Made available in DSpace on 2016-06-02T19:56:33Z (GMT). No. of bitstreams: 1 2364.pdf: 1102875 bytes, checksum: b267e8f11265cae314ce120c5c408adf (MD5) Previous issue date: 2008-10-30 / The yield ethanol (hP) is one of the most important parameters to be evaluated in the industrial process of alcohol fermentation. Because most plants neither present a reasonable level of automation, nor perform frequent analytical measures during the batch, the hP is currently calculated by methodology based on by-products formed in the process. Such calculation is questioned as to its accuracy because it almost always presents high values of hP. Still, with regard to the process of alcohol fermentation, the kinetic modeling of the fermentation process has been studied extensively. However, the proposed models have not properly represented the industrial process in which the fermenter initially containing the inoculum ( preculture ) is fed with a flow of wort (feeding culture medium) until its filling, following the fermentation, in batch, until the substrate exhaustion. The aim of the present study is propose a new methodology for calculating the yield ethanol (hP) based on material balances. The hP results obtained in the pilot plant (80.5 to 87.9%) and industrial plant (84.2 and 92.1%) presented by the new method tracks near values. However, values of hP obtained by the present methodology were lower than those obtained by the traditional by-products method, currently used by almost all sugar and alcohol mills. Still, a kinetic model was proposed to describe the conventional process of fermentation considering the existence of two cellular performances during the process, with regard to the consumption of substrate and to the production of ethanol. The model properly described the performance of the concentration of cells (CX), substrate (CS) and ethanol (CP) over all cultures, showing that the assumptions considered when preparing the model were concerned about the reality of the phenomenon observed. Finally, estimated values of kinetic parameters of the proposed model showed little variation in different cultures as well as values within a range of literature. / O rendimento em etanol (hP) é um dos parâmetros mais importantes a ser avaliado no processo industrial de fermentação alcoólica. Devido ao fato da maioria das usinas não apresentar um nível de automação razoável, nem realizar medidas analíticas freqüentes durante as bateladas, o hP é atualmente calculado por metodologia baseada nos subprodutos formados no processo, cálculo questionado quanto a sua precisão, pois gera quase sempre altos valores de hP. Ainda, no que diz respeito ao processo de fermentação alcoólica, a modelagem cinética do processo fermentativo tem sido estudada extensivamente. No entanto, os modelos propostos não têm representado adequadamente o processo industrial no qual a dorna inicialmente contendo o inóculo ( pé de cuba ) é alimentada com uma vazão de mosto até seu enchimento, seguindo a fermentação, em batelada, até o término do substrato. O presente trabalho teve como objetivo principal a proposta de uma nova metodologia de cálculo do rendimento em etanol (hP) com base nos princípios de balanço material. Os resultados de hP obtidos na planta piloto (80,5 a 87,9%) e na industrial (84,2 e 92,1%) pela nova metodologia apresentaram faixas próximas de valores. No entanto, estes valores obtidos de hP foram inferiores àqueles obtidos pela metodologia tradicional por subprodutos, utilizada atualmente pela quase totalidade das usinas de açúcar e álcool. Ainda, foi proposto um modelo cinético para descrever o processo convencional de fermentação alcoólica considerando a existência de dois comportamentos celulares ao longo do processo, no que diz respeito ao consumo de substrato e à produção de etanol. O modelo descreveu adequadamente os comportamentos das concentrações de células (CX), substrato (CS) e de etanol (CP) ao longo de todos os cultivos, mostrando que as hipóteses consideradas na elaboração do modelo foram pertinentes quanto à realidade do fenômeno observado. Por fim, os coeficientes cinéticos estimados do modelo proposto apresentaram uma pequena variação nos diferentes cultivos, bem como valores dentro de uma faixa de literatura.

