changes in techniques, methods and regulations. The comparison considers key environmental indicators such as carbon emissions, resource extraction, obtained using the Environmental Footprint LCA methodology. The paper highlights the sustainability inherent in traditional earth architecture and contrasts it with the challenges and benefits of contemporary construction practices. The findings question the idea of sustainability as a contemporary issue and reveal insights into the ecological trade-offs between historical and modern construction methods, emphasizing the potential for re-integrating sustainable practices from the past into modern architecture. This study contributes to ongoing discussions on sustainable construction and the relevance of traditional methods in reducing environmental impact. , Authors AQG, IGG, FAM and FSG gratefully thank the Spanish Ministry of Science, Innovation and Universities for funding the project Earth4Future “Sustainable reuse of earthen architecture and its lessons for contemporary architecture” (PID2022139154OB-I00) under the Proyectos de Generación de Conocimiento 2022 call. Authors AQG and IGG gratefully thank the Vicerrectorado de Investigación of the Polytechnic University of Valencia for funding the project “Parametrización de impactos sociales en vivienda: Análisis de Ciclo de Vida Social” (PAID-06-23).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Parametrizing social performance of new and existing housing projects through Social Life Cycle Assessment

  • Quintana-Gallardo, Alberto|||0000-0002-8911-595X
[EN] In the construction industry, decisions made at the architectural design level significantly impact the life of the building's users and the surrounding society. Aspects such as the effect of the building envelope on the thermal comfort of the users, the indoor air quality, the distribution of spaces, and the choice of materials and finishes are often studied without considering the effect these decisions have on the users throughout the life of the building. These aspects are often studied theoretically and relegated to a secondary role because of the difficulty of considering them holistically and objectively. This study's objective is to create a methodology that allows us to know how decisions at the project level affect the users prior to its construction. This is achieved by adapting the Social Life Cycle Assessment (S-LCA) methodology. S-LCA is a methodology used to objectively measure human actions' effects on society and the environment. The methodology is based on developing and defining a series of indicators that allow the adaptation of the Social Life Cycle Assessment (S-LCA) methodology to the architecture and construction sector. The methodology builds on the United Nations Environmental Programme guidelines and the Level(s) Framework. The indicators are chosen to analyze the vulnerability, well-being, habitability, and comfort associated with housing. The result is a series of social indicators that will be subsequently implemented in developing a holistic methodology of Social Life Cycle Assessment in the field of architecture., The authors gratefully thank the Vicerrectorado de Investigación of the Polytechnic University of Valencia for funding the project “Parametrización de impactos sociales en vivienda: Análisis de Ciclo de Vida Social” (PAID06-23). Author AQG gratefully acknowledges receiving funding from the Spanish Ministry of Universities and the Polytechnic University of Valencia, under the “Plan de Recuperación Transformación y Resiliencia” (investment funding from the European Union Next Generation EU), according to Royal Decree 289/2021, of April 20 (BOE of 22 April 2021).