APRÈN. Portal de la producció docent del professorat de la UPC

The accelerating adoption of electric vehicles (EVs) necessitates the development of efficient and sustainable electric vehicle charging station (EVCS) infrastructure. This thesis addresses the challenges posed by EVCS integration into power systems, focusing on grid impacts and mitigation strategies. Key methods include the use of photovoltaic (PV) systems, battery energy storage systems (BESS), and active power filters to reduce grid stress and improve system stability. A comprehensive technical handbook is presented, guiding the design of PV systems, the selection and assembly of battery packs, and their integration into EVCS. A case study evaluates six operational scenarios, accounting for uncertainties such as demand variability, operating hours, land availability, and grid strength at the point of common coupling. These scenarios assess the feasibility of different energy penetration levels from PV, BESS, and the grid. Economic and environmental analyses highlight the profitability of each scenario and quantify the sustainability benefits, including carbon emission reductions and energy efficiency gains. The findings demonstrate that integrating renewable energy and storage technologies can significantly enhance EVCS viability while reducing grid dependence. The study concludes with recommendations for extending battery pack lifespan to improve long-term system performance.

As the development of Industry 4.0 technologies, utilization of Internet-of-Things (IoT) devices also have increased. Along with increased demand, energy consumption of increased indus- try 4.0 technologies, have led researchers to concerns regarding the carbon footprint that these devices left behind. With development of piezoelectric energy harvesters (PEH), with suitable interface, energy generated from vibrational kinetic energy can be stored for later use. This development have motivated researchers in pursue for high efficient energy harvesters that can be used in low power application. In this paper, recent technological advances in field of power conditioning of PEH devices is examined and two fundamental approaches are analyzed and simulated. Popular due to its electrical characteristics, Synchronised Switch Harvesting on Inductor (SSHI) and Synchronized Electrical Charge Extraction (SECE) are two widespread methods used as an interface in PEH devices, with latter providing constant power output along the frequency range. Paper provides parametric evaluation of novel self-powered P-SSHI inter- face proposed by [8], along with application of this configuration to other interfaces, S-SSHI and SECE. Proposed NSP P-SSHI interface, have showed better output performance among all candidates with output power of 501 μW. Moreover, in this paper, impact of capacitance values in NSP configuration also examined, which was omitted in original paper. For experimental evaluation of the project, dedicated test-bench printed on PCB was manufactured.

This thesis investigates the economic potential of energy flexibility from the perspective of end customers. The study included the development of Python optimization model based on linear programming used to analyze different scenarios of energy management. The scenarios include the use of solar PV systems with and without battery storage and studied impact of participation in different energy markets – SPOT markets and Frequency Containment Reserve, allowed by aggregation of energy flexibility. The objective is to assess how different technologies and strategies can optimize energy costs and enhance the economic viability of flexible energy systems for end users.

Simulations have been utilized to confirm the initial algorithm ideas and verify their feasibility. The algorithm will be executed within memory-constrained micro-controllers, adhering to strict hardware limitations. The final implementation will be coded in C to ensure compatibility with the micro-controller environment. The principles of test-driven development (TDD) will be adhered to, ensuring the creation of a robust, error-free code base. A test jig consisting of a Raspberry Pi and two Nucleo boards will be used to conduct automated tests. To conclude, manual testing with the already integrated code in a real motor will confirm the analytic results.

The main objective of this project is to automate the routing of medium and low voltage cables, as well as instrumentation and control, fire and telecommunication cables. To achieve this, Python will be used as a fundamental tool, facilitating the determination of the trays or ducts through which these cables must pass using Dijkstra algorithm. Before getting into the main topic, this paper provides a brief introduction to the concept of routing, highlighting the typical challenges that may arise and the relevance of its automation. This will be followed by details of how the code was configured and developed. The case study will focus on a plant under construction in Grudziądz, Poland. This facility is a combined cycle power plant, equipped with a 560 MW gas turbine, currently under development in the Polish province of Kujawy-Pomorskie. The project will be executed in compliance with international industry standards, specifically IEEE and IEC standards. Finally, a demonstration of the Python code will be presented, together with an analysis of possible improvements and future targets to be considered.