Klimatförändringar har direkt påverkan på glaciärers tillväxt och avsmältning. Genom att studera glaciärers massbalans kan framtida riskbedömningar, för bland annat havsytans stigning, uppskattas. Skillnaden i glaciärens ackumulation och ablation ger den så kallade nettobalansen, som avslöjar om glaciären ökat eller minskat i massa under det gångna året. Storglaciären belägen i Tarfaladalen, öster om Kebnekaise, är en av världens mest välstuderade glaciärer. Här används den så kallade skandinaviska metoden för dokumentering av glaciärens massbalans. Metoden utgörs till stor del av manuella sonderingar i form av punktmätningar direkt på glaciärens yta. Då Storglaciären har upp mot 300 mätpunkter är inhämtningen av data både omfattande och tidskrävande. I detta kandidatarbete utfördes ett försök till optimering för den skandinaviska metoden på Storglaciären, med syftet att förhoppningsvis underlätta och effektivisera metoden. Punkter för mätdata avlägsnades för att undersöka om de resterande mätpunkterna gav ett godkänt värde för nettobalansen. För godkänt resultat av experimentet skall optimeringen ge ett representativt värde för storglaciärens nettobalans uppskattad med hela datamängden. Beräkningarna för nettobalansen är beroende av snöns densitet på glaciären, därav utfördes även densitetsmätningar i fält vid Tarfalas forskningsstation. Experimentet gav resultatet att metoden kan optimeras genom att var fjärde punkt för mätdata avlägsnades. Ytterligare studier kan dock utföras för att utforma mer specifika mönster för mätpunkter, som då skulle kunna ge ett ännu säkrare resultat för Storglaciärens nettobalans. / Climate changes has a direct impact on glaciers’ growth and melting. By studying the mass balance of glaciers, future risk assessments can be estimated by for example sea level rise. The difference between the glacier’s accumulation and ablation gives the net balance, which reveals if the glacier has increased or decreased in mass during the past year. Storglaciären, located in the Tarfala valley east of Kebnekaise, is one of the world's most studied glaciers. The Scandinavian method is used for documentations of the glacier’s mass balance. The method consists of manual probings at specific points on the glacier’s surface. Since Storglaciären has up to three hundred measurement points, the collection of the mass balance-data is both comprehensive and time consuming. For this bachelor project, an experiment was made to optimize the Scandinavian method at Storglaciären. The project was carried out with the aim of hopefully facilitating the method and to make it more efficient. Measurement points on the glacier were removed to investigate if the remaining points gave an approved value for the net balance. The requirement for the experiment to be approved is that the optimization shall provide a representative value for the net balance of Storglaciären, with all its measurement points included. The calculations for the net balance is dependent on the snow density on the glacier’s surface. Fieldwork at Tarfala research station was thereby carried out to collect data for the density on Storglaciären. The experiment gave the result that the Scandinavian method can be optimized. If every fourth measurement point on Storglaciären is removed, the glacier’s net balance is still representative. Further studies can be conducted to design specific patterns for removal of specific measurement points, which probably would provide even more accurate results for Storglaciären’s net balance.

During the past few years a vast amount of research has been done to increase the understanding of the complex ecosystem of the Baltic Sea. Modelling and simulations are important tools to increase knowledge of the system. A suitable model must be simple to use and the parameters and variables needed in the model must be easy to access. In this paper a dynamical mass-balance model, CoastMab, which is validated for smaller coastal areas, has been used to predict concentrations and transports of phosphorus in three large coastal areas - the Gulf of Finland, the Gulf of Gdansk and the Gulf of Riga. CoastMab uses ordinary differential equations to regulate inflow, outflow and internal flows. To reflect seasonal variations in temperature and different types of flows the model has a temporal resolution of a month. The main purposes of this paper have been to evaluate CoastMab, predict the concentrations and the transports of phosphorus in each coastal area and to analyse how much and why the results differ from empirical data. The performed simulations show the importance of a correct calculation of the wave base. This is due to its influence on the division between surface and deep water as well as the division of areas of accumulation and areas of erosion and transport. The retention times of surface and deep water calculated by the model also is of great importance because of their direct influence on the flow of water between the study area and the sea outside the area. This together with the concentration of phosphorus determines the flow of phosphorus in and out of the study area. The dynamical mass-balance model is considered to work well in the three studied areas even though the areas are outside the model domain in several respects. Even if the predictions of the phosphorus concentration in the gulf of Riga was less accurate than in the other study areas, the prediction was inside the interval of standard deviation that has been calculated from empirical data. The major flow of phosphorus in the model was in the cross section between the Baltic proper and the modelled coastal area. Since the model is easy to use and requires only a small number of obligatory input variables it would be appealing to continue development of the model to handle coastal areas of the same size and larger than the study areas in this paper. / Under senare år har mycket forskning utförts för att öka förståelsen av det komplexa ekosystem Östersjön utgör. Viktiga redskap för att öka förståelsen för systemet är modellering och simulering. För att en modell ska vara lämplig att använda är det viktigt att den inte är för komplicerad och att de parametrar och variabler som används i modellen är lättillgängliga. I detta arbete har en dynamisk massbalansmodell, CoastMab, som är validerad för mindre kustområden använts för att modellera fosforkoncentrationer och fosforflöden i tre större kustområden - Finska viken, Gdanskbukten och Rigabukten. CoastMab reglerar genom ordinära differentialekvationer inflöden, utflöden och interna flöden. För att kunna hantera säsongsvariationer i temperatur och olika typer av flöden har modellen en temporär upplösning på en månad. Syftet med arbetet har främst varit att se hur väl CoastMab predikterar fosforkoncentrationen och fosforflöden inom respektive område, samt att analysera hur mycket och varför prediktionerna skiljer sig från empiriska data. De simuleringar som utfördes visade att djupet på den teoretiska vågbasen, som modellen beräknar, är av stor betydelse eftersom den i sin tur bestämmer fördelningen yt- och djupvatten samt fördelningen av ackumulationsbottnar och erosions- och transportbottnar. De utbytestider för yt- och djupvatten som modellen beräknar är också av största vikt eftersom de har en direkt inverkan på vattenflödet mellan studerat område och havet utanför, vilket i sin tur tillsammans med fosforkoncentrationen bestämmer fosforflöden in och ut ur det studerade området. Den dynamiska massbalansmodellen kan anses fungera väl för samtliga tre studerade områden, trots att de ligger utanför modellens domän i flera avseenden. Även om prediktionen av fosforkoncentrationen i Rigabukten blev sämre än för de två övriga områdena, låg prediktionen inom det intervall för standardavvikelsen som beräknats utifrån empiriska data. De största fosforflödena i modellen förekom för samtliga områden i gränssnittet mellan egentliga Östersjön och det modellerade kustområdet. Enkelheten i att använda modellen och det låga antalet obligatoriska drivvariabler gör att den är väl värd att utveckla för att hantera områden av samma eller större storlek än de nu studerade områdena.

