Wave energy converters are devices that are designed to extract power from ocean waves. Existing wave energy converter technologies are not financially viable yet. Control systems have been identified as one of the areas that can contribute the most towards the increase in energy absorption and reduction of loads acting on the structure, whilst incurring only minimal extra hardware costs. In this thesis, control schemes are developed for wave energy converters, with the focus on single isolated devices. Numerical models of increasing complexity are developed for the simulation of a point absorber, which is a type of wave energy converter with small dimensions with respect to the dominating wave length. After investigating state-of-the-art control schemes, the existing control strategies reported in the literature have been found to rely on the model of the system dynamics to determine the optimal control action. This is despite the fact that modelling errors can negatively affect the performance of the device, particularly in highly energetic waves when non-linear effects become more significant. Furthermore, the controller should be adaptive so that changes in the system dynamics, e.g. due to marine growth or non-critical subsystem failure, are accounted for. Hence, machine learning approaches have been investigated as an alternative, with a focus on neural networks and reinforcement learning for control applications. A time-averaged approach will be employed for the development of the control schemes to enable a practical implementation on WECs based on the standard in the industry at the moment. Neural networks are applied to the active control of a point absorber. They are used mainly for system identification, where the mean power is related to the current sea state and parameters of the power take-off unit. The developed control scheme presents a similar performance to optimal active control for the analysed simulations, which rely on linear hydrodynamics. Reinforcement learning is then applied to the passive and active control of a wave energy converter for the first time. The successful development of different control schemes is described in detail, focusing on the encountered challenges in the selection of states, actions and reward function. The performance of reinforcement learning is assessed against state-of-the-art control strategies. Reinforcement learning is shown to learn the optimal behaviour in a reasonable time frame, whilst recognizing each sea state without reliance on any models of the system dynamics. Additionally, the strategy is able to deal with model non-linearities. Furthermore, it is shown that the control scheme is able to adapt to changes in the device dynamics, as for instance due to marine growth.

Thesis. 1977. Ocean E.--Massachusetts Institute of Technology. Dept. of Ocean Engineering. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING. / Includes bibliographical references. / Ocean E.

Thesis (Ocean E)--Massachusetts Institute of Technology, Dept. of Ocean Engineering, 1980. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING / Includes bibliographical references. / by Pierre Francois Koch. / Ocean E

Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Ocean Engineering, 1981. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING. / Bibliography: leaves 124-126. / by Paul Sclavounos. / Ph.D.

Um mecanismo móvel instalado no interior de tanques, e conectado à estrutura por molas é proposto para atenuar os efeitos de sloshing em diferentes níveis de preenchimento. Para realizar o estudo e desenvolvimento deste dispositivo, foram propostas duas ferramentas de análise. A primeira ferramenta é baseada em uma formulação desenvolvida analiticamente que calcula a resposta das forças nas paredes do tanque no domínio da frequência. Esta formulação foi desenvolvida como um sistema mecânico de dois graus de liberdade, onde são representados o corpo móvel e o sloshing descrito como um sistema do tipo massa-mola. A segunda ferramenta é uma abordagem utilizando simulações numéricas baseada em um método de partículas de cálculo de escoamento. O método numérico utilizado foi o Moving Particle Semi-implicit (MPS) que calcula o comportamento do fluido interagindo com o corpo móvel e as paredes do tanque sem restrições de movimento. Resultados qualitativos do método analítico foram analisados e mostraram-se bastante consistentes. Resultados quantitativos dos métodos analíticos e numéricos foram comparados com os resultados presentes na literatura e entre si com uma boa concordância. O método analítico é eficiente no dimensionamento inicial do corpo móvel considerando apenas efeitos lineares, e o método numérico é mais indicado para análises mais detalhadas onde efeitos não-lineares podem ser considerados. A integração dos efeitos do corpo móvel no sloshing mostra que o dispositivo de supressão é eficaz em diferentes razões de preenchimento do tanque na atenuação dos efeitos gerados pelo sloshing. / A movable mechanism installed inside tanks and connected to the structure by springs is proposed to attenuate the sloshing effects for different filling ratios. To carry on the investigation and the development of this device, two analysis approaches were elaborated. The first approach is based on an analytical formulation that describes the coupled motion calculating the sloshing force on the walls in the frequency domain. This formulation was developed as a mechanical system with two degrees of freedom representing the moving body and the sloshing described as a mass-spring system. The second tool is an approach that uses numerical simulations based on a particle method to calculate the fluid flow called Moving Particle Semi- implicit method (MPS). This numerical approach calculates the fluid behavior interacting with the moving body and the tank walls without motions restrictions. Qualitative results of the analytical method were analyzed and they were consistent. Quantitative results of the analytical and numerical approaches were compared with the results found in the literature with good agreement. The analytical method is useful to define the moving device initial design considering only linear effects, and the numerical approach is indicated to more detailed analysis where non-linear effects can take place. The integration of the effects of sloshing with the moving body shows the effectiveness of the device for different filling ratios to attenuate the sloshing loads.