Globally, the energy-intensive building, particularly residential and commercial, contributes significantly to environmental pollution through fossil fuel consumption and green house gas emissions. Rising demand of energy leads to increase prices, inducing frequency and voltage deviations that impact electric grids. Urgent measures, including adopting innovative technologies, are imperative to mitigate climate change effects and achieve zero Carbon Dioxide (CO2) emissions. In this context, we focus on the synergistic potential of rooftop Photovoltaic (PV) systems to cost-effectively address residential-scale energy challenges and reduce emissions. We design a grid-connected solar PV with Battery Backup (―PV+BB‖) system in bidirectional mode for the environment of Barcelona, Spain. The economic viability of this ―PV+BB‖ system relies on effective energy dispatching schedules, considering diverse load profiles, system sizes, generation patterns, degradation data, and utility rates. We employ two heuristic dispatch strategies; manual and automatic dispatchs (peak shaving & retail rate) within the System Advisor Model (SAM) tool to evaluate energy dispatching strategies under different utility rates. The integrated algorithm is tested with a nonlinear generic electrochemical battery model. The study aims to contribute insights into optimizing residential energy consumption by giving priority to the cheapest energy source and reducing energy bills through strategic technology integration. The findings indicate that the retail rate dispatch algorithm excels over SAM‘s current algorithms, especially in situations where managing demand charges and energy trading are critical, and when dealing with substantial battery degradation costs. This underscores its efficacy in optimizing energy usage and reducing operational expenditures in residential environments. Overall, this setup can significantly reduce CO2 emissions. By not only lowering energy costs but also minimizing carbon emissions, it has a positive environmental impact.

The protection issues surrounding multiterminal High Voltage Direct Cur- rent (MT-HVDC) grids based on Voltage Source Converters (VSC) are ex- amined in this thesis, with an emphasis on the function of hybrid HVDC circuit breakers in reducing DC-side faults. The study emphasises how im- portant it is to isolate and detect faults quickly, accurately, and specifically as HVDC networks become more intricate as a result of the growing inte- gration of renewable energy sources. The study evaluates the performance of hybrid HVDC circuit breakers under various fault scenarios by compre- hensive modelling and simulation using Simulink, demonstrating their ef- ficiency in shortening fault clearing times and improving system stability. Undervoltage, overcurrent, and current/voltage derivative-based fault de- tection algorithms are compared, and the results show that voltage-based techniques perform better in high-resistance fault scenarios.Hybrid HVDC circuit breakers, which combine the reliability of mechanical disconnectors with the speed of solid-state devices, are identified as a promising solu- tion for modern MT-HVDC grids. This research contributes valuable insights into the design and application of hybrid HVDC circuit breakers and protection algorithms, showing that they significantly improve the operational safety and reliability of HVDC grids. Future studies ought to concentrate on lowering power losses that occur during regular operations and improving the scalability of hybrid HVDC circuit breaker designs for bigger grids. It is projected that improvements in self-triggering mechanisms and circuit breaker technology will further reduce conduction losses and speed up operation. Furthermore, the inte- gration of machine learning (ML) and artificial intelligence (AI) into fault detection algorithms presents a promising opportunity to enhance detec- tion efficiency and precision, all the while accommodating dynamic grid conditions. The effect of communication latency on problem detection may be lessened by investigating communication-independent protection systems, such as travelling wave-based methods. For a viable deployment, extensive testing and validation in real HVDC grid environments are nec- essary, and complicated failure scenarios should be simulated in order to better inform future implementations.By addressing these issues, HVDC grid protection will be improved, along with system efficiency, depend- ability, and capacity to handle the increasing needs of renewable energy integration.