O estudo teve como objetivos a quantificação da interceptação das chuvas pelas copas das árvores e a estimativa da evapotranspiração real pelos métodos dos balanços hídrico e de cloreto. Para isso, estabeleceu-se o monitoramento de duas microbacias hidrográficas experimentais localizadas na Floresta Ombrófila Densa (Mata Atlântica), uma no Parque Estadual das Fontes do Ipiranga (PEFI) e a outra no Parque Estadual da Serra do Mar - Núcleo Cunha, ambas no Estado de São Paulo. As interceptações estimadas para o IAG-PEFI e para a microbacia B-Cunha foram de 21,6% e 24,0%, respectivamente. Os resultados das análises de regressão linear simples, entre a precipitação e a transprecipitação apresentaram coeficientes de determinação elevados para o IAG-PEFI (92,0%) e Cunha (88,6%), indicando que a variável independente explica maior proporção da variação da variável dependente. Para a interceptação foi aplicado o teste de correlação de Spearman (rs), sendo de 0,5449 para o IAG-PEFI e de 0,5243 para Cunha, indicando uma moderada correlação positiva, significativa para o nível de 5%. A pequena diferença encontrada na interceptação das chuvas pelas florestas para as duas localidades pode ser explicada pelas aberturas dos dosséis, no IAG-PEFI de 34,2% e Cunha de 31,1%. No ano hídrico de 2000 na microbacia do IAG-PEFI, a precipitação e o deflúvio totais foram de 1.279,6 mm e 273,4 mm. A evapotranspiração real alcançou 1.006,2 mm, representando 78,6% da precipitação total, indicando uma baixa produção hídrica (21,4%). Para os anos hídricos de 2001, 2002 e 2003 na microbacia B-Cunha, as evapotranspirações reais alcançaram 696,4 mm (42,8%), 769,5 mm (47,7%) e 748,4 mm (44,0%), com média de 738,1 mm (44,8%), indicando uma boa produção hídrica média (55,2%). Os fluxos totais de cloreto via deposição úmida nas microbacias do IAG-PEFI e B-Cunha foram de 7,26 e 7,57 kg.ha-1.ano-1, enquanto os fluxos totais através do rio alcançou 7,15 kg.ha-1.ano-1 e 7,96 kg.ha- 1.ano-1. Os resultados mostraram que os fluxos de entrada do íon foram muito próximos, indicando que o processo de urbanização que o entorno do PEFI sofreu não está influenciando nas precipitações. Na microbacia do IAG-PEFI a evapotranspiração real foi de 1.000,1 mm, correspondendo a 78,2% da precipitação total, valor este muito próximo ao método do balanço hídrico. Na microbacia BCunha a evapotranspiração real para os anos hídricos de 2001, 2002 e 2003 foi de 898,6 mm (55,2%), 696,7 mm (43,2%) e 693,7 mm (40,8%), e a média para os três anos de 763,0 mm (46,3%), valor ligeiramente superior ao obtido pelo balanço hídrico. Para as duas microbacias, os fluxos das entradas e das saídas do cloreto foram praticamente iguais. Então, podemos inferir que o íon não é um produto do intemperismo, e que seu envolvimento em troca iônica ou processos biológicos atingiu o equilíbrio. Portanto, esse íon pode ser considerado conservativo. A estimativa da evapotranspiração real pelo escoamento de base do mês de junho de 2000 na microbacia do IAG-PEFI, através do método do cloreto apresentou valor de 989,0 mm (77,3%). Já, na microbacia B-Cunha para agosto do ano hídrico de 2001, a evapotranspiração real foi de 450,5 mm (27,7%), valor muito abaixo do obtido pelos outros métodos, possivelmente pelo pequeno número de amostras realizadas. Para os meses de agosto e julho dos anos de 2002 e 2003 os resultados foram satisfatórios, alcançando 707,8 mm (43,9%) e 719,4 mm (42,3%). / The present study aims the quantification of the interception of the rains by the trees canopy and the determination of the real evapotranspiration by the methods of the water and the chloride balances. For that, it was established the monitory of two experimental catchments situated at Dense Ombrófila Forest (Atlantic Forest) one of them, in Parque Estadual das Fontes do Ipiranga (PEFI) and the other in Parque Estadual da Serra do Mar- Cunha Nucleus both in São Paulo State. The interceptions estimated for the IAG-PEFI and for the B-Cunha catchments were of 21.6% and 24.0%, at the same time. According to the D`Agostino-Pearson test for the transformed values, of the precipitation and the transprecipitation it was verified normal distribution at the level of significance 5%. The analysis results of simple linear regression presented coefficient of elevated determination for the IAG-PEFI (92.0%) and Cunha (88.6%). The analysis through the test of D`Agostino-Pearson for the absolute values and transformed from interception do not followed normal distribution. Thus, it was applied the test of correlation of Spearman, being of 0.545 for the IAG-PEFI and the 0.524 for Cunha showing a positive moderate correlation for the level of 5% significant. The small difference found in the interception of the rains through the forests for the two locations might be explained by the canopy openings in IAG-PEFI of 34.2% and Cunha of 31.1%. During the water year of 2000 in the catchment of IAG-PEFI, the precipitation and the total streamflow were of 1,279.6 mm and 273.4 mm. The real evapotranspiration reached 1,006.2 mm, representing 78.6% from total precipitation, indicating a low water production (21.4%) For the water years of 2001, 2002 and 2003 in the B-Cunha catchments, the real evapotranspiration reached 696.4 mm (42.8%), 769.5 mm (47.7%) and 748.4 mm (44.0%), with average of 738.1 mm (44.8%), indicating a