Search

The development in the area of robotics and automation has generated a greater demand for autonomous systems. This reality, as well as the development of devices with increasingly lower energy consumption, has been a factor that encourages the research and development of methods for autonomous generation of renewable energy that have a low cost and are capable of powering several small-scale systems. One possible solution for this type of application is electromagnetic generators. This work focuses on the use of artificial intelligence algorithms to better develop the innovative concept of electromagnetic power generators with magnetic levitation architecture. It is important, for its effective application, that the device is able to adapt its performance with the dynamic change of the mechanical excitation. The main objective of the project is to predict the behavior of these generators using an artificial neural network for this purpose because it is a solution capable of predicting with high accuracy the behavior of nonlinear systems. For this project, an algorithm was developed in MatLab. The configuration called NARX was chosen, and different configurations were tested, as well as two different types of training algorithms, with the objective of obtaining the best possible result. For training the neural network, simulated data obtained previously to this work was used, in the end the network was tested trying to predict the behavior of the generator with its experimental data. The results show that the network was able to predict the behavior of the generator in more than 80% of the cases satisfactorily. These results indicate the possibility of developing a neural network capable of performing self-adaptation of a mechanical vibration generator, making it capable of generating energy for a wide range of excitation frequencies. O desenvolvimento na área da robótica e automação tem gerado uma maior procura por sistemas autónomos. Essa realidade, assim como o desenvolvimento de dispositivos com um consumo energético cada vez menor, tem sido factor de incentivo à pesquisa e desenvolvimento de métodos de geração autónoma de energia renovável que tenham um baixo custo e sejam capazes de alimentar diversos sistemas de pequena escala. Uma possível solução para esse tipo de aplicação são os geradores eletromagnéticos. Esse trabalho tem por foco o uso de algoritmos de inteligência artificial para melhor desenvolvimento do inovador conceito de geradores eletromagnéticos de energia elétrica com arquitetura em levitação magnética. É importante, para que sua aplicação seja efetiva, que o dispositivo seja capaz de adaptar seu desempenho com a alteração dinâmica da excitação mecânica. O principal objetivo do projeto é prever o comportamento desses geradores utilizando uma rede neuronal artificial para esse fim por ser uma solução capaz de prever com alta precisão o comportamento de sistemas não lineares. Para esse projeto, foi desenvolvido um algoritmo no MatLab. Optou-se pela configuração denominada NARX e foram testadas diferentes configurações, além de dois tipos diferentes de algoritmos de treino, com o objetivo de se obter o melhor resultado possível. Para o treino da rede neuronal, foram utilizados dados simulados obtidos previamente a este trabalho, no fim a rede foi testada tentando prever o comportamento do gerador com seus dados experimentais. Os resultados mostram que a rede foi capaz de prever o comportamento do gerador em mais de 80% dos casos de forma satisfatória. Esses resultados indicam a possibilidade do desenvolvimento de uma rede neuronal capaz de realizar a auto-adaptação de um gerador de vibrações mecânicas, tornando-o capaz de gerar energia para uma grande gama de frequências de excitação. Mestrado em Engenharia de Automação Industrial

Buildings represent 40% of the Union’s final energy consumption; the member states should establish a strategy to improve the energy performance in buildings and reduce the consumption of non-renewable primary energy. In Spain, the implementation of the Technical Building Code (CTE) compels to install solar thermal collectors in new buildings providing a minimum solar contribution of domestic hot water (DHW). In north and center European countries, e.g. Denmark, Germany and Austria, new installations also supply heat for the space heating needs. The approach of central solar heating plants with seasonal storage (CSHPSS) is the storage of solar thermal energy from the period of higher offer (summer) to be consumed in the periods of higher demand (winter). These installations are integrated into district heating systems that supply heat for a large number of dwellings and reach a solar fraction of 50% or higher. In this thesis the experience gained in Europe on centralized solar district heating systems with seasonal storage will be transferred to the Spanish situation, in order to establish the conditions and criteria for installing these systems in Spain in the midterm. The main objective of this thesis is the proposal and design of CSHPSS that could be able to provide a high fraction of thermal energy demand with solar thermal energy for different climatic areas. These systems should be feasible from a technical viewpoint, economically