Dipòsit Digital de la Universitat de Barcelona

Dades

Docència

Institucional

Mediateca

Programari

Recerca

Tesis Doctorals

Treballs de l'alumnat

El Dipòsit Digital de la Universitat de Barcelona és el repositori institucional que conté en format digital els materials derivats de l'activitat docent, investigadora i institucional de la comunitat universitària.

Enviaments recents

Desarrollo de un workflow integrado entre softwares de modelización de reservorios (MRST) y de simulación técnico-económica (GEOPHIRES) para la evaluación de sistemas geotérmicos con análisis de sensibilidad. Estudio sobre el modelo geotermal en 3D de la Fossa del Camp
(2026-06) Valdés Larraín, Diego José; xxx
The open-source evaluation of deep geothermal projects with industrial applications faces a structural gap between subsurface reservoir conditions and its surface financial analysis, which has traditionally been bridged through manual and inefficient tool coupling. This work develops, implements, and validates an automated workflow for the coupling of computational tools, orchestrated via Python. The workflow integrates the 3D thermo-hydraulic reservoir simulation of the MRST (MATLAB Reservoir Simulation Toolbox) software with the techno-economic evaluation engine provided by GEOPHIRES-X. This eliminates human intervention in data integration, transforming simulations into auditable, traceable, and reproducible numerical processes. The architecture of the developed workflow relies on advanced sequential control through on-disk data serialization and synchronous batch-mode calls. It incorporates an algorithmic bypass of GEOPHIRES-X, overriding its native homogeneous subsurface simplifications to enforce the direct assimilation of petrophysical data and CAPEX correction factors calculated in MRST. To validate the workflow and the necessity of this coupling, the pipeline was applied to a case study in the Jurassic aquifer of La Fossa del Camp (Tarragona, Spain), a carbonate basin with an estimated geothermal resource at a depth of 2,000–3,000 meters, with temperatures ranging from 70–90 °C and high permeabilities of up to 2,930 mD. This case evaluates a J-type directional doublet with a 1,200-meter well separation designed to supply industrial heat to the Mas Batller industrial park in Reus over a 30-year operational horizon. The main result is the quantification of the bias introduced by analytical models when applied to heterogeneous basins. The standalone GEOPHIRES analytical model overestimates the NPV by more than tenfold and underestimates the LCOH by 72% relative to the coupled workflow results. This divergence is caused by the software's inability to capture the real confinement of the thermal plume, alongside the petrophysical and geometric characteristics of the system. Consequently, the actual extracted thermal power is reduced by 39%, triggering cascading errors across other financial indicators. This discrepancy shifts the project's classification from highly profitable to marginally viable, a finding evidenced by the implemented parametric sensitivity analysis (OFAT). The analysis demonstrates that the reservoir's high transmissivity renders the local geology a non-limiting variable. This shifts the financial risk toward the operational efficiency of the surface plant, where the production flow rate and reinjection temperature emerge as parameters with the greatest leverage on the evaluated financial indicators. In conclusion, the developed workflow establishes itself as a transferable and open-source methodological contribution, remaining entirely reservoir-agnostic. Its primary value lies in demonstrating that the gap between physical simulation and techno-economic evaluation using open-source tools directly distorts the financial reliability of geothermal projects, mitigating risks in capital allocation.
Modelización y simulación del potencial de almacenamiento geológico de CO₂ con MATLAB Reservoir Simulation Toolbox (MRST): caso de estudio de la estructura Lopín
(2026-06) Quintero Tejada, Camilo Antonio; Gil Ortiz, Marc; Herms, Ignasi
This Final Master’s Thesis develops and implements a numerical model for geological CO₂ storage in the Lopín structure, located in the Ebro Basin, in the northeast of the Iberian Peninsula. The simulation is performed using a vertical-equilibrium implementation developed in MATLAB (MathWorks), within the open-source MATLAB Reservoir Simulation Toolbox (MRST) developed by SINTEF and based on the logic of its MRST/co2lab module. The model is built from geological and petrophysical data from the European PilotSTRATEGY project, and integrates the geometry of the Buntsandstein B1 reservoir, petrophysical properties, hydraulic boundary conditions and the main CO₂ retention mechanisms. A total of 15 cases are simulated, organized according to three dimensions of variation: the P10, P50 and P90 probabilistic realizations of the geological model, the injection-well configuration and the hydraulic boundary conditions. The reference scenario corresponds to the P50 realization, with two active wells and mixed boundary conditions. In this case, the model allows 8.46 Mt of CO₂ to be injected over 30 years while keeping the maximum reservoir pressure below the geomechanical safety thresholds defined for the site. The maximum injectable mass under the bottom-hole pressure criterion reaches 10.71 Mt in the reference scenario. The results show that lateral hydraulic connectivity is one of the factors with the greatest influence on the dynamic response of the system. Mixed boundary conditions, which allow lateral pressure dissipation, increase the maximum injectable mass by a factor of 1.6 to 1.9 compared with the closed model. Therefore, the main