Orientador: José Paschoal Batistuti / Banca: Rubens Monti / Banca: Alice Yoshiko Tanaka / Resumo: Recentemente, a indústria de alimentos está preocupada em desenvolver novos produtos, usando ingredientes inovadores, que promovam benefícios à saúde. O uso de proteínas de leguminosas como uma alternativa para a proteína animal tem sido estudada. O tremoço branco (Lupinus albus L.) apresenta elevada concentração de proteína e óleo, principalmente ácidos graxos insaturados e poliinsaturados, fibra alimentar, açúcares e vitaminas B1 e B2. Por outro lado, o tremoço não pode ser consumido diretamente porque possui alcaloides, que além de serem tóxicos, atribuem gosto amargo ao grão. Como a maioria dos alcaloides é solúvel em água, a quantidade de alcaloide do grão pode ser diminuída através de água corrente, fervente ou salmoura. Sucos de frutas são bastante consumidos no Brasil, especialmente no verão, não somente pelo sabor, mas como fonte de hidratação e por serem ricos em nutrientes. Os sucos de frutas tropicais conquistam cada vez mais o mercado consumidor, sendo o Brasil, um dos principais produtores. A pitanga, fruto da pitangueira (Eugenia uniflora L.) é uma planta frutífera nativa do Brasil (das regiões Sul e Sudeste e que tem se adaptado favoravelmente às condições climáticas e edáficas da região Nordeste), da Argentina e do Uruguai. A pitanga é uma fruta cuja polpa apresenta excelentes condições para industrialização, devido ao seu alto rendimento, aroma agradável e sabor exótico. O objetivo do presente estudo foi o de verificar, através da metodologia de superfície de resposta, as melhores condições para a utilização simultânea de extrato aquoso de tremoço branco e de suco tropical de pitanga, no desenvolvimento de uma bebida. Para tanto, as variáveis independentes foram representadas pelo volume de extrato aquoso de tremoço (mL) e pelo volume de suco de pitanga... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Recently, the food industry is concerned about developing new products, using innovative ingredients that promote health benefits. The use of vegetable protein as an alternative to animal protein is being studied. The white lupine (Lupinus albus L.) has high concentrations of protein and oil, especially non saturated fatty acids and polynonsaturated fats, dietary fiber, sugars and vitamins B1 and B2. On the other hand, the lupine bean can't be consumed straight because it has alkaloids which besides being toxic, add a bitter taste to the bean. As most alkaloids are water soluble, the amount of alkaloid in the beans can be reduced by running water, boiling or brine. Fruit juices are widely consumed in Brazil, especially in the summer, not only for their flavor, but as a source of hydration and because they are rich in nutrients. The tropical fruit juices are conquering the market more and more, with Brazil being one of the main producers in the world. The Pitanga (Eugenia uniflora L.) is the fruit of the Pitangueira, a native plant of southern Brazil (which has adapted positively to weather and soil conditions in the country's Northeast), Argentina and Uruguay. Pitanga is a fruit which pulp presents excellent conditions for industrialization, due to its high yield, pleasant aroma and exotic flavor. The aim of the present study was to verify, through the response surface methodology, the optimum conditions for the combined use of aqueous extract of white lupine and tropical pitanga juice in the development of a drink. Therefore, the independent variables were represented by the volume of aqueous extract of lupine (mL) and by the volume of pitanga juice (mL). The dependent variables (responses) were obtained through sensory test of acceptance ("appearance", "aroma", "taste" and "overall acceptability"). The combined... (Complete abstract click electronic access below) / Mestre

