O objetivo deste trabalho foi analisar o comportamento hidráulico de sistemas de abastecimento de água (SDAA), via redes neurais artificiais (RNA). Tendo em vista que o uso de um modelo de simulação hidráulica é inviável para operações em tempo real em SDAA devido à carga computacional que impõe, a abordagem adotada visa capturar o seu conhecimento de uma forma mais eficiente por meio de uma RNA de camada oculta única. Desta forma os conjuntos de dados provenientes de simulações hidráulicas realizadas utilizando o conhecido e consagrado software EPANET combinando diferentes situações e avaliando as variáveis que compõem os dois sistemas distintos, foram tabuladas e inseridas em forma de rotinas de programação desenvolvidas na plataforma do software Scilab. Os conjuntos de dados (entradas e saídas) utilizados para treinar a RNA, foram divididos em conjuntos de treino, validação cruzada e testes, aplicada a dois SDAAs distintos. O primeiro denominado Anytown modificado tratou de um sistema hipotético, cuja finalidade foi avaliar a metodologia e fundamentar o conhecimento. Posteriormente aplicou-se a metodologia em outro SDAA, real e mais complexo, utilizado na conferência de calibração WDSA Water Distribution System Analysis, realizada na cidade de Tucson, Arizona (EUA) entre os dias 12 e 15 de setembro de 2010. Desta maneira foram obtidos valores simulados de potência energética consumida no bombeamento, níveis de reservação e pressões, muito próximos dos valores reais para os dois SDAAs estudados, comprovando que as RNAs identificadas podem ser consideradas ferramentas eficientes na substituição dos modelos de simulação hidráulica convencionais. / The objective of this study was to analyze the hydraulic systems of water supply (WSS), using artificial neural networks (ANN). Since the use of a hydraulic simulation model is impractical for real-time operations due to the computational load WSS imposing, the approach aims to capture their knowledge more efficiently through a single hidden layer of RNA. Thus the data sets from hydraulic simulations conducted using the known and established software EPANET combining different situations and evaluating the variables that make up the two different systems were tabulated and entered in the form of programming routines developed in Scilab software platform. The data sets (inputs and outputs) used to train the ANN, were divided into training sets, cross validation and testing, applied to two distinct WSS. The first called Anytown modified dealt with a hypothetical system, whose purpose was to evaluate the methodology and knowledge base. Later the methodology was applied in another WSS, real and complex, used in the calibration WDSA conference Water Distribution System Analysis, held in Tucson, Arizona (USA) between 12 and 15 September 2010. Thus were obtained simulated values of the power energy consumed in pumping, pressures and reservation levels very close to actual values for the two WSS studied, proving that the identified ANN can be considered effective tools to replace in the conventional hydraulic simulation models.