acceptable, and with a low environmental impact. That is, this thesis unveils the requirements for the feasibility of CSHPSS and is intended to foster their development in Spain. In order to reach this objective, it has been performed a revision of the state of the art of district heating systems, with emphasis to: i) solar district heating systems and CSHPSS; ii) design and calculation methods that could be used for new systems in Spain; iii) economic data and results from existing solar district heating systems and CSHPSS in Europe and worldwide; and iv) environmental assessment methodologies and analysis performed for solar thermal components and systems. An original calculation method for the analysis, design and evaluation of these installations from technical, economic and environmental points of view has been developed. The variation of solar radiation along the day and the year and the monthly distribution of the residential sector demand are considered. The main advantage of the method developed, compared to other methods, is the simplification of the calculation process and the utilization of simple climatic and demand data. The method developed has also been used to perform parametric analyses that have served to obtain new design criteria for different locations. The technical viability of these installations is not enough argument to motivate their development. The investment cost of these installations has therefore been analyzed according to the main design parameters (area of solar collectors and seasonal storage volume) and validated with results from real projects in north European countries. Moreover, this thesis analyzes the environmental impact of these installations using the Life Cycle Assessment (LCA) methodology. This impact assessment not only considers the consumption of fuels and electricity for the production of energy but also the consumption of materials for the construction of the plant. Three different environmental assessment methods have been used to determine the impacts generated and avoided by a CSHPSS: i) emission of greenhouse gases and their contribution to the global warming; ii) consumption of primary energy; and iii) environmental indicator IMPACT 2002+, which encompasses a significant range of environmental burdens. Based on the previous thermal, economic and environmental models and analyses developed, appropriate design criteria for CSHPSS in different geographical areas have been established. It has been concluded that design criteria are strongly dependent on the local climatic and demand conditions. Therefore, CSHPSS designs for north European countries cannot be applied in south Europe. Furthermore, it has also been concluded that CSHPSS have a considerable potential in Spain; i.e. it is interesting to build CSHPSS in those regions of Spain with significant heating demand, because they can supply heat to large communities at a competitive cost with a low environmental impact. Finally, from the calculation and analysis tools developed in the thesis, a software application with a friendly user interface has been developed to pre-design CSHPSS. The software is mainly oriented to European locations and provides the thermal performance, economic cost and environmental impact of the evaluated CSHPSS. Los edificios representan el 40% del consumo de energía final de la Unión Europea; los estados miembros deben establecer una estrategia para mejorar la eficiencia energética de los edificios y reducir el consumo de energía primaria no renovable. En España con la entrada en vigor del Código Técnico de la Edificación (CTE) se ha pretendido sustituir energía procedente de combustibles fósiles por energía solar y se obliga a la instalación de sistemas solares térmicos para proporcionar una contribución solar mínima anual a la demanda de agua caliente sanitaria (ACS). Sin embargo en los países del centro y norte de Europa como Dinamarca, Alemania y Austria, que destacan por su aprovechamiento de la energía solar, vemos que parte de sus nuevas instalaciones aportan energía solar térmica para cubrir también las necesidades de calefacción. El interés de las centrales solares térmicas con acumulación estacional consiste en el aprovechamiento del exceso de captación solar en el periodo de mayor oferta (verano) para su consumo en el periodo de mayor demanda (invierno). Estas instalaciones se integran en sistemas de calefacción de distrito que proveen energía térmica a un elevado número de viviendas alcanzando una fracción solar elevada (> 50%). En esta tesis se adapta al caso de España la experiencia obtenida en Europa acerca de los sistemas de calefacción solar de distrito con acumulación estacional y se esclarecen las condiciones y criterios que harían interesante su implantación a medio plazo en nuestro país. El principal objetivo de esta tesis consiste en proponer y prediseñar sistemas de calefacción solar de distrito con acumulación estacional para distintas zonas climáticas y diferentes tamaños de distrito, que sean: i) técnicamente viables, ii) económicamente rentables, y iii) con bajo impacto ambiental. En otras palabras, esta tesis desvela y establece los requisitos para que los sistemas de calefacción solar de distrito con acumulación