O composto de lodo de esgoto, classificado como fertilizante orgânico Classe D (MAPA, 2009) é um material rico em nutrientes, especialmente nitrogênio (N) e fósforo (P), e possui elevado conteúdo de matéria orgânica. Contudo, ainda não há relatos na literatura brasileira sobre o uso do composto de lodo de esgoto na cultura da cana-de-açúcar. Assim como, faltam pesquisas para adequar as doses de adubos nitrogenado e fosfatado a serem aplicadas em área agrícola tratada com composto. O objetivo do presente trabalho foi avaliar os efeitos de aplicação e reaplicação de doses de composto de lodo de esgoto e de adubos minerais nitrogenado e fosfatado sobre as propriedades químicas do solo e a produtividade, características tecnológicas e os teores de N e P na folha, caldo e colmo da cana-soca de segundo e terceiro corte. Na aplicação do composto, segunda soca, foram aplicadas quatro doses de composto de lodo de esgoto (0; 27; 54 e 82 t ha-1, base seca), quatro de N (0; 32,3; 64,7 e 98 kg ha-1) e duas de P2O5 (0 e 30 kg ha-1) e na reaplicação, terceira soca, foram aplicadas quatro doses decrescentes do composto (0; 7,8; 15,5 e 23,3 t ha-1, base seca), quatro de N (0; 44,6; 89 e 135 kg ha-1) e duas de P2O5, (0 e 54 kg ha-1) em delineamento em blocos casualizados, em esquema fatorial, com três repetições. Na aplicação do composto, segunda soca, cujo requerimento de N foi considerado como 98 kg ha-1, a produtividade de colmos (TCH) aumentou de 77 a 104 t ha-1 em função das doses do composto. A aplicação do composto pelo critério do nitrogênio promoveu valores de TCH, produtividade de açúcar (TAH), açúcares redutores (AR), açúcar total recuperável (ATR) e açúcares redutores totais (ART) semelhantes aos do tratamento com adubo mineral convencional. Na reaplicação do composto, terceira soca, cujo requerimento de N foi considerado como 135 kg ha-1, a TCH aumentou de 59 a 80 t ha-1 em função das doses do composto. Porém, a maior dose do composto, cuja taxa de mineralização do N foi considerado de 30%, deve ser complementada com 89 kg ha-1 de N mineral para produzir valores de TCH, TAH, AR e ATR semelhantes aos do tratamento com adubo mineral. As doses do composto de lodo de esgoto resultou em incrementos nos teores de N total, C-orgânico, P, pH, Ca, K, soma de bases e capacidade de troca catiônica do solo, em relação ao controle absoluto. Quanto aos teores de N e P nas soqueiras, de modo geral, não houve efeito dos tratamentos, exceto quanto ao teor de N na terceira soca que foi crescente em função das doses de N mineral. Conclui-se, com base no estudo de superfície de resposta, que: 1) a aplicação do composto de lodo de esgoto no cultivo da cana-soca, considerando taxa de mineralização do N de 10% e o requerimento agronômico de N, pode substituir de forma adequada a aplicação dos adubos minerais nitrogenado e fosfatado, e 2) o uso de dose de composto inferior a da dose estabelecida pelo critério do N deve ser proporcionalmente complementada com nitrogênio mineral, seja em condições da primeira e, ou, da segunda aplicação sucessiva / The sewage sludge compost classified as an organic fertilizer Class D (MAPA, 2009) is a material rich in nutrients, especially nitrogen (N) and phosphorus (P), and have high content of organic matter. However, there is a lack of research in the Brazilian literature about the use of the sewage sludge compost on the sugar cane crop. There is also a lack of research to adequate the nitrogen and phosphate fertilizer doses to be applied in agricultural fields treated with compost. The objective of this study was to evaluated effect of levels of sewage sludge compost and mineral sources of nitrogen and phosphorus on the soil chemical proprieties, productivity and technological characteristics and the content of N and P on leaves, juice and stalks of the second and third ratoon cane. On the first application of compost, second ratoon cane, were applied four doses of sewage sludge compost (0; 27; 54 and 81.7 t ha-1, dry base), four of N (0; 32.3; 64.7 and 98 kg ha-1) and two of P2O5 (0 and 30 kg ha-1) and in the reapplication of the compost, third ratoon, the compost was applied in decreased levels 0; 7,8; 15,5 e 23,3 t ha-1, dry base, four of N (0; 44,6; 89 and 135 kg ha-1) and four of P2O5, (0 and 54 kg ha-1) in randomized block design, in factorial design, with three replications. In the first application, second ratoon, which N requirement was considered as 98 kg ha-1, the stalk productivity (TSS) increased from 77 to 104 t ha-1 in function of