Esta dissertação de mestrado propõe o estudo e a implementação de um método de deformação usando modelos tridimensionais sem o uso de malhas baseado na técnica Smoothed Particles Hydrodynamics (SPH), que consiste num sistema de resolução de equações diferenciais para aplicação de conceitos físicos para simular deformação de tecidos moles. A opção pelo método sem malha para processo de deformação é apresentado nesta dissertação como alternativa a um dos métodos mais comuns em simulação de deformação de tecidos, o método massa-mola, explorando questões referentes ao uso de recursos computacionais. Para chegada a definição do método foram analisados os temas envolvendo métodos de deformação, modelos baseados em pontos e o SPH como plataformas para alcançar o desenvolvimento do método proposto pela dissertação. Como forma de comprovar as propriedades do método desenvolvido foi realizada a implementação e testes levando em consideração os modelos de deformação e a interação em tempo real num ambiente de simulação que contempla a deformação de uma mama, levando em conta a comparação com o método massa-mola, o uso de recursos do próprio método em função do aumento de detalhe e do uso de objeto com múltiplas propriedades / This master thesis proposes a study and implementation of deformation method using tridimensional models without edge composed meshes based on Smoothed Particles Hydrodynamics (SPH) technique, that consists on diferential equation solving system to reproduce physical concepts to simulate soft tissue deformation. The option for a meshless method to deformation process is shown in this thesis as an alternative to a very common method in tissue deform simulation, the mass-spring method, reviewing a comparison based on computational resources. To achieve a method definition were analyzed fields of study involving deformation methods, point-based models and SPH as platforms to build and deploy the proposed method for this thesis. To show the characteristics for this developed deformation method was realized the implementation and tests based on deformation models and real time interaction on a simulation environment that includes a breast deformation, taking in account the comparison to mass-spring, number of points of the cloud model and multiple properties

Previous studies have shown that thermal feedback from protostars plays a key role in the process of low-mass star formation. In this thesis, we explore the effects of protostellar heating on the formation of stellar clusters. We describe new methods for modelling protostellar accretion luminosities and protostellar evolution in calculations of star formation. We then present results of a series of numerical simulations of stellar cluster formation which include these effects, and examine their impact. We begin by investigating the dependence of stellar properties on the initial density of molecular clouds. We find that the dependence of the median stellar mass on the initial density of the cloud is weaker than the dependence of the thermal Jeans mass when radiative effects are included. We suggest that including protostellar accretion luminosities and protostellar evolution may weaken this dependence further, and may account for the observed invariance of the median stellar mass in Galactic star-forming regions. Next, we investigate the effects of including accretion feedback from sink particles on the formation of small stellar groups. We find that including accretion feedback in calculations suppresses fragmentation even further than calculations that only include radiative transfer within the gas. Including feedback also produces a higher median stellar mass, which is insensitive to the sink particle accretion radius used. Finally, we compare calculations of small stellar clusters which model the evolution of protostars using a live stellar model with those which use a fixed stellar structure. We find that the dynamics of the clusters are primarily determined by the accretion luminosities of protostars, but that the relative effects of protostellar evolution depend on the accretion rate and advection of energy into the protostar. We also demonstrate how such calculations may be used to study the properties of young stellar populations.

Particle-particle interactions are of paramount importance in every multi-body system as they determine the collective behaviour and coupling strength. Many well-known interactions like electro-static, van der Waals or screened Coulomb, decay exponentially or with negative powers of the particle spacing r. Similarly, hydrodynamic interactions between particles undergoing Brownian motion decay as 1/r in bulk, and are assumed to decay in small channels. Such interactions are ubiquitous in biological and technological systems. Here I confine multiple particles undergoing Brownian motion in narrow, microfluidic channels and study their coupling through hydrodynamic interactions. Our experiments show that the hydrodynamic particle-particle interactions are distance-independent in these channels. We also show that these interactions affect actively propelled particles via electrophoresis or gravity, resulting in non-linear transport phenomena. These findings are of fundamental importance for understanding transport of dense mixtures of particles or molecules through finite length, water-filled channels or pore networks.