The study provides an innovative means to sustainable hydrogen production with the integra- tion of a hybrid photovoltaic system. There is, thus, a growing pressure worldwide to change from the conventional sources of fossil fuel to cleaner energy sources on account of the alarming concerns related to climate change. Hydrogen is seen as one of the most promising candidates, since it is a versatile and environmentally friendly substance when used as fuel. However, the conventional hydrogen-producing methods often use fossil fuels, thus undermining the envi- ronmental benefits of using hydrogen. The model to be developed in this research is supposed to be effective and cost-effective in harnessing solar energy for water electrolysis to allow the production of clean hydrogen gas. This research has two major objectives: first, to fill the gap in the need for sustainable hydrogen production methods; second, to show that solar energy can be a clean source applicable in hydrogen generation. This can be achieved by designing and implementing an efficient solar-energy-harvesting hybrid system that is capable of converting such energies into electrical power through electrolysis. It simulates the system's performance from a series of simulations and experimental tests, including but not limited to solar irradiance, electrolyzer efficiency, and hydrogen production rates. The presented research results offer evidence that the hybrid PV system exhibits promising ca- pabilities for sustainable hydrogen generation. Through proper design of system components and optimization of the parameters with which the system operates, considerable enhancement in terms of efficiency and cost-effectiveness is achieved compared to conventional methods. Moreover, renewable sources of energy contribute to the environmental sustainability of hy- drogen production because its production reduces greenhouse gas emissions and saves fossil fuel resources. The results will highlight the potential of hybrid PV systems as a viable solution for sustain- able hydrogen generation, offering a pathway towards a cleaner and more resilient energy fu- ture. This research contributes to the ongoing efforts to combat climate change and advance the transition towards renewable energy sources, underscoring the importance of innovation and interdisciplinary collaboration in addressing global environmental challenges.

Aquest treball de fi de màster investiga el paper dels convertidors multiport (MPCs) en la millora de la resiliència i l'eficiència de les xarxes de distribució. La recerca se centra en la reducció de l'energia no subministrada (ENS) i en com aquests poden facilitar la integració de recursos energètics distribuïts (DER) a la xarxa. Utilitzant com a banc de proves el sistema de 33 busos de l'IEEE 33 modificat, s'han realitzat simulacions per avaluar el rendiment dels MPCs sota diversos escenaris, incloent-hi fallades simples i dobles de línia i modes d'operació grid-forming i grid-following de les unitats de generació distribuïda (DG). Tot i que en el cas base el rendiment elèctric de la xarxa de distribució ja és acceptable, l'estudi revela que els MPCs milloren significativament el rendiment de la xarxa, especialment en la regulació de voltatge i la reducció de pèrdues. L'anàlisi de la viabilitat econòmica indica que els MPCs d'escala parcial proporcionen una solució més rendible per a la millora del rendiment de la xarxa respecte a aquells a escala total. A més, la recerca avalua el potencial d'ingressos de la participació de sistemes d'emmagatzematge d'energia amb bateria (BESS) en els mercats elèctrics, destacant els beneficis de la participació al mercat diari i el de serveis auxiliars. Els resultats donen suport a l'adopció dels MPCs com una estratègia viable per a modernitzar les xarxes de distribució per a acomodar la creixent integració dels DER i facilitar aconseguir els objectius de sostenibilitat en l'àmbit energètic.

Finding an optimal point of operation that reduces the operational cost of a power system while it complies with all the technical limitations is a complex problem that every grid operator has to tackle. To solve this optimization problem, many strategies have been developed, such as the DC-OPF (Direct Current Optimal Power Flow), which performs some simplifications over the power flow equations and constraints to linearize and solve easily the problem, or the AC-OPF (Alternating Current Optimal Power Flow), which is the most exact method, but also more computationally expensive. It encapsulates the economic dispatch, by the means of a cost objective function, and all the technical constraints that the grid has to respect. The work developed in this thesis presents an AC-OPF solver integrated in GridCal, an opensource power system analysis tool developed by the Spanish TSO, Redeia. This work has been encapsulated under the SIROCO project, an initiative focused on adding and improving functionalities to make GridCal the go-to software when it comes to planning the Spanish network. The solver has been developed in Python, based on the Matpower formulation, and adds important features such as transformer setpoints optimization, DC links, reactive power limitations and some improvements with respect to Matpower's original solver convergence.