good water production (55.2%). The chloride by wet deposition total flux in the IAG-PEFI and B-Cunha catchments were of 7.26 kg.ha-¹.year-1 and 7.57 kg.ha-¹.year-¹ and while the total flux through the streamflow were 7.15 kg.ha ¹.year-¹ and 7.96 kg.ha-¹.year-¹. The results showed that the ion enter flux were much closed, indicating that the urbanization process suffered by PEFI spilled is not influencing in the precipitations. In the IAGPEFI catchment, the estimated real evapotranspiration was of 1,000.1 mm, that correspond the 78.2% of the total precipitation a much closed value to the water balance. In the B-Cunha catchment the estimated real evapotranspiration for the water years of 2001, 2002 and 2003 was of 898.6 mm (55.2%), 693.7 mm (43.2%) and 693.7 mm (40.8%), and the medium for the three years of 763.0 mm (46.3%), value a little superior to that obtained in the water balance. For the two catchments, the input and output chloride flux were basically the same. Then, we could infer that the ion is not a product of intemperism, and that its involvement in ionic change, or biologic process reach the balance, however, this ion may be consider conservative. The estimative of real evapotranspiration by the base flow for June 2000 in the IAG-PEFI catchment, through the chloride method presented the value of 989.0 mm (77.3%). But in the B-Cunha catchment for August of the water year of 2001, the real evapotranspiration was of 450.5 mm (27.7%), value much lower obtained by the other methods, possible by the small number of sample realized. For the months of August and July of the 2002 and 2003 water years, the results were satisfactory, reaching 707.8 mm (43.9%) and 719.4 mm (43.3%).

[pt] O mercúrio usado no garimpo do ouro na região amazônica é liberado para a atmosfera, solo e rios. Uma vez na atmosfera, o metal é oxidado e imediatamente se deposita. Na água, ocorre a transformação para o metilmercúrio principalmente pela ação de microrganismos. A formação do metilmercúrio aumenta a dispersão e biodisponibilidade do elemento no ambiente aquático. O metilmercúrio pode ser incorporado pelo plancton entrando, assim, na cadeia alimentar. A concentração do metal aumenta a medida que se ascende nos níveis tróficos da cadeia, atingindo os valores mais elevados em peixes carnívoros como o tucunaré. Dessa forma, as emissões de mercúrio provocam a contaminação dos recursos naturais e aumentam os riscos para a saúde dos consumidores habituais de pescado. O objetivo deste trabalho é testar um modelo de bioacumulação de mercúrio em peixes para estimar concentrações em predadores de topo da cadeia alimentar. O modelo pode ser usado como ferramenta para a gestão ambiental de ecossistemas aquáticos potencialmente contaminados com mercúrio. Este objetivo foi atingido através da combinação dos modelos de balanço de massa de Trudel e bioenergético de Wisconsin, aplicados em espécies tropicais do gênero Cichla (tucunaré) da bacia do rio Tapajós. O modelo bioenergético de Wisconsin foi usado para determinar as taxas de consumo de alimento a partir de dados de crescimento estimados. Os parâmetros usados nos modelos foram obtidos na literatura. A habilidade da modelagem na predição dos teores de mercúrio em tucunarés foi avaliada através da comparação com dados de campo obtidos nos anos 1992 e 2001, no rio Tapajós e no sistema de lagos Maicá. Os melhores resultados foram alcançados para os espécimes coletados no ano 1992, o que parece estar relacionado com uma melhor estimativa da concentração de mercúrio no alimento neste ano. / [en] The mercury used in the artisanal mining (garimpo) of gold in the Amazon region is emitted to the atmosphere, soil and rivers. Once in the atmosphere, the metal is oxidized and immediately deposited. In the water, the transformation to methylmercury takes place mostly due to the action of microorganisms. The formation of methylmercury increases the dispersion and bioavailability of the element in the aquatic environment. The methylmercury can be assimilated by the plankton and enters the food chain. The concentration of the metal increases further up in the trophic levels of the chain and reaches the highest values in carnivorous fishes like tucunaré. Thus, the mercury emissions cause the contamination of natural resources and enhance the risks to the health of regular fish consumers. The objective of this work is to test a model for bioaccumulation of mercury in fishes. The model calculates concentrations in top predators of the food chain and can be used as a tool for the management of aquatic ecosystems potentially contaminated with mercury. This objective was accomplished through the combination of the Trudel mass balance and Wisconsin bioenergetics models, applied to tropical species of the Cichla (tucunaré) genre of the Tapajós river basin. The Wisconsin bioenergetics model was utilized to determine the rates of food consumption from estimated fish growth data. The parameters used in the models were found in the literature. The ability of the models to predict mercury contents in tucunarés was evaluated by comparison with the field data, obtained in 1992 and 2001, in the Tapajós River and in the Maicá lakes. The best results were attain ed for specimens collected in 1992 what seems to be related to a better estimate of mercury concentration in the food in this year.