sewage sludge compost rate. The sewage sludge compost application rate established by nitrogen criteria promoted values of promoted values of stalk productivity (TSS), sugar productivity (TSH), reducing sugars (RS), total recoverable sugars (TRS) and total reducing sugars (TRS) similar to the treatment with mineral fertilizer. In reapplication of compost, whose N request was considered 135 kg ha-1, the TSS increased from 59 to 80 t ha-1 in function of compost application rate and the highest mineralization rate was 30%, should be complemented with 89 kg ha-1 of mineral N to produce values of TSS, TSH, AR, Pol cane, Pol juice and similar to conventional mineral fertilizer. The sewage sludge compost application rate increased N total, organic carbon, P, pH, Ca, K, bases sum and cation exchange capacity soil contents in relation to the control treatment. As for the contents of N and P in the ratoon, in general, there was no effect on the treatments, except for the N content in the third ratoon that increased in function of N fertilizer. Concludes, based on the response surface study that: 1) the application of sewage sludge compost in cultivation of ratoon cane, whereas N mineralization rate of 10% and the agronomic application of N, can replace the adequately the application of nitrogen and phosphorus mineral fertilizer; 2) the use of lower dose of sewage sludge compost established by nitrogen criteria should be proportionally supplemented with mineral nitrogen, either in terms of first and second successive application

[pt] Compostos de boro são utilizados na indústria metalúrgica, microeletrônica, de vidros, na agricultura, etc. Esse elemento é um micronutriente essencial no desenvolvimento de microrganismos, plantas, animais e humanos. No entanto, pode ser tóxico em grandes concentrações e por isso necessita ser removido de águas e efluentes. No Brasil, o limite padrão é de 0,5mg/L para águas doces de classe I e II, estabelecido pela resolução do Conama 357/2005. Por sua vez, a resolução do Conama 430/2011 estabelece um padrão de lançamento de efluentes de 5mg/L. A presente dissertação tem como objetivo geral estudar a remoção de boro por eletrocoagulação para o tratamento de águas e efluentes contendo boro, utilizando uma célula com um arranjo de 4 eletrodos de alumínio (2 catodos e 2 anodos) monopolares em paralelo. Os resultados obtidos mostram que o processo de eletrocoagulação é uma alternativa viável para a remoção de boro, e alcança eficiências em torno de 70 por cento (em pH inicial igual à 4; densidade de corrente igual à 18,75mA/cm2 e tempo de eletrólise igual à 90min). O modelo cinético que melhor descreve a remoção de boro foi o de pseudo-primeira ordem. O modelo de Langmuir se ajustou muito bem aos dados experimentais obtidos. O valor de qm obtido pelo modelo de Langmuir refletiu a elevada capacidade de adsorção máxima (qm é igual à 334mg/g). Através das análises por Microscopia Eletrônica de Varredura (MEV) e por Espectroscopia de Energia Dispersiva (EDS) na superfície dos eletrodos, observou-se a presença de corrosão do tipo localizada nos catodos e a corrosão do tipo uniforme nos anodos. A morfologia do lodo produzido no processo de eletrocoagulação foi analisada por MEV, indicando a presença de uma morfologia heterogênea na superfície, enquanto que a análise por Difração de Raios-X (DRX) apresentou picos largos característicos de um material amorfo e a fase de alumínio predominante foi a boehmita, AlO(OH), finalmente, através da análise por Espectroscopia por Perda de Energia de Elétrons (EELS) foi possível a detecção do boro no lodo, além da detecção de alumínio e de oxigênio. Diante dos resultados obtidos no estudo de otimização a partir da Metodologia de Superfície de Resposta (RSM) constatou-se que o desenvolvimento de um modelo matemático por análise de regressão possibilitou a avaliação do efeito das variáveis independentes (densidade de corrente, pH inicial e tempo de eletrólise) e as suas interações na remoção de boro. / [en] Boron compounds are used in the metallurgical industry, microelectronics, glassware, agriculture, etc. This element is an essential micronutrient in the development of