The objective of the Thesis is to study and analyze the modeling and control of a two-level VSC converter in grid following mode within a point-to-point HVDC link. To do that, the two converters and the link between them have been modeled. All the control algorithms are detailed, as well as their parameterization, implementation and validation. Special attention is paid to the network support functionalities offered by the converters, such as frequency and AC voltage droop, but without losing sight of the physical limits of the converter and the different techniques to avoid exceeding them. On the other hand, the response of the HVDC link is compared with that of a synchronous generator model, detailing the importance of inertia and of the latter's behavior as a voltage source. The interaction between the generator and converter support droops when facing a mismatch between generation and demand is studied. Finally, the stability limits of the system are analyzed for variations in the inertia and droop characteristics of the converter.

Accurate battery modelling is the key to effective energy management, as it plays an important role in extending battery lifespan and ensuring operational safety. Within the field of lithium- ion (Li-ion) battery cell technology, precise estimation of state-of-charge (SOC) and state-of-health (SOH) is fundamental for achieving optimal performance. This thesis embarks on an exploration of the different techniques for estate and health monitoring. By navigating these techniques, it is intended to clarify the challenges, opportunities and advances in the field of lithium-ion battery cell modelling. The proposed modelling framework is validated using experimental data obtained from cylindrical Li-NMC (Lithium Nickel Manganese Cobalt Oxide) cells. Diverse test procedures were conducted under different conditions to characterize their performance. The analysis of these tests is also an important aspect of this thesis, as it sheds light on the behaviour of Li-NMC cells and provides insight into their performance characteristics in various scenarios. Departing from the literature, this Master's thesis reviews the fundamentals of secondary batteries, with a particular focus on Li-NMC battery cells. Within the cell structure, various elements are discussed, offering a detailed understanding of their properties and their roles. Moreover, this thesis addresses the ageing mechanisms, examining factors such as calendar ageing, and cycle ageing. Additionally, it provides an overview of different battery cell modelling approaches, covering equivalent circuit models (ECM), electrochemical models, and data-driven models. The selection of the models and techniques are based on the available resources and it aligns with the aim of providing robust solutions for accurate battery modelling. In this context, two different ECMs have been developed. One is primarily based on a dynamic load profile, enabling online monitoring, while the other relies on Electrochemical Impedance Spectroscopy (EIS), an offline method. Furthermore, a differential analysis is conducted, establishing a relationship between the performance during charging tests and ageing mechanisms

An Electric Power System (EPS) is a function of multivariable order in which the dynamic response is influenced by a wide range of devices with different characteristics and response times. Stability is a condition of balance between opposite active forces in the system. Depending on the network topology, operating conditions and the shape and magnitude of the disturbances, these opposing forces may experience the sustained imbalance that leads to various forms of instability. Currently, the EPS is changing its behavior from slow to fast dynamics due to the insertion of renewable energy sources and HVDC systems with power electronic converters. Additionally, the stability of the EPS can be assessed through offline simulation and mainly by reducing the detail of the elements to improve simulation times. However, these simplifications are no longer valid with the emergence of power converters in electrical systems. Indeed, phasor-type simulations are not capable of capturing fast dynamics, therefore they must be migrated to EMT-type simulations. For this reason, the simulation technique called Co- simulation that addresses this problem seeks to improve simulation times without losing the precision of the system state variables. This thesis presents different types of Co-simulation in the current EPS from coupling programs of different natures to models in different time scales. This work has also developed the different types of simulations and how they interact in the Co-simulation environment. On the other hand, the entire VSC structure has been developed which is a key element that has changed the behavior of the system dynamics. Finally, with all of the above, a way to represent the Co-simulation in EPS has been implemented. In this model, it has been concluded that fast dynamics in the EPS such as short circuits, Co- simulation provides better simulation times rather than EMT simulation but with errors. On the other hand, slow events such as changes in demand/generation continue to be valid phasor simulations since they are less demanding in simulation time and maintain the precision of the EMT model