Varmare klimat är något som kontinuerligt påverkar glaciärer planeten över och har under det senaste århundrandet vidtagit en global ökning av negativa massbalanser. Denna trend kan bidra till variationer i havsnivån och orsaka problem med översvämningar över hela världen. Övervakning och observation av världens glaciärer är därför väldigt viktigt och genom att skapa modeller som tillåter insyn i glaciärernas respons till atmosfärisk fluktuation går det att åstadkomma en djupare förståelse för hur den globala uppvärmningen kommer att utvecklas. I detta projektet återskapas massbalansen för Brewster Glacier, Nya Zeeland, för perioden 2005 – 2015 genom tillämpning av graddagar. Modellen har försetts med data från ett nyligen genomfört nedskalningsprojekt som har producerat atmosfäriska data för perioden 1979 – 2015 vilket tillåter modellen att köras. Resultaten som presenteras kommer att bidra till större inblick i graddags-modellen om huruvida tillvägagångssättet att använda enkel temperatur- och nederbördsdata är tillförlitligt för att porträttera massbalansen för glaciärer. / Warmer climate is something that is continuously affecting glaciers across the planet and has during the last century gained a global increase in negative mass balances. This trend has contributed to sea-level rise and had other impacts on water resources. Monitoring and observing the worlds glaciers is therefore very important and by creating models that allows insight in the glacier response to atmospheric fluctuation, it is possible to obtain deeper knowledge on how the global warming will develop. In this project, the mass balance of Brewster Glacier, New Zealand, is reconstructed for the period 2005 – 2015 using a degree-day approach. The input for the model has been obtained from a recent downscaling project that has produced atmospheric data over the period 1979 – 2015 that allows the model to run at a daily temporal resolution. The results presented are going to contribute to a greater insight in the degree-day model as to whether the approach of using simple air temperature and precipitation data are reliable to portray the mass balance of glaciers.

L’objectif de notre étude est de modéliser un réservoir géothermique. Si nous supposons que le réservoir géothermique n’est composé que d’eau pure le transfert de matière et d’énergie est classiquement décrit par deux équations de conservation : la conservation de la matière et la conservation de l’énergie. À ces deux équations vient s’ajouter la vitesse du fluide classiquement donnée par la loi de Darcy tandis que les propriétés thermodynamiques, obtenues grâce à des équations théoriques ou empiriques (les équations d’état), ferment le modèle mathématique. Dès lors, ce modèle fermé, il existe différents schémas de résolution. Le premier est de résoudre en pression et température puis de procéder à un changement de variables lors du passage de monophasique à diphasique ou de diphasique à monophasique. TOUGH2 utilise le couple pression-saturation dans la zone diphasique.La seconde approche est de résoudre en pression et enthalpie afin d’accroître la stabilité lors de la transition entre l’état monophasique et l’état diphasique (voir Hydrotherm). Nous avons adopté la seconde option, résoudre en pression et enthalpie. De plus la résolution spatiale est faite avec les volumes finis.La modélisation d’un réservoir géothermique fait intervenir des équations fortement dépendantes l’une de l’autre. Cependant nous avons fait le choix de découpler la résolution afin de se libérer de la complexité de la résolution du système couplé. En effet, cette méthode possède l’avantage d’être moins consommatrice de mémoire puisque nous travaillons toujours avec le même nombre de données, mais dans une matrice deux fois moins importante. Nous montrerons que cette méthode demeure suffisamment précise pour une utilisation aussi bien dans le domaine industriel que dans celui de la recherche.Nous offrons à l’utilisateur une grande liberté grâce à l’implémentation de plusieurs méthodes : Euler implicite, explicite, Runge-Kutta ou BDF2 pour les solveurs temporels ou GMRES et BICGSTAB pour les solveurs linéaires. Nous pouvons gérer des conditions aux limites très variées telles que des flux nuls (décrivant une frontière qui n’échange pas de matière avec l’extérieur) ou une condition mixte (un Dirichlet sur la pression et un Dirichlet ou condition « sortie libre » sur la température… Cette dernière situation décrit une zone de recharge ou de décharge. Nous avons développé un outil multilangage : Python,Fortran et C++ (une implémentation de l’IAPWS provenant du projet freesteam incluant la zone supercritique). Tous ces langages sont orientés objet. L’IAPWS est l’outil permettant de calculer les propriétés physiques inconnues et par conséquent il ferme le système.Enfin