Doctoral School

Reading date: 19/02/2026

• MEDRANO DÍAZ, MANUEL ALEJANDRO: Estimación de series de tiempo de imágenes mediante técnicas de aprendizaje profundo
  
  Author: MEDRANO DÍAZ, MANUEL ALEJANDRO
  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: 20/01/2026
  Reading date: 19/02/2026
  Reading time: 16:00
  Reading place: Aula Maestría de Ciencias de la Computación, Instituto Tecnológico de Culiacán, MéxicoEnlace Videoconferencia: https://meet.google.com/cya-jyje-zeq
  Thesis director: PUIG CAYUELA, VICENÇ | RODRÍGUEZ RANGEL, HÉCTOR
  Thesis abstract: An image time series (ITS) is a chronologically ordered sequence of images showing the spatial change of its elements over time. Satellite images of meteorological events can be treated as ITS by displaying their values through pixel color intensity.Estimating from a ITS with a deep learning model is a complex problem that requires analyzing various configurations that determine how images are processed. It is a computationally intensive problem, which, due to its non-deterministic characteristics, requires combining different parameter configurations to extract spatio-temporal relationships. Thus, the complexity of the problem increases as the dimensions to be evaluated increase.To solve this problem, a robust and scalable conceptual model for ITS estimation is proposed that extracts the spatiotemporal relationships between pixels and their neighborhoods. Based on the specifications of the proposed model, a methodological proposal is developed that allows the estimation of meteorological maps using deep learning models. The proposed methodology is implemented through the design of a software architecture that translates abstract elements into software components, allowing the methodology to be evaluated through the use of deep learning models in different case studies.In the case study of the United States (US) drought monitor, experimentation with deep learning models based on ConvLSTM and Multi-CNN mostly yielded an F1-score of over 0.90 for the estimation of step t+1, with the best model obtaining an F1-score of 0.9953. Due to the high memory demand of the data dimensions, together with the physical limitations of the hardware equipment, dimension reduction techniques were applied to the images. Using the fragmentation technique with the ConvLSTM architecture, an F1-score of 0.9684 was obtained by reducing the dimension of the samples by 48%. By applying recursive and direct multi-step estimation strategies, medium-term estimates could be made. However, due to the complexity of the spatiotemporal analysis, there is an accumulated error that affects the quality of the medium-term estimate.In a second case study on US standardized precipitation index (SPI) maps, a ConvLSTM architecture is used to estimate step t+1. The results show that the best learning model obtains an F1-score of 0.5268, while the Naïve model obtains an F1-score of 0.3408. The results demonstrate the capabilities of deep learning models to extract spatiotemporal relationships from a sequence of images, laying the foundation for a branch of research focused on image estimation.
  

• RODRÍGUEZ ROMERO, CARLOS EDUARDO: Analysis of coupled hydro-mechanical processes in double-structure geomaterials for nuclear waste storage
  
  Author: RODRÍGUEZ ROMERO, CARLOS EDUARDO
  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: Department of Civil and Environmental Engineering (DECA)
  Mode: Normal
  Deposit date: 04/12/2025
  Reading date: 19/02/2026
  Reading time: 11:00
  Reading place: ETSECCPB. UPC, Campus NordBuilding C2. Classroom: 212C/Jordi Girona, 1-308034 Barcelona
  Thesis director: VAUNAT, JEAN | GENS SOLE, ANTONIO
  Thesis abstract: The safe long-term isolation of high-level radioactive waste requires engineered barriers capable of maintaining low permeability and mechanical stability under complex thermo-hydro-mechanical (THM) conditions. Among candidate materials, compacted bentonite exhibits a distinctive double-structure behaviour, governed by the coexistence of micro- and macro-porous domains. This thesis focuses on the analysis of coupled hydro-mechanical processes in double-structure geomaterials, with particular attention to bentonite mixtures of blocks and pellets, as used in buffer systems for deep geological repositories. The research first reviews the geomechanical basis of double-structure soils and identifies the experimental evidence supporting their dual-porosity nature. A constitutive THM framework is then developed, extending the existing double-structure formulation to incorporate: (i) the parameter ακ to control microstructural deformation; (ii) a fabric-dependent structuration law to represent the memory and degradation of compression; and (iii) frictional resistance at block–pellet and block–wall interfaces.The model was implemented and calibrated using laboratory and mock-up experiments from the BEACON project, including the MGR22, MGR23, and MGR27 experiments, the EPFL path-dependent tests and the POSIVA test. Numerical simulations successfully reproduced the evolution of swelling pressure, void ratio, dry density, water content and water intake observed experimentally. The results confirmed that friction plays a decisive role in the redistribution of stresses between pellets and blocks, while microstructural evolution governs the long-term homogenisation process. The enhanced formulation captured partial density homogenisation and the persistence of microstructural porosity, in agreement with laboratory observations.Overall, the thesis provides an improved understanding of the coupled hydro-mechanical behaviour of double-structure bentonites and proposes a robust constitutive framework capable of reproducing their key features under repository-relevant conditions. The work highlights the necessity of considering both microstructural evolution and frictional effects in predictive models for bentonite barriers, thus contributing to the reliability of long-term safety assessments of deep geological repositories.
  

