Author: GIULIODORI PICCO, AGUSTINA
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN CIVIL ENGINEERING
Department: (DECA)
Mode: Article-based thesis
Deposit date: 06/11/2023
Reading date: 30/11/2023
Reading time: 11:00
Reading place: Sala OCZ-CIMNE. CIMNE - Edifici C1 Campus Nord UPC C/ Gran Capità, S/N 08034 Barcelona
Thesis director: SOUDAH PRIETO, EDUARDO | HERNANDEZ ORTEGA, JOAQUIN ALBERTO
Committee:
     PRESIDENT: CONTI, MICHELE
     SECRETARI: OTERO GRUER, FERMÍN ENRIQUE
     VOCAL: MARTORELL LOPEZ, JORDI
Thesis abstract: The advancements in material science and innovative manufacturing techniques have created substantial possibilities for improving the functionality of medical devices, tailoring them to specific needs. However, the increased complexity in terms of geometry and materials underscores the growing demand for predictive modeling tools that can accurately evaluate the mechanical behavior of such devices. To address this issue, this thesis proposes a robust multiscale methodology capable of addressing intricate structural scenarios that would typically be challenging to analyze using traditional methods. We strove to develop a computational tool that effectively incorporates fine-scale localized effects within a coarse-scale model, enabling the assessment of the mechanical behavior of complex structures while keeping low computational costs. Through the examination of the specific case of covered braided stents, we illustrate how traditional methods are not suitable for the analysis of such devices. While some analytical formulations and additional approaches can be developed to improve the results, we advocate for a shift away from the development of these ad-hoc solutions in favor of a robust and general approach that can be applied to a wide range of medical devices. In this context, we propose a multiscale approach that can be regarded as a special type of Finite Element (FE) technology, which leverages machine learning to construct a model that incorporates features learned from (pre-computed) full-order models. This implies that our method substitutes certain polynomial-based FE matrices with `empirical' operators derived from data obtained in previous computational experiments. We demonstrate throughout this thesis that our approach leads to an accurate representation of fine-scale heterogeneities and nonlinearities within the coarse-scale model, achieved by a proper combination of domain decomposition and reduced-order modeling techniques.

Author: PRADES VALLS, ALBERT
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN GEOTECHNICAL ENGINEERING
Department: (DECA)
Mode: Article-based thesis
Deposit date: 15/09/2023
Reading date: 30/11/2023
Reading time: 11:00
Reading place: Sala de Graus, Av Doctor Marañón 44-50, Escola Politècnica Superior d'Edificació de Barcelona Campus SUD.
Thesis director: LANTADA ZARZOSA, MARIA DE LAS NIEVES | COROMINAS DULCET, JORDI
Committee:
     PRESIDENT: PRAT CATALAN, PERE
     SECRETARI: BLANCH GÓRRIZ, XABIER
     VOCAL: COPONS LLORENS, RAMON
Thesis abstract: The term rockfall applies to blocks or masses of rock that, after separating from a steep wall, experience free fall and thenimpact against the ground, with the consequent rebound and possible rolling and sliding of rock fragments. Thisphenomenon has great destructive potential due to the high velocities, and thus high kinetic energy, that blocks may reachduring their propagation and may threaten buildings, civil infrastructure and transportation networks. The phenomenonbecomes more complex if fragmentation during rockfall propagation is considered. The present thesis aims to contribute toknowledge of the rockfall process by using geomatic techniques and developing specific models that may contribute to themonitoring of rock cliffs.Monitoring plays an important role in managing risk due to rockfalls in a wide spectrum of applications. Terrestrial or aeriallaser scanning (TLS or ALS) and photogrammetry are geomatic techniques that are frequently used to estimate thedetachment volume over time. These observations allow the expected magnitude frequency of the cliff to be estimated, whichis required to quantitatively evaluate risk. In this thesis, an automated and low-cost continuous time-lapse fixed positioncamera system was developed and implemented to monitor a basaltic cliff. The aim was to test its ability to detect precursoryrockfall movements and quantify volume changes if the rock mass detachment occurs. A methodology was developed tohandle a set of images at different times to detect possible precursory movements. The procedure consists in matching theimages with a reference master image after filtering the vegetation and correcting illumination differences and small cameramovements. The fitted images are used for a forward-backward correlation process that is carried out to detect deformation. Ifdisplacement is detected, a 3D model is built to quantify it. The image process system can detect small precursormovements of around a centimeter.Several real-scale tests were performed in a quarry to analyze the trajectories of falling rock blocks and their fragmentation,which were recorded by three high-speed video cameras. In this thesis, a code was specifically developed to track the 3Dtrajectory of blocks and fragments and to measure velocities before and after the impact. Two different programs wereimplemented, one for the blocks that break and one for those that do not. The captured parameters allow the velocitydistribution of fragments, the restitution coefficients, and the energy balance of the impact to be obtained. The results of thesetests were incorporated into the rockfall propagation model RockGIS.A final issue that is addressed is the characterization of roughness, which is a fundamental parameter of the propagation ofblocks on slopes made up of accumulations of fragments (e.g. talus deposits). One of the limitations of considering a lumpedmass approach when simulating block propagation is that its relative position at impact with the slope is unknown, as thegeometry is not explicitly taken into account. In this thesis, different algorithms are developed to quantify volumes of rockfragments and to define the spatially distributed roughness of the terrain using high resolution 3D point clouds collected withTLS or photogrammetry.

Author: PÉREZ PÉREZ, ESVAN DE JESÚS
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN AUTOMATIC CONTROL, ROBOTICS AND VISION
Department: Department of Automatic Control (ESAII)
Mode: Change of supervisor
Deposit date: 29/09/2023
Reading date: 30/11/2023
Reading time: 16:00
Reading place: Sala conferencias Joaquín Montes Edificio Z, México
Thesis director: PUIG CAYUELA, VICENÇ | LÓPEZ ESTRADA, FRANCISCO RONAY
Committee:
     PRESIDENT: ODGAARD, PETER FOGH
     SECRETARI: NEJJARI AKHI-ELARAB, FATIHA
     VOCAL: REYES REYES, JUAN
Thesis abstract: In recent years, wind power has emerged as a promising alternative to fossil fuels, with wind turbines widely used to generate clean energy. However, like complex machinery, wind turbines are prone to faults that can lead to costly downtime and maintenance expenses. To address this challenge, this thesis has proposed a comprehensive approach for fault diagnosis in wind turbines using machine learning, structural analysis, and interval and zonotope observers. These techniques complement each other to generate a hybrid fault diagnosis method.Structural analysis has provided valuable information on the redundant relationships between the components and measurable variables of the wind turbine. These redundant relationships are identified with an adaptive neurofuzzy inference system (ANFIS) and its improved multi-output (MANFIS) version. The training of neurofuzzy systems is based on completely fault-free data, unlike other methods that require learning the different possible faults. As a result of the identification with ANFIS and MANFIS, a set of Takagi Sugeno convex models is obtained. Interval and zonotopic observers are designed for fault detection from the convex models. The isolation is produced by evaluating the fault signature matrix under different metrics.To evaluate the effectiveness of the proposed methods, they were evaluated under different fault scenarios and compared with other diagnostic approaches reported in the literature.

Author: AGUILAR VILLAJOS, JAVIER
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN BUSINESS ADMINISTRATION AND MANAGEMENT
Department: Department of Management (OE)
Mode: Normal
Deposit date: 20/09/2023
Reading date: 01/12/2023
Reading time: 12:00
Reading place: Sala de Reunions, Planta 0 de l'edifici C5 - Campus Nord - UPC
Thesis director: PONS PEREGORT, OLGA
Committee:
     PRESIDENT: GASCÓ GASCÓ, JOSE LUIS
     SECRETARI: SUNYER TORRENTS, ALBERT
     VOCAL: GONZALEZ RAMIREZ, Mª DE LOS REYES
Thesis abstract: This research project is to discuss. contrast and demonstrate the need to influence the training and development of emotional intelligence competences in the civil engineering profile. as well as to propase a model for the assimilation and integration of competences within the framework of management functions and to analyse the relationship between the emotional intelligence of civil engineers in Spain and their leve! of professional satisfaction.Based on documentary research and semi-structured interviews with civil engineering professionals. the emotional intelligence competencies are identified and grouped according to the dimensions of Golemans model (1998). Subsequently. a questionnaire adapted from Bar-On's (1997) measurement instrument, the "Emotional Quotient lnventory" (EQ-i). is developed.The scope of the study focused on civil engineering professionals who are members of the association of civil engineers and technical engineers of public works throughout Spain.The methodological approach was quantitative and exploratory, with descriptive and correlational analyses. with surveys of 220 civil engineers. The main results show that there is a positive and significant relationship between emotional intelligence and professional satisfaction of the respondents, and that there are significant differences in their professional satisfaction according to the years of experience.Based on both the bibliographical study and the results obtained in the questionnaires. the relevance of El competences for the development of Civil Engineering is sustained. both for the people who practise it and for the organisations where they work and the social systems they impact with their work.

Author: CORTES DUARTE, MARIA CAROLINA
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN SUSTAINABILITY
Department: University Research Institute for Sustainability Science and Technology (IS.UPC)
Mode: Normal
Deposit date: 29/09/2023
Reading date: 01/12/2023
Reading time: 16:00
Reading place: Aula TR5 2.05 en la ESEIAAT, Campus de Terrassa
Thesis director: CREMADES OLIVER, LAZARO VICENTE | GÓMEZ YEPES, MILENA ELIZABETH
Committee:
     PRESIDENT: TEJADA TEJADA, MACARENA
     SECRETARI: AMANTE GARCIA, BEATRIZ
     VOCAL NO PRESENCIAL: YEZID ROJAS, NÉSTOR
Thesis abstract: Environmental inequality metrics require complex analyses between social, economic and environmental variables at different scales. The current research develops metrics within this framework to determine correlations between socio-environmental position and air quality in highly densified areas, focusing on the urban environment of the city of Bogota, Colombia (South America). In addition to the variables related to public health, there are other variables that have also been proposed to measure environmental inequality, such as population density by UPZ, population of working age, health institutions, active pensioners, per capita income, companies created, demography of the victim population by locality, food voucher, total enrollment in official schools and finally, rate of school dropout in official schools, increasing the contributions to the concept of environmental justice framework. .During the research, tools were provided to build a methodological proposal that promotes the well-being and life quality of Bogota residents, integrating socio-environmental factors to the current socioeconomic status, among them, exposure to particulate matter.