limiting factor is not the available pore volume, but the pressure evolution during injection. The trapping inventory analysis shows that, at the end of the simulated period, corresponding to 1,030 years, mobile free CO₂ under stratigraphic confinement represents approximately 77 % of the stored mass. Residual trapping accounts for around 18 %, structural trapping for approximately 4 %, and dissolution in brine for less than 1 %. These results indicate that most of the plume remains mobile at the simulated time scale, although it remains confined beneath the caprock and within the modelled domain. Overall, the results confirm the technical feasibility of the Lopín structure for a pilot phase of geological CO₂ storage. They also identify the hydraulic characterization of faults, validation with well data and the incorporation of geomechanical coupling as priorities for subsequent evaluation stages
Análisis del apagón ibérico desde una perspectiva ambiental
(2026-06) Piscitelli, Franco Martín; Daví-Arderius, Daniel
This Master’s Thesis investigates the environmental impact resulting from the ’reinforced operation’ implemented in the Iberian electricity system following the blackout on April 28, 2025. Using data from the ESIOS platform, this research evaluates how the prioritization of inertial stability —following the loss of robust synchronous capacity— has transformed electricity dispatch management. By developing stability and fragility indices, the study demonstrates that the transition to a decarbonized model reached a critical point where the lack of rotational inertia compromised operational security. The findings confirm a paradigm shift: the System Operator has drastically reduced risk events (alerts and alarms) through massive interventions prioritizing synchronous assets. However, this increased resilience has come at a significant environmental toll, resulting in an excess of 2,679,418 tCO2-eq emitted due to technical constraints between the blackout and the end of 2025. The study concludes by proposing digitization and the use of Grid Forming technologies as the necessary path to decouple grid security from fossil fuel dependency
Análisis de Ciclo de Vida del Sistema Eléctrico Español: Comparación de Tecnologías Renovables mediante Sphera GaBi y SimaPro
(2026-06) Gaviria Rosso, Natalia; Martínez López, Mònica; Serra Rada, Joaquim
The transition towards renewable electricity systems represents a crucial pathway for reducing greenhouse gas emissions and advancing the decarbonization of industrial processes. This study examines the environmental performance of the Spanish renewable electricity system projected for 2025 using Life Cycle Assessment (LCA) and explores its influence on electric arc furnace (EAF) steel production. The modelling was conducted with datasets from the ecoinvent 3.11 database and the LCA software tools Sphera GaBi and SimaPro. The renewable electricity mix was developed based on representative Spanish technologies, including wind, photovoltaic, concentrated solar, hydropower, and biomass. Environmental impacts were assessed using the IPCC 2013 GWP100, ReCiPe Endpoint (H), and Primary Energy Demand methods. Three scenarios were considered: a baseline steel production case, a renewable electricity scenario reflecting the projected Spanish mix, and a 100% wind-powered scenario. Additional analyses addressed technological contributions, historical emission trends, and seasonal variability in renewable electricity generation across Spain. The findings reveal that replacing conventional electricity with renewable sources substantially reduces greenhouse gas emissions, primary energy demand, and impacts on human health, ecosystems, and resource depletion. The wind-based scenario achieved the best overall environmental performance. Moreover, the comparison between Sphera GaBi and SimaPro showed consistent environmental trends, with quantitative differences arising from methodological features specific to each platform. Taken together, these results underscore the potential of renewable electrification as an effective strategy to mitigate environmental impacts and promote the decarbonization of energy-intensive industrial systems.
Predicció de degradació en sistemes d’emmagatzematge electroquímic d’energia mitjançant intel·ligència artificial
(2026-06) Casas Cros, Àlex; Calvo de la Rosa, Jaume; Calderón Díaz, Alejandro
This work develops and evaluates a machine learning-based methodology to predict the State of Health (SOH) of lithium-ion batteries using the NASA Battery Dataset. The study is based on experimental discharge cycles, from which a tabular dataset is constructed, the SOH is defined as the target variable, and a conservative data quality review is applied. The methodology follows a progressive structure: first, a model is built within a base domain composed of batteries tested under relatively homogeneous experimental conditions; then, the analysis is extended to domains with different temperatures, currents, and discharge profiles. The results show that, within the base domain, Linear Regression achieves the best overall performance, with a mean R² of 0.9629. However, when more heterogeneous domains are incorporated, Gradient Boosting provides better relative performance, especially after feature enrichment and hyperparameter optimization. Nevertheless, the final model still shows generalization limitations, particularly in the low-temperature domain and in the external test performed on a discharge profile not included during training. Overall, this work shows that SOH prediction does not depend only on the algorithm used, but also on the representativeness of the dataset, the experimental domain, and the validation strategy applied