estacional sean una alternativa interesante, contribuyendo de este modo al desarrollo de estas instalaciones en España. Para poder alcanzar este objetivo se ha llevado a cabo una revisión del estado del arte de los sistemas de calefacción de distrito, haciendo especial énfasis en: i) sistemas de calefacción solar de distrito con acumulación estacional; ii) métodos de cálculo y diseño que puedan ser empleados para la propuesta de sistemas de estas características en distintas zonas geográficas de España; iii) datos y resultados económicos de sistemas de calefacción solar de distrito con acumulación estacional actualmente existentes en Europa y en el mundo; y iv) datos y metodologías para su evaluación ambiental y análisis de ciclo de vida (ACV). Se ha desarrollado un método de cálculo original para el análisis, diseño y evaluación de estas instalaciones desde un punto de vista técnico, económico y ambiental. El método desarrollado considera la variación de la radiación solar a lo largo del día y del año y la distribución mensual de la demanda térmica en el sector residencial. La principal ventaja del método desarrollado frente a otros trabajos es la simplificación del proceso de cálculo y la utilización de datos climáticos y de demanda fáciles de encontrar. Con el método desarrollado se han realizado análisis paramétricos que han servido para definir nuevos criterios de diseño para distintas localizaciones. La comprobación de la viabilidad técnica de estas instalaciones no supondría por si solo un argumento suficiente para impulsar su desarrollo, por tanto se analiza el coste de inversión en función de las principales variables de diseño de la instalación (área de captación solar y volumen del acumulador estacional). Más allá, esta tesis analiza el impacto ambiental de estas instalaciones utilizando la metodología del análisis de ciclo de vida. La evaluación ambiental realizada considera los efectos de los consumos de combustibles y electricidad para la producción de energía térmica, y además el consumo de materiales para la construcción de la planta. Se han utilizado tres métodos diferentes para determinar los impactos ambientales generados y los beneficios ambientales alcanzados: i) emisión de gases de efecto invernadero y su contribución al cambio climático; ii) consumo de energía primaria; y iii) cálculo del indicador ambiental IMPACT 2002+ que abarca una gran variedad de aspectos ambientales. Gracias a los modelos desarrollados y a los análisis llevados a cabo considerando el comportamiento físico, económico y ambiental de los sistemas de calefacción solar de distrito con acumulación estacional, se han definido criterios de diseño adecuados para diferentes zonas geográficas. Una de las principales conclusiones alcanzadas es que el diseño correcto de estos sistemas depende fuertemente de las condiciones climáticas y de la demanda de las viviendas. Por tanto, los diseños aplicados en el norte de Europa no pueden ser trasladados al sur de Europa. Asimismo se concluye que los sistemas de calefacción solar de distrito con acumulación estacional son viables técnica y económicamente en las zonas de España con elevado consumo de calefacción; es decir, estos sistemas cuentan en nuestro país con un potencial elevado para atender las necesidades de calefacción de grandes comunidades a un coste competitivo y además con bajo impacto ambiental. Finalmente indicar que a partir de los modelos y herramientas de análisis elaborados se ha desarrollado una aplicación informática de fácil manejo para el pre-diseño de sistemas de calefacción solar de distrito con acumulación estacional, principalmente orientado a localidades europeas, que proporciona el comportamiento térmico del sistema y estima su coste económico e impacto ambiental.

A notable issue in the field of dielectric elastomers is the characterization of composite materials with improved electromechanical coupling destined for mechanical-to-electrostatic energy converters. To this aim, random composites, where ceramic fillers with high dielectric constant are dispersed in a silicone matrix, represent an interesting option. Currently, the most promising reinforcing materials to be immersed in a silicone matrix, already tested for soft dielectric actuators, are lead magnesium niobate–lead titanate (PMN–PT) and lead zirconate-titanate (PZT). To estimate the performance improvement entailed by the composite device with respect to the homogeneous matrix, a typical four-phase cycle is considered in the model, where nominal load and electric charge are alternately held constant. Different materials are being studied: a composite consisting of a matrix in poly-dimethyl-siloxane (PDMS) reinforced with PMN–PT, assuming a filler concentration of 10% in volume and a PDMS–PZT composite with a 1% volume fraction of the ceramic component. In comparison with pure PDMS, the PDMS–10%PMN–PT allows an increase of more than 60% in the harvested energy per unit volume, while the PDMS–1%PZT composite, entailing a minor improvement, here in the range 23.5–37.4%, exhibits a better performance in terms of generated energy per unit weight. These results provide a guide for the choice and design of materials suitable for the realization of enhanced energy harvesters.