A homogeneidade do solo em sistema de plantio direto, é um assunto questionável, pois alguns autores têm considerado que, com o passar do tempo, há aumento da homogeneidade do solo, outros têm verificado a redução. A hipótese deste trabalho é que em sistema de plantio direto consolidado existe correlação e dependência espacial dos atributos físicos-hídrico do solo e do desenvolvimento vegetativo da soja, sendo a cokrigagem colocalizada interpolador mais representativo destas correlações. O objetivo foi avaliar a variabilidade espacial de atributos físicos do solo e do desenvolvimento vegetativo da soja em sistema de plantio direto, adotado há mais de 19 anos. O trabalho foi desenvolvido na fazenda-escola da Universidade Estadual de Ponta Grossa - Paraná. A área de estudo tem como cultura a soja e está inserida em um relevo cuja declividade máxima, no sentido da pendente, é de aproximadamente 10 %. Nesta área, foram avaliados atributos físicos e hídrico do solo, como: densidade do solo (Ds), granulometria (areia e argila) e condutividade hidráulica saturada (Kfs). Avaliou-se, ainda, atributos de planta: altura da planta, estádio reprodutivo e stand. Para tais avaliações, foi demarcado um grid com espaçamento 10 x 10 metros, onde as avaliações foram realizadas em cada ponto. A análise dos dados foi efetuada por geoestatística, utilizando o pacote de programas GEOSTAT, para todas as variáveis que apresentaram dependência espacial. Foi obtido mapa de krigagem, e para todas as que apresentaram correlação, mapas de cokrigagem e cokrigagem colocalizada. A precisão de tais mapas foi obtida por meio dos menores valores de variância e a raiz quadrada do erro médio (RMSE). Verificou-se existência de dependência espacial na área em estudo, sendo a declividade um dos fatores responsáveis pela variação e o outro fator pode ser atribuído ao manejo uniforme adotado na área. Existiu correlação direta e positiva entre Kfs e areia e negativa com argila. Os atributos que influenciaram positivamente o desenvolvimento da planta foram a Kfs e, negativamente, a densidade do solo (Ds). Quanto aos métodos de estimação, o que obteve o mapa mais representativo da condição real, para a maioria das variáveis estudadas, foi a cokrigagem colocalizada. Apenas para a correlação argila x areia não houve ganho no uso da cokrigagem colocalizada, por isso, a cokrigagem ordinária foi a mais indicada. / The homogeneity of the soil in no-tilled system is being a highly controversial question, as some authors have claimed that there is an increasing in the soil homogeneity over time, while others have proven it to be reduced. The hypothesis is that in established no-tillage system is no correlation and spatial dependence of the physical-hydric soil properties and soybean vegetative development, cokriging-located interpolator being most representative of these correlations The objective was to evaluate the spatial variability of the soil physical attributes and soybean vegetative growth along 19-years of no-tillage system. The study was carried out at the farm-school from the Universidade Estadual de Ponta Grossa - Paraná. The area was cultivated with soybean in a topossequence with a maximum slope of10 %. In this area, were evaluated physical and hydrical soil properties, e.g. soil bulk density (Ds), soil texture (sand and clay) and saturated hydraulic conductivity (Kfs). It was also evaluated plant characteristics such as plant height, reproductive stage and standFor these assessments, it was delimited a 10 x 10 meters grid, where the assessments were carried out in each point. The data analysis was performed using a specific geostatistic software GEOSTAT, for all spatial-dependent variables. A kriging map was obtained from the results, and for all those correlated variables cokriging maps and cokriging-located have been also performed. The maps accuracy were achieved from the smallest values of variance and square root of the mean error (RMSE). The results indicated the existence of spatial dependence in the study area with the slope and soil management the mainly dependent factors, wherein in the latter the spatial variability is disregarded. Kfs was direct and positively correlated with sand, and a negatively with clay. The soil attribute that influenced positively the plant development was Kfs, whilst the bulk density (Ds) influenced negatively. With regard to the estimative methods, cokriging-located produced the most accurate and representative map of the real conditions for most of the variables. The low correlation between sandy and clay fractions makes the ordinary cokriging is being more appropiate than the cokriging-located.

A ocorrência de elevados índices de perdas por vazamentos em sistemas de abastecimento de água é frequentemente ligada às inúmeras quebras de tubulações originadas em decorrência de elevados níveis de pressão atuantes. O controle efetivo de tais perdas requer a detecção e localização dos vazamentos para uma rápida reparação do sistema. Neste sentido, a consideração de análises hidráulicas no escoamento transitório tem se revelado especialmente útil aos propósitos de calibração e detecção de vazamentos. Tal análise pode revelar substancial quantidade de informações sobre as propriedades físicas e o nível de integridade do sistema, pois as ondas de pressão resultantes são afetadas pelos diversos dispositivos e fenômenos, incluindo os vazamentos. Assim, o presente trabalho visa o estudo dos transitórios hidráulicos para a calibração e detecção de vazamentos em modelos de sistemas de distribuição de água. Para tanto, são utilizados dados de redes hipotéticas e de laboratório, e modelos inversos resolvidos por métodos de busca global e local. Devido o emprego de tubos plásticos no circuito de laboratório, o clássico modelo da coluna elástica demonstrou-se ineficiente na reprodução do comportamento hidráulico de tal sistema. Resultados satisfatórios foram obtidos somente com um modelo hidráulico que considere o comportamento viscoelástico dos materiais dos tubos do circuito experimental. / The occurrence of large leakage losses in water supply systems has been frequently linked with pipe breaks resulting from high pressures levels in pipes. Efficient location of leakages is required in order to effectively control water losses and quickly repair the system. Analysis of hydraulic transients has been particularly useful for calibration and leak detection purposes. System observation for such analysis can reveal a substantial amount of information concerning physical properties and the integrity of the system, since water hammer waves are affected by different features and phenomena, including leaks. Thus, this research focuses on hydraulic transients for leak detection and calibration of water distribution system models. These objectives are achieved using data obtained from experiments performed on an experimental facility and numerical experiments on hypothetical networks. Inverse methods were based on both global and local search methods. Classic water hammer theory proved to be imprecise in describing the observed behavior of the hydraulic system composed of plastic pipes. Satisfactorily results were obtained with a hydraulic transient solver considering viscoelastic behavior of the pipe material.