microorganisms, plants, animals and humans. However, it can be toxic in high concentrations and therefore needs to be removed from water and effluent. In Brazil, the standard limit is 0.5mg/L for Class I and II freshwaters, established by Conama Resolution 357/2005. In turn, the Conama 430/2011 resolution establishes an effluent discharge standard of 5mg/L. The present dissertation aims to study the removal of boron by electrocoagulation for the treatment of water and effluents containing boron, using a cell with an arrangement of 4 monopolar aluminum electrodes (2 cathodes and 2 anodes) in parallel. The results show that the electrocoagulation process is a viable alternative for the removal of boron and reaches efficiencies around 70 per cent (at initial pH equals to 4, current density equals to 18.75mA/cm2 and electrolysis time equals to 90min). The kinetic model that best describes the removal of boron was pseudo-first order. The Langmuir model fitted very well to the experimental data obtained. The value of qm obtained by the Langmuir model reflected the high maximum adsorption capacity (qm equals to 334mg/g). Through the analysis by Scanning Electron Microscopy (SEM) and Energy Dispersive Spectroscopy (EDS) on the surface of the electrodes, it was observed the presence of pitting corrosion in the cathodes and uniform corrosion in the anodes. The morphology of the sludge produced in the electrocoagulation process was analyzed by SEM, indicating the presence of a heterogeneous surface morphology, while the X-ray diffraction (XRD) analysis showed broad peaks characteristic of an amorphous material and the predominant aluminum phase was boehmite, AlO(OH), finally, through the Electron Energy Loss Spectroscopy (EELS) analysis, it was possible to detect boron in the sludge, as well as aluminum and oxygen. In view of the results obtained in the optimization study from the Response Surface Methodology (RSM), it was verified that the development of a mathematical model by regression analysis made possible the evaluation of the effect of the independent variables (current density, initial pH and time of electrolysis) and their interactions in the removal of boron.

Diante do moderno paradigma de projeto otimizado de grandes sistemas de engenharia, este trabalho apresenta um ambiente de projeto que inclui uma série de inovações que permitem otimizar problemas multidisciplinares utilizando colaboração remota pela Internet. Muitos trabalhos têm sido desenvolvidos nesse sentido, o que motivou uma parceria internacional entre o Departamento de Engenharia Naval da Universidade de São Paulo e a Ecole Centrale de Nantes. O objetivo desta parceria foi unir a experiência brasileira em projetos de sistemas complexos de engenharia com a experiência francesa em otimização multidisciplinar. Sendo este trabalho fruto de uma colaboração internacional, é importante ter em mente a obrigação existente para que a pesquisa desenvolvida atenda a ambas as expectativas. O objetivo é, a um só tempo, viabilizar a integração de subsistemas conhecidos em um ambiente de projeto que permita a otimização multidisciplinar de sistemas complexos de engenharia e desenvolver contribuições pontuais relevantes no uso de superfícies de resposta, projeto distribuído e no desenvolvimento de estratégias de otimização multidisciplinares. O relatório apresentará uma descrição do problema, seguida da revisão bibliográfica sobre os principais assuntos envolvidos na pesquisa. A descrição do trabalho desenvolvido ocupará os dois próximos capítulos. No primeiro, o trabalho é apresentado privilegiando a visão global do sistema e a integração dos diferentes subsistemas desenvolvidos em um enfoque claramente influenciado pela tradicional visão sistêmica da Engenharia Naval. No capítulo seguinte as principais contribuições pontuais desenvolvidas são detalhadas ao estilo Francês. Por fim é apresentada a aplicação deste trabalho em diversos problemas matemáticos e de engenharia. / Facing the modern paradigm of optimized projects of large engineering systems, this research work presents a project environment that includes a series of innovations that allow the optimization of multidisciplinary