• SAMADI GHARAJEH, MOHAMMAD: Mapping of Real-Time Computation to Parallel Platforms
  
  Author: SAMADI GHARAJEH, MOHAMMAD
  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 COMPUTER ARCHITECTURE
  Department: Department of Computer Architecture (DAC)
  Mode: Normal
  Deposit date: 20/01/2026
  Reading date: 19/02/2026
  Reading time: 11:00
  Reading place: C6-E101
  Thesis director: PINHO, LUIS MIGUEL | ROYUELA ALCÁZAR, SARA
  Thesis abstract: Time-critical systems are a special type of application that must complete computing tasks within specific time constraints to guarantee the required level of service. As these systems have become more complex in recent years, they have increased performance requirements, which necessitates the need to use more powerful hardware platforms, providing higher performance. The performance of these systems can be improved by utilizing multi-core processors and parallel programming models (e.g., OpenMP). However, the predictability and schedulability of these parallel applications easily become intricate due to the potentially complex structure of execution graphs and the several resources available in modern architectures. Therefore, mapping tasks to computing resources needs to be performed efficiently in these applications to improve the work-conserving of the mapping process and the load-balancing of the job queues. This process can lead to reduced application response time (and WCRT), as well as its variability.To address the problems mentioned above, this thesis tackles the challenge of maintaining system predictability and schedulability while maximizing performance in real-time parallel computing systems. Accordingly, it (i) proposes task-to-thread mapping methods based on heuristics that exploit knowledge about the behavior of predictable parallel applications, considering both the offline design of the system as well as the online execution phase, to reduce the variability of response times and the WCRT, as well as to improve the average response time in OpenMP applications, (ii) uses efficient methods for measuring parallel tasks in terms of schedulability and predictability as well as discovering the longest execution path through the parallel execution (i.e., WCRT), (iii) evaluates the impact of the configurations of mapping algorithms (e.g., static and dynamic mapping) and hardware platforms on the task execution time and application response time, and (iv) proposes and evaluates the performance of task-to-accelerator mapping using different heuristics to reduce the response time by executing high-workload tasks in the accelerator.Evaluation results, based on simulations and experiments using random graphs and real-world applications, demonstrate that the new approaches, in most cases, minimize response time (and WCRT) and reduce the variability of response times compared to existing mapping approaches. In addition, a prototype implementation of the main heuristics is evaluated using real-world applications, showing that the WCRT and response time variability obtained using the new methods are lower than those obtained using LLVM's default scheduler in most configurations. These achievements show that the proposed methods can improve the predictability and schedulability of real-time parallel applications.
  

Reading date: 20/02/2026

• BELSA CARNÉ, BLANCA MARIA: Engineering Catalyst-Ionomer Interfaces for Carbon-Efficient CO2 Electrolysis and Technology Prospects
  