This is a phd thesis

Since the damming and channelization of the lower Long Tom River in the 1940s and 1950s, the quality and quantity of habitat for coastal cutthroat trout and spring Chinook salmon in the watershed has dramatically diminished. In order to better understand the potential for stream restoration, this study uses 2D hydraulic modeling to determine the impact of reconnecting historical meander bends to the main stem of the lower Long Tom River on localized flooding, sediment erosion and deposition, and salmonid physical habitat. These models compare the current conditions to two restoration scenarios that allow for fish passage given 1, 2, and 5-year flood events at two study sites. This study reveals important variations in the impact of restoration between the study sites and the reconnection methods. It also suggests that there is the potential for a large increase in the area of accessible habitat with stream restoration.

Interruption to water services and low water pressure conditions are commonly observed problems in water distribution systems (WDSs). Of particular concern are the unplanned events, such as pipe bursts. The current regulation in the UK requires water utilities to provide reliable water service to consumers resulting in as little as possible interruptions and of as short possible duration. All this pushes water utilities toward developing and using smarter responses to these events, based on advanced tools and solutions. All with the aim to change network management style from reactive to a proactive, and reduce water losses, optimize energy use and provide better services for consumers. This thesis presents a novel methodology for efficient and effective operational, short time response to an unplanned failure event (such as pipe burst) in a WDS. The proposed automated, near real-time operational response methodology consists of isolating the failure event followed by the recovery of the affected system area by restoring the flows and pressures to normal conditions. The isolation is typically achieved by manipulating the relevant on/off valves that are located closely to the event location. The recovery involves selecting an optimal combination of suitable operational network interventions. These are selected from a number of possible options with the aim to reduce the negative impact of the failure over a pre-specified time horizon. The intervention options considered here include isolation valve manipulations, changing the pressure reducing valve’s (PRV) outlet pressure and installation and use of temporary overland bypasses from a nearby hydrant(s) in an adjacent, unaffected part of the network. The optimal mix of interventions is identified by using a multi-objective optimization approach driven by the minimization of the negative impact on the consumers and the minimization of the corresponding number of operational interventions (which acts as a surrogate for operational costs). The negative impact of a failure event was quantified here as a volume of water undelivered to consumers and was estimated by using a newly developed pressure-driven model (PDM) based hydraulic solver. The PDM based hydraulic solver was validated on a number of benchmark and real-life networks under different flow conditions. The results obtained clearly demonstrate its advantages when compared to a number of existing methods. The key advantages include the simplicity of its implementation and the ability to predict network pressures and flows in a consistently accurate, numerically stable and computationally efficient manner under both pressure-deficient and normal-flow conditions and in both steady-state and extended period simulations. The new real-time operational response methodology was applied to a real world water distribution network of D-Town. The results obtained demonstrate the effectiveness of the proposed methodology in identifying the Pareto optimal network type intervention strategies that could be ultimately presented to the control room operator for making a suitable decision in near real-time.