problems using remote collaboration through the internet. Many research works have been done around this subject, which have motivated an international partnership between the Departamento de Engenharia Naval da Universidade de Sao Paulo and Ecole Centrale de Nantes. The objective of this partnership was mix the Brazilian expertise in design of engineering systems with the French expertise in multidisciplinary optimization. Being the result of an international collaboration, it is important to keep in mind the existing compromise to fulfill both expectations. The objective is altogether to make the integration of known systems viable in a project environment that allows multidisciplinary optimization of complex engineering systems with the creation of focused contributions that are relevant on the use of response surfaces, design collaborative and on the development of strategies of multidisciplinary optimization. The report will present a description of the problem, followed by a bibliographic revision about the main issues involved in the research. The description of the work is covered on the two following chapters. On the first one, the work approaches a global view of the system and the integration of different developed sub-systems under a perspective clearly influenced by the traditional systemic vision of Naval Engineering. On the next chapter, the main punctual contributions are detailed in the French style. Finally, the application of this work is presented in several mathematical and engineering problems.

A degradação oxidativa dos lipídeos é um fator preponderante que limita a vida útil dos alimentos, reduzindo suas qualidades nutricionais e organolépticas. A utilização de antioxidantes durante o processo de fabricação pode diminuir consideravelmente a extensão da oxidação lipídica, desencadeada pela reação de radicais livres. A crescente preocupação pelo consumo de alimentos mais saudáveis e livres de aditivos sintéticos tem recebido maior atenção pelas indústrias, que atualmente tem procurado por alternativas mais naturais, como o uso de antioxidantes provenientes de especiarias, como uma forma promissora de substituição dos compostos sintéticos. O objetivo deste trabalho foi estudar as condições de obtenção de extratos hidroalcoólicos de cravo (Syzygium aromaticum), canela (Cinnamomum verum) e pimenta dedo-de-moça (Capsicum annuum) e sua aplicação em óleos de soja refinado, armazenado em embalagens de vidro, em comparação com amostras adicionadas de antioxidante sintético TBHQ e amostras livres de antioxidantes (controle), durante sete meses, simulando a comercialização em supermercados. Foram verificados também as propriedades antioxidantes dos extratos de especiarias produzidos nas condições selecionadas através das metodologias de DPPH e ABTS. O estudo do efeito da temperatura de extração e do grau de hidratação do etanol sobre o rendimento de compostos fenólicos totais dos extratos destas especiarias foi realizado com aplicação de planejamento experimental e metodologia de superfícies de resposta. As condições para uma extração eficiente dos compostos fenólicos do cravo, da canela e da pimenta foram o emprego de etanol a 49,5 % (v/v). As melhores temperaturas estiveram entre 45 a 50 °C para o cravo e canela, e 60 °C para a pimenta. Os extratos de cravo, canela e pimenta dedo-de-moça preparados nas condições otimizadas, foram adicionados nas proporções de 50 mg kg-1, 100 mg kg-1, 150 mg kg-1 e 200 mg kg-1 às amostras de óleo de soja refinado e submetido a um teste acelerado em estufa a 65 °C juntamente com amostras adicionadas nas mesmas proporções de TBHQ e amostras isentas de antioxidantes (controle), para verificação da concentração mais eficiente no retardar a oxidação. Em seguida, após os resultados do teste de estufa, os óleos adicionados de 100 mg kg-1 de extratos, TBHQ, e isento de antioxidantes foram acondicionados em embalagens de vidro transparente de 250 mL e armazenados sob condições de comercialização (incidência de luz fluorescente por 15 horas / 24 horas e 331,71 Lux). Os resultados de Índice de Peróxido, Acidez e Absortividade na faixa do ultravioleta evidenciaram que os extratos