[pt] Neste trabalho estudamos o comportamento dinâmico de um gás granular a baixa densidade do ponto de vista hidrodinâmico. Fizemos também uma análise de instabilidade linear das equações de Navier-Stockes para tais sistemas. Finalmente, construímos um modelo de sistema de celulas dinâmicas para simular computacionalmente esses sistema e o aplicamos a um gás granular, a fim de entender um pouco mais sobre seus comportamentos dinâmicos, sob diversas condições de contornos externos. / [en] In this work we study the dynamic behavior of a granular gas at low density from the hydrodynamic point of view. We also perform a linear instability analysis of the Navier- Stockes for such systems. Finally we construct a cell dynamic systems model to simulate computationally these systems and apply it for a granular gas in order to understand a little more about its dynamic behavior, under several boundary conditions.

Esta pesquisa teve como objetivo estudar a hidrodinâmica, a transferência de massa e a inativação de microrganismos indicadores durante o processo de ozonização de esgoto sanitário tratado previamente em reator UASB. A pesquisa foi dividida em três grupos de ensaios, experimentos I, II e III, nos quais foram medidas as concentrações de bolhas de ozônio, os campos de velocidades instantâneos e médios, os diâmetros das bolhas de ozônio e realizados ensaios de desinfecção com ozônio para inativação de microrganismos indicadores - E. coli, colifagos e Clostridium perfringens - e remoção de DQO. Estes ensaios foram realizados tanto em batelada (experimento I e II) quanto em fluxo contínuo (experimento III), e as dosagens de ozônio aplicado foram de 10 e 20 mg/L e os tempos de contato de 5, 10, 15 e 20 minutos. As concentrações, velocidades e diâmetros das bolhas de ozônio foram proporcionais às vazões de ozônio aplicadas e variaram, respectivamente, de 0,23 a 5,26%, de 6,84 a 10,81 cm/s e de 0,260 a 1,781 mm. A quantidade de ozônio consumido variou entre 50,5% e 99,3%, em relação à dose de ozônio aplicado, e seu valor foi decrescente com o aumento da vazão de ozônio. Quanto à inativação dos microrganismos indicadores, C. perfringens apresentou-se como o mais resistente à ação bactericida do ozônio seguido por E. coli e colifagos. Foram inativados 1,7 log de C. perfringens (Ct = 80 mg.min/L); 6,1 log de E. coli (Ct = 133 mg.min/L); e 100% de colifagos (Ct = 38 mg.min/L). O controle das características de ensaios - dose e vazão aplicadas e tempo de contato -, bem como das características hidrodinâmicas das bolhas de ozônio podem otimizar o processo de ozonização de esgoto sanitário tornando-o atrativo técnica e economicamente. / This research had as objective study hydrodynamics, mass transfer and indicating microrganisms inativation during the ozonization process of sanitary sewer treated previously in UASB reactor. The research was divided in three groups of assays, experiment I, II e III, in which had been measured the concentrations, velocity and diameter of bubbles and carried through assays of disinfection with ozone for indicating microrganisms inativation - E. coli, coliphages and C. perfringens - and COD removal. These assays had been carried through in batch (experiment I and II) how much in continuous flow (experiment III), and the dosages of applied ozone had been of 10 and 20 mg/L and the times of contact of 5, 10, 15 and 20 minutes. The concentrations, velocity and diameters of ozone bubbles had been proportional to the applied ozone outflows and had varied, respectively, of 0,23 to 5,26%, of 6,84 to 10,81 cm/s and 0,26 to 1,781 mm. The amount of consumed ozone varied between 50,5 and 99,3%, in relation to the dosage of applied ozone, and its value was decreasing with the increase of the ozone outflow. For the indicating microorganisms inativation, C. perfringens was presented as most resistant to the bactericidal action of ozone followed for E. coli and coliphages. Had been inactivated 1,7 log of C. perfringens (Ct = 80 mg.min/L); 6,1 log of E coli (Ct = 133 mg.min/L); and 100% of colifphages (Ct = 38 mg.min/L). The control of the characteristics of assays - dosage and outflow applied and time of contact -, as well as of the hydrodynamic characteristics of the ozone bubbles they can optimize the ozonization process of sanitary sewer becoming it attractive technique and economically.