The selection of mortar for conservation of historic and heritage buildings can be challenging. Achieving compatibility with the historic fabric, durability and efficient use of materials within a practical timeframe often requires the use of hydraulic lime-based mortars which set more rapidly than the more traditional air lime mortars. These are considered to be more compatible with historic fabric than cement-based mortars, although, due to the modern production techniques and their natural variability, a deeper knowledge of their chemical and physical properties is needed to minimise damage due to incompatibility and make the decision process easier and safer. Natural hydraulic lime (NHL) binders are currently classified under EN 459-1:2015 in three designations, NHL 2, NHL 3.5 and NHL5, with the suffix representing the minimum compressive strength (in MPa) of a standard mortar mix at 28 days. The performance of NHL binders, manufactured by burning a naturally impure limestone, can be difficult to predict due to the inherent variability of both their physical and chemical characteristics. At the same time, the tolerance values for each classification allow for binders with significantly compressive strength differences to be classified by the same designation. The main aim of this research was to study a range of NHL binders, understand and quantify the variability of their characteristics and to establish how these properties influence the performance of mortars cured under standard and simulated weather conditions. In the first stage of the project, a selection of NHL binders from different origins and distinct designation were rigorously examined through physical, chemical and mineralogical characterisation to elucidate surface area, particle size distribution, oxide composition and crystalline phase composition. The characteristics of the binders were found to vary greatly, particularly amongst binders from the same classification and distinct origins, and in one particular case even from the same origin. A change of properties over time was also identified, binders manufactured in different years could have very different properties, even though, as far as could be ascertained from the packaging, it was the same product. Starting from a selection of 11 NHLs and 1 hydrated lime, the next step involved the manufacture of mortar samples using a sand aggregate appropriate for a conservation mortar with 1:2 ratio (binder:aggregate by volume). Sufficient water was added to produce a spread by flow table of 165 ± 10 mm. These mortars were cured under standard conditions and for a smaller group of binders under simulated weather conditions. For the standard cure conditions, the properties of the binders were compared to the physical properties in terms of strength (from 7 to 1080 days), porosity, capillary water absorption, water vapour permeability and freeze-thaw resistance of mortars made with the binders. The carbonation was also studied by phenolphthalein stain after all the flexural strength tests and after 2 years by XRD. The mortars under climate simulation were studied in terms of mechanical properties (up to 360 days) and carbonation. For comparison purposes, cement-lime (1:1:6 and 1:2:9 cement:lime:aggregate volumetric ratio), lime-metakaolin (MK) (with MK addition of 5, 10 and 20% of the lime mass) and lime putty mortars were manufactured to the same workability as the NHL mortars. These were studied in terms of strength up to 360 days, porosity and water absorption by capillarity action. The strength of the studied mortars does not follow the classification of the binders, with one binder, specified as NHL 2, resulting in a stronger mortar than another binder specified as NHL 5, and one NHL 3.5 mortar surpassing all the other mortars in terms of mechanical strength. The mechanical strength was found to correlate with the hydraulic phases, alite and belite, identified within the binders. The relative long-term performance of the mortars manufactured with the different binders can therefore be predicted based on the mineral properties rather than the standard classification. Pore related properties, such as water vapour permeability and water absorption by capillarity, were found to be related to the water/binder ratio of the NHL mortars. Later in the project, using the standard cured mortars data, a model was developed to predict compressive strength based on the proportion of crystalline phases present in the mortars, the surface area and the water/binder ratio. This model, applied to the studied mortars, was found to predict, with low error, the measured performance of the mortars, meaning that the model can be used as tool to predict mortar strength. The outcomes of this thesis demonstrated that with sufficient knowledge of the underlying chemistry of NHL binders, it is possible to establish the relative performance of mortars, thus making the decision on which binder to use easier and safer for the historic fabric.

Seabed gouging by ice, also known as ice scouring, is a common feature of the Arctic and sub Arctic regions of the planet as well as in Antarctica. It is a phenomenon which occurs when ice moves while in contact with the seabed. Ice gouging is of economical significance due to the probability of disruption of seabed structures such as subsea pipelines. Small scale laboratory tests were conducted at 1g to investigate the scour force produced when a scaled iceberg model scours a test bed in dry and saturated conditions. The tests were conducted for a range of scour depths, scour widths, frontal rake angles and soil conditions to study the performance of a rigid indenter (iceberg keel) scouring a test bed. Furthermore, the tests were also conducted at various speeds to study the rate effect in sand. The effect of the drifting speed on the drag force was found to be important: a sandy seabed scoured by an iceberg with a mean drifting speed of 0.1 m/s can generate scour loads twice as large as the static loads. The methods used currently to predict ice scour loads consider only the static loads under drained conditions and these should be revised. The PIV (Particle Image Velocimetry) technique was utilized to study the sub-gouge deformation and the soil failure mechanism associated with ice gouging. The soil resistance and the sub gouge deformation results obtained in the laboratory were compared with centrifuge investigations (the PRISE and PIRAM programs) in order to examine the viability of extrapolating the results from the model scale to a prototype scale.