nous avons appliqué le modèle sur le bassin parisien, France, sur plusieurs systèmes 1D et un autre système 2D réalisés par Coumou avec la plateforme CSMP++. Le bassin parisien est un réservoir exploité pour produire de la chaleur par le biais du pompage d’une eau à 70 _C et réinjecté à 40 _C. Les simulations 1D permettent de visualiser le déplacement d’un front de chaleur en haute enthalpie. La simulation 2D montre la convection naturelle de l’eau dans une faille. Chaque simulation a été comparée aux résultats obtenus avec un autre code (CSMP++, HYDROTHERM ou TOUGH2) et les résultats sont en accord. / The purpose of our study is to model a geothermal reservoir. When geothermal reservoirs are assumed to be composed of pure water, the transfer of mass and energy is classically described by two balance equations: the mass balance equation and the energy balance equation. In addition to those equations, fluid velocity is classically given by the Darcy law while thermodynamic properties, inferred from theoretical or empirical equations of state, are used to close the mathematical system. Once this system is closed, there exist different solutions. The first one is to solve for pressure and temperature with a variable switch to saturation in the two-phase region (e.g. TOUGH2). The second one is to solve for pressure and enthalpy to increase the stability of phase transition between single and two-phase states (e.g. Hydrotherm). We adopted the second option. We solve the system by using a splitting method — to get rid of the complexity of coupling equations — and a finite volume method for the spatial discretization. We offer some freedom to users thanks to the implementation of several methods like explicit or implicit Euler, Runge-Kutta or BDF2 for time solvers or GMRES and BICGSTAB for the linear solver. We can handle several boundary conditions like no-flow — describing a boundary which cannot exchange matter withthe exterior — or like a mixed-therm condition — a Dirichlet condition to the pressure and a Dirichlet or an outflow condition to the temperature in order to describe a recharge or a discharge zone — …Selecting object-oriented languages, we developed a multi-language framework, combining Python, Fortran and a C++ implementation of IAPWS (from the freesteam project) including the supercritical equations. To close the system physical propertiesare determined by the IAPWS- IF97 thermodynamic formulation. We’ve applied this simulation model to the dogger in Paris, France, to several onedimensional systems and a two-dimensional one made by Coumou with the CSMP++platform. The dogger is a reservoir exploited to produce heat by pumping water at 70 _C and reinjecting it in the reservoir at 40 _C. In the one-dimensional systems we wanted to observe the process of heat transfer from a higher temperature boundary to a smaller one in a high-energy domain. The last simulation shows the natural convection of water in a fault. For every simulation we compared the solutions we found with another code (TOUGH2 or CSMP++) and they agreed.

In a particle based geometallurgical model, the behavior of the particles can be used for forecast the products and quantify the performance of the different ore types within a deposit. The particle tracking is an algorithm developed by Lamberg and Vianna 2007 whose aim is to balance the liberation data in a mineral processing circuit composed by several processing units. Currently, this tool is being developed for the HSC Chemistry software by Outotec.The objective of this study is to understand and evaluate the particle tracking algorithm in a geometallurgical test for iron ore. To achieve this objective, the liberation data is balanced in a Davis tube test circuit. A total of 13 samples from Leveäniemi iron ore were process in a Davis tube circuit.The magnetite is the main mineral in the Leveäniemi iron ore samples. Its high recovery in the Davis tube circuit along with the V, Ti and Mn suggest that these elements are present in the magnetite lattice. These penalty elements in the iron concentrates cannot be avoided at the stage of mineral concentrations.The washing effect of the Davis tubes controlled by the rotational and longitudinal agitation of the tube perturb the particles agglomeration between the pole tips of the electromagnet. A higher agitation frequency and amplitude will wash away most of the gangue minerals and also fine grained magnetite.In this work, the particle tracking is depicted and implemented in a magnetic separation circuit for high liberated material. The liberation data was balanced in a way that the particle classes can be followed through circuit and their recoveries can be calculated. Nevertheless, the algorithm requires further validation and analysis of its limitations in terms of resolution and reproducibility.