The main objective of this project is to carry out the sizing of a storage system connected to a solar photovoltaic installation whose purpose is to supply the demand of an irrigation system. This project will be based on the existing infrastructure located in l'Espluga Calba, in the province of Lleida. In order to achieve the desired objective, a technical and economic analysis will be carried out, as well as an optimisation of the installation and simulation of the behaviour of the battery to be sized. For this reason, two clearly defined scenarios will be evaluated: one in which the installation will be connected to the Spanish electricity grid and another in which it will be totally isolated from it. To carry out all this, a simulation and numerical analysis tool will be used with its own computer language, Matlab. Within each of the scenarios, and as far as possible, multiple analyses will be carried out in order to understand the impact of certain input parameters, such as the maximum available budget and the different space restrictions that exist in photovoltaic installations, in obtaining the results. Finally, it can be concluded that the implementation of this project is not only viable from an economic and technical point of view, but also becomes a key element in promoting the use of solar photovoltaic installations in the field of agriculture. This will reduce dependence on the electricity grid, help to achieve energy self-sufficiency and, finally, reduce carbon dioxide emissions

Aquest treball final de màster té com a finalitat mostrar una nova metodologia que és possible seguir a l'hora de dissenyar un recuperador d'energia elèctrica que vulgui ser instal·lat en les subestacions de tracció de la xarxa ferroviària convencional. L'anàlisi aquí presentada pot ser replicada per a altres línies i/o subestacions de tracció on es vulgui implantar aquest tipus de mesures. Un recuperador d'energia no és més que un convertidor de potència que permet convertir una potència d'entrada en corrent continu en una potència de sortida en corrent altern, d'aquesta manera, l'energia procedent de la catenària pot ser retornada de nou a la xarxa o utilitzada per a satisfer altres consums. D'aquesta manera es permet millorar l'eficiència energètica de la xarxa ferroviària pel fet que s'evita malgastar l'energia alliberada pels trens en el moment de la frenada. Aquesta energia, tal com estan dissenyades ara mateix les subestacions de tracció si es malgasta en la majoria dels casos, es crema en unes resistències a bord del tren. A més a més d'estudiar una nova metodologia es mostren els estudis econòmics i ambientals que sorgeixen de dur a terme aquest tipus de mesures. Per aconseguir tot l'anterior, s'analitza una línia de la xarxa ferroviària ubicada a Barcelona. Aquesta línia està alimentada per una tensió de 3 kV en contínua i a més a més és una de les línies principals d'aquesta ciutat. Gràcies a l'anàlisi de les dades corresponents a aquesta línia, és possible obtenir uns resultats viables i que puguin servir com a punt de partida per altre tipus d'estudis

This paper studies the performance of Machine Learning and Deep Learning algorithms in the task of state estimation for low-voltage distribution networks. As renewable energy technolo- gies are increasingly integrated into these grids, monitoring the grid becomes more challenging for system operators. We explore the applications of novel Artificial Neural Networks for grid observability, due to their known generative power and low data requirements. Additionally, the performance of ANNs is compared with gradient boosting Machine Learning regression algo- rithms. The objective of this document is to demonstrate the effectiveness of these AI methods to overcome the lack of reliable measurements in low voltage grids, in order to accelerate the integration of renewable energies at a localized level and contributing to the decarbonization of our society

The objective of this master's thesis has been to modify an energy control and management model developed within the state competitive project Flexibility of distributed energy resources to optimise the operation of distribution grids FLEXRED by José Fernando Forero Quintero in his PhD Thesis to incorporate the flexibility services for grid congestion in a specific geographical area into the real-time domestic energy management software (FLEMS). For such aim realises an enquiry of information of the ongoing rule and defines the theoretic frame from which will develop distinct study cases. The first case study distinguishes between flexible and inflexible loads, so that in the event of insufficient potential, some loads can be disconnected by means of battery management and photovoltaic generation in order to fulfil the flexibility service. For this purpose, a control signal is generated which determines whether and when the loads have to be disconnected. In the second case study, it is proposed to measure the delivered flexible energy based on a reference value. This corresponds to the average consumption forecast by the Home Energy Management System (HEMS) programme for the period prior to the flexibility service. In addition, the number of variables taken into account by the simulation is reduced, combining the shiftable loads and the inflexible loads in a single block. Once the simulation models have been developed, the energy and economic results are analysed. In terms of energy, it is necessary to check whether the flexibility requirement is met and what measures the system takes to comply with it. In economic terms, the cost of not modifying the HEMS schedule, modifying it without complying without any energy model and modifying it using the FLEMS schedule with or without providing a flexibility service must be taken into account. In this way, the different results obtained are compared and the most cost-effective model for the user is determined