O transiente hidráulico na adução de Pequenas Centrais Hidrelétricas (PCH) é um importante tema de estudo para a correta escolha e dimensionamento das estruturas de amortecimento, que protegem o circuito adutor da variação abrupta da pressão interna causada pela variação na velocidade média do escoamento. Este trabalho propõe a utilização da hidrodinâmica computacional (CFD) para avaliar a oscilação de massa ou nível d\'água no interior da estrutura de proteção denominada Chaminé de Equilíbrio e comparar esses resultados com a solução teórica analítica, com os critérios de dimensionamento definidos em norma técnica e com os dados de medições experimentais. O problema foi modelado no software Ansys CFX que utiliza o método dos volumes finitos (MVF) para resolver os campos de velocidade e pressão do escoamento, tanto em regime permanente, quanto em regime transiente. Aplicando-se o modelo CFD proposto ao circuito hidráulico didático, os resultados foram validados com a formulação teórica e com as medições experimentais do nível d\'água na chaminé. O modelo foi aplicado num estudo de caso que consiste num circuito hidráulico de PCH. Os resultados da oscilação de massa no interior da chaminé para este estudo de caso confirmam o emprego da fluidodinâmica computacional para avaliar tais escoamentos em regime transitório. / The hydraulic transient in the circuit of Small Hydroelectric Power Plants (SHP) is an important study subject for the correct choice and design of damping structure which protects the conduit from an abrupt variation of the internal pressure caused by the changes in the average flow velocity. This work proposes the use of computational fluid dynamics (CFD) to evaluate the mass oscillation or water level inside the protection structure called Surge Tank and compares these results with theory, design criteria defined by the technical standard, and experimental data. The problem was modeling in Ansys CFX software that uses the finite volume method (FVM) to solve velocity and pressure flow fields, in both steady state and transient flow. The proposed model used in a hydraulic circuit led to results validation with experimental measurements of water level surge tank. For the proposed case study, the response of the numerical CFD simulation confirms the use of computational fluid dynamics to evaluate such transient flows.

Orientador: Fernando de Azevedo Silva / Banca: João Zangrandi Filho / Banca: Anselmo Monteiro Ilkiu / Resumo: O objetivo desta pesquisa foi apresentar um método de análise para a solução de problemas de contato por elementos finitos e o cálculo dos componentes mecânicos que fazem parte do projeto de uma roda de aço. Foi demonstrado passo a passo à execução de um cálculo analítico da roda de uma Comporta Vagão, calculando as cargas radial e axial presentes, verificando a tensão de Hertz, as tensões localizadas no cubo da roda, as tensões atuantes no eixo e a seleção do rolamento. Após esta etapa retiraram-se os parâmetros iniciais que foram utilizados no cálculo analítico para buscar o cálculo por elementos finitos. Assim, foram definidas as superfícies de contato, a geometria dos componentes, a malha, as restrições, as condições de contorno e por fim a análise dos resultados encontrados antes e após o projeto. Dessa maneira, podê-se determinar com precisão os pontos de concentração de tensão para eliminação as falhas. Com a busca deste cálculo, usando a análise por Elementos Finitos, respaldado pelo cálculo analítico, foi possível reduzir o diâmetro da roda para obter ganhos em massa para este tipo de equipamento que usa uma quantidade alta de rodas (acima de 100 rodas), sem alteração da confiabilidade das peças durante seu funcionamento. / Abstract: The objective of this research was to present an analysis method for the solution of contact problems for finite elements and the calculation of the mechanical components that are part of the project of a steel wheel. It was demonstrated step by step the execution of an analytic calculation of the wheel of a Fixed Wheel Gate, calculating the radial and axial loads present, verifying the Hertz pressure, the concentrated stresses in the hub of the wheel, the acting stresses in the shaft and hub, and the selection of the roller bearings. After this stage the initial parameters were used in the analytic calculation in order to obtain the calculation using finite elements. After this, it was defined the contact surfaces, the geometry of the components, the mesh, the restrictions, the real constraints and finally the analysis of the results found before and after the project. After this calculation, the other analysis was established in order to determine the points of concentration stress accurately for the elimination of fails. In order to obtain the calculation, using Finite Elements, by the analytic calculation it was possible to reduce the diameter of the wheel to obtain earn in mass for this type of equipment that uses a high amount of wheels (above 100 wheels), without alteration of the reliability of the parts during the operation. / Mestre