Abstract The objective of this thesis was to obtain new information about mechanisms of thermomechanical pulp refining in the inner area of a refiner disc gap by studying inter-fibre refining and by calculating the distribution of energy consumption in the refiner disc gap. The energy consumption of thermomechanical pulping process is very high although theoretically a small amount of energy is needed to create new fibre surfaces. Mechanisms of refining have been widely studied in order to understand the high energy consumption of the process, however, phenomena in the inner area of disc gap has had less attention. It is likely that this important position is causing high energy consumption due to the high residence time of pulp located there. The power distribution as a function of the refiner disc gap was calculated in this work. The calculation was based on mass and energy balances, as well as temperature and consistency profiles determined by mill trials. The power distribution was found to be dependent on segment geometry and the refining stage. However, in the first stage refiner with standard refiner segments, a notable amount of power was consumed in the inner area of the disc gap. Fibre-to-fibre refining is likely to be the most important mechanism in the inner area of disc gap from the point of view of energy consumption. In this work the inter-fibre refining was studied using equipment for shear and compression. Fibre-to-fibre refining was found to be an effective way to refine fibres from coarse pulp to separated, fibrillated and peeled fibres if frictional forces inside the compressed pulp were high enough. It was proposed that high energy of today’s thermomechanical pulping process could derive from too low frictional forces that heated pulp and evaporated water without any changes in fibre structure. The method to calculate power distribution and results of fibre-to-fibre refining experiments may give ideas for developing today’s thermomechanical pulp refiners’ or for developing totally new energy saving mechanical pulping processes.

Die vorliegende Arbeit beschäftigt sich mit rohstofflichen und verfahrenstechnischen Einflussfaktoren bei der Biomassepyrolyse. Ausgehend von der Entwicklung einer kleintechnischen Festbettpyrolyseapparatur, erfolgten experimentelle Untersuchungen an 26 biogenen Einsatzstoffen unter verschiedenen Prozessbedingungen. Die Apparatur erlaubt eine vollständige Bilanzierung und Gewinnung aller Produkte, zudem können Einflüsse durch sekundäre Reaktionen in der Gasphase minimiert werden. Die Einsatzstoffe, welche u. a. auch Zellstoff, Xylan und Alkali-Lignin einschließen, wurden hinsichtlich brennstofftechnischer und physikalischer Eigenschaften sowie der Stoffgruppenzusammensetzung charakterisiert. Sie repräsentieren eine große Bandbreite möglicher Zusammensetzungen. Bei den Prozessparametern wurde die Pyrolysetemperatur im Bereich von 200 bis 750 °C, die Aufheizrate zwischen 5 und 100 K/min, die Feststoffverweilzeit von 0 bis 30 min sowie die Partikelgröße (0 bis 5 mm) variiert. Aus den Untersuchungen zum Einfluss der Prozessparameter für die verschiedenen Einsatzstoffe wurden unter Anwendung einer geeigneten Bilanzierungsmethodik geschlossene Masse- und Elementbilanzen für jeden Versuchspunkt aufgestellt. Unter den Prozessvariablen konnte die Temperatur erwartungsgemäß als wichtigste Einflussgröße identifiziert werden. Der zweistufige Zersetzungsverlauf der Biomassen ermöglicht die mathematische Beschreibung der temperaturabhängigen Ausbeuten mittels der zweistufigen Boltzmann-Funktion für den gesamten Temperaturbereich mit hohen Bestimmtheitsmaßen. Die rohstofflichen Einflussgrößen wurden unter Anwendung der Rangkorrelationsmethode nach Spearman und der Produkt-Moment-Korrelation nach Pearson mit den definierten Zielgrößen (Ausbeuten, Produktzusammensetzung, Kokseigenschaften, Heizwerte, Energieeinbindung) bei verschiedenen Pyrolysetemperaturen korreliert. Neben der Stoffgruppenzusammensetzung konnten bei den rohstofflichen Einflussfaktoren die Gehalte an Alkalien sowie der Gesamtgehalt an potentiell katalytisch aktiven Bestandteilen (Na, K, Mg, Ca, Fe) als Haupteinflussgrößen identifiziert werden. Korrelationen ergeben sich auch für brennstofftechnische Eigenschaften, wobei neben dem Flüchtigen- und dem Aschegehalt, das O/C-Verhältnis signifikant ist. Die gefundenen statistischen Zusammenhänge können weitestgehend mechanistisch begründet werden. Zur Quantifizierung der ermittelten Zusammenhänge für die Zielgrößen wurden multiple Regressionsmodelle erstellt und anhand von Bestimmtheitsmaß, Informationskriterium und mittleren Modellfehlern bewertet. Somit konnten 42 Regressionsgleichungen für die Produktausbeuten bei verschiedenen Pyrolysetemperaturen entwickelt werden, die auf den Gehalten verschiedener Stoffgruppen und dem Gesamtgehalt an katalytisch aktiven Elementen basieren. Weitere 56 Regressionsgleichungen stehen für die Berechnung von Teer/Öl-Elementarzusammensetzung, Kokszusammensetzung, Teer/Öl-Heizwert sowie Energieeinbindung im Koks bei verschiedenen Pyrolysetemperaturen zur Verfügung. Die Prognoseeignung der Gleichungen wurde anhand eines weiteren Datensatzes für Apfeltrester überprüft. Für die Koks-, die Gas- und die Kondensatausbeute sowie die genannten Produkteigenschaften ergab sich eine gute Vorhersagequalität, die