  Author: BELSA CARNÉ, BLANCA MARIA
  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 PHOTONICS
  Department: Institute of Photonic Sciences (ICFO)
  Mode: Normal
  Deposit date: 15/01/2026
  Reading date: 20/02/2026
  Reading time: 10:30
  Reading place: ICFO Auditorium
  Thesis director: GARCÍA DE ARQUER, FRANCISCO PELAYO
  Thesis abstract: The electrochemical reduction of CO2 (CO2E) offers a promising route to convert greenhouse gas emissions into value-added chemicals and fuels. However, achieving performance metrics that enable the technoeconomic and sustainable viability of CO2E remains challenging. This is especially acute in the case of multicarbon products (C2+), important precursors for energy fuels and manufacturing, where achieving combined selectivity and carbon utilisation under industrially relevant conditions is challenged by undesired competing reactions. This thesis explores the design and implementation of new strategies to modulate electrochemical interfaces in CO2E to overcome this barrier. These are based on the implementation of ionomer coatings that specifically address key reactants and intermediates in CO2E. A key contribution is the development and mechanistic elucidation of ion management channels (IMCs), formed by co-distributing cation and anion exchange ionomers (CEIs and AEIs) within the catalyst layer. This architecture enables local regulation of hydroxide and cation populations, mitigating *OH poisoning and enhancing *CO adsorption, critical steps for promoting C–C coupling and C2+ product formation.The ionomer–catalyst interface is comprehensively characterised using SEM–EDS, FTIR, XPS, KPFM, contact angle measurements, cyclic voltammetry, and EIS. In situ Raman spectroscopy reveals the dynamic evolution of surface species, confirming that excessive *OH accumulation suppresses C2+ selectivity, while IMCs restore favourable interfacial conditions. These insights are correlated with improved electrochemical performance, carbon efficiency, and stability across a wide range of operating conditions, including highly acidic environments.The IMC concept is further implemented in membrane electrode assembly (MEA) devices operating under neutral pH. Preliminary results demonstrate improved performance and reduced cell voltages for IMC-based electrodes, indicating compatibility with scalable reactor platforms and commercially viable components.The thesis concludes with a broader analysis of the challenges facing CO2E at scale. Key bottlenecks, such as the reliance on iridium anodes and fluorinated membranes, are critically assessed, and material and performance targets for gigaton-scale deployment are proposed. A techno-economic and life-cycle analysis outlines trade-off between performance, cost, and sustainability, while global scaling efforts are reviewed. Benchmarking protocols are proposed to bridge the gap between laboratory research and industrial implementation.Together, this work advances a cohesive framework for interfacial engineering in CO2E, linking molecular-level understanding to device-scale integration, and providing pathways toward industrial deployment.
  

• CUELLAR SANTIAGO, FATIMA: Optical and visual quality of presbyopia-correcting intraocular lenses
  
  Author: CUELLAR SANTIAGO, FATIMA
  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 OPTICAL ENGINEERING
  Department: Department of Optics and Optometry (OO)
  Mode: Article-based thesis
  Deposit date: 22/12/2025
  Reading date: 20/02/2026
  Reading time: 11:30
  Reading place: Auditori Joan Salvadó del Ce (CUV) Facultat d'Òptica i Optometria de TerrassaAvgda. 22 de Juliol, 660. 08222 Terrassameet.google.com/wbu-qxqi-hkj
  Thesis director: MILLAN GARCIA VARELA, MARIA SAGRARIO
  Thesis abstract: The development of new presbyopia-correcting intraocular lens (IOL) designs, with their manufacturing and launching on the market makes optical characterization highly recommendable, allowing for a more objective and quantitative understanding of their properties before implantation, thus avoiding the influence of subjective and individual factors. International regulations require the evaluation of IOL optical and visual quality in both laboratory and clinical settings.This investigation focuses on the interaction of new presbyopia-correcting IOLs designs with ocular optical aberrations (in normal and astigmatic corneas), the effect of IOL decentration and tilt on image quality, and the potential risk of unwanted optical phenomena. In all these cases, studying the optical behavior of a set of diffractive trifocals, extended-depth-of focus (EDOF), and enhanced monofocal (EM) IOLs under controlled in vitro conditions serves as a valuable complement to clinical studies. The optical quality of IOLs is evaluated through preclinical metrics based on the modulation transfer function (MTF), the estimation of postoperative visual acuity, range of vision and halo. Characterization was conducted using an optical bench equipped with a model eye consisting of a saline-filled wet cell in which the commercial presbyopia-correcting IOLs under study are immersed one by one. Adaptive optics was used to introduce controlled amounts of corneal aberration. This configuration allows for an objective, patient-independent assessment and enables control over factors such as pupil size, corneal aberrations, and lens alignment, aspects difficult to control in clinical practice. The experimental setups approximately reproduced the conditions under which the lens is implanted in the eye, in accordance with current international standards (ISO 11979-2-2024 and ANSI Z80.35-2018). An upgrade of the optical setup to allow for depth of field scanning in the object space has been initiated with the characterization of tunable focus lenses (FTL).To complement and validate the characterization of intraocular lenses (IOLs) in the optical bench, this work also includes studying visual quality using the SimVis Gekko visual simulator, which allows dynamic reproduction of vision with different intraocular lens designs before the cataract surgery. We also provide clinical results, obtained through collaborative research, that demonstrate excellent agreement with the laboratory findings. These data include visual assessments (with and without astigmatism) of pseudophakic patients implanted with the same presbyopia-correcting IOLs tested in the optical laboratory.