jedoch stark von der verwendeten Gleichung abhängt. Die Validierung mit Literaturdaten konnte aufgrund fehlender Datensätze, die sowohl die notwendigen Rohstoffparameter als auch Produktausbeuten und -eigenschaften enthalten, nur anhand der Koksausbeute erfolgen. Für verschiedene Biomassen und biogene Reststoffe führte dies zu einer guten Anpassung. Die mathematische Beschreibung der Ausbeuten und bestimmter Produkteigenschaften über Regressionsgleichungen auf Grundlage von Rohstoffparametern stellt einen vielversprechenden Ansatz für die Vorhersage der maximalen Ausbeuten bei bestimmten Bedingungen dar. Dies ermöglicht eine Abschätzung zur Einsatzeignung von Biomassen bzw. biogenen Reststoffen für verschiedene Anwendungszwecke. Bisher existiert kein derartiges Modell zur Vorhersage der definierten Zielgrößen. Grundsätzlich wäre die Entwicklung einfacher Gleichungen mit wenigen, einfach bestimmbaren und standardisierten Parametern erstrebenswert. Die Ergebnisse haben jedoch gezeigt, dass Ein-Variablen-Modelle die Trends zwischen den Biomassen aufgrund der komplexen Zusammenhänge zwischen Pyrolyseverhalten und Rohstoffparametern häufig nicht richtig wiedergeben können. Für robuste Modelle sind somit mindestens zwei unabhängige Modellparameter mit idealerweise gegensätzlichem Einfluss notwendig. / The intention of this work was an intensive study of the influence of feedstock properties and process variables on biomass pyrolysis. Due to a lack in consistent data sets, including various feedstock parameters as well as product yields, compositions, and further properties, a laboratory fixed bed reactor was developed to overcome this problem. The pyrolysis reactor was used for experiments with 26 biogenous feedstock under variable process conditions. The reactor is suitable to assure nearly closed mass balances and a complete product recovery. Furthermore, it allows the minimization of secondary reactions. The used feedstock, which include cellulose, xylan, and lignin amongst others, represent a broad range of possible compositions and were intensively characterized by determination of fuel and physical properties as well as biopolymer composition. The varied process parameters are: temperature between 200 and 700 °C, heating rate in the range of 5 to 100 K/min, solid residence time from 0 to 30 min, and particle size up to 5 mm. Closed mass and element balances were done for every set of parameters. As expected, amongst process variables the temperature was identified as the main factor influencing biomass pyrolysis. The temperature depending products yields could be fitted well by the double boltzmann approach due to the two-stage pyrolytic decomposition of biomass. Correlation of feedstock properties with different target parameters, including yields, product composition, heating values, remaining energy content in char, and char properties, was done by Spearman´s rank correlation and Pearson´s correlation for different temperatures. Biopolymer composition as well as alkaline content and total content of potential catalytic elements (Na, K, Ca, Mg, Fe) were identified as main factors influencing biomass pyrolysis product yields and compositions. Further correlations arise with fuel properties like volatile matter and ash content besides O/C atomic ratio. The obtained correlations can be mainly related to pyrolysis mechanisms. The received relationships were quantified by means of multiple regression models. Model evaluation was done by coefficient of determination, information criteria and mean squared errors. 42 regression models, based on different biopolymer contents and the total content of catalytic elements, were provided for the mathematical description of product yields for different process temperatures. Another 56 equations are suitable for the calculation of product properties like tar/oil and char composition, tar/oil heating value, and remaining energy content in the char at different temperatures. The predictability of the regression models was proved using another data set for apple pomace. The yields of char, gas, and condensate as well as the aforementioned product properties can be predicted very well, although, the predictability varies with the applied equation. Validation of the models by literature data was only possible for the char yield, because of the mentioned lack in suitable and complete data sets. Application of regression model to fixed bed char yields for different biomass and biogenous residues from literature resulted in a good predictability. Mathematical description of pyrolysis product yields and properties by means of regression models based on feedstock parameters is a promising approach to predict maximum yields at defined conditions and, therefore, to make an estimation of suitability of the biomass to different applications. Up to now such models do not exist. In general, the development of simple equations based on a few standardized parameters which are easy to determine is worthwhile. Hence, the results showed that the overall trend between different biomass feeds was often not predicted correctly using one-parameter models. This is due to the complex relationships between pyrolysis behavior and feedstock properties. Consequently, at least two parameter models, where the variables show the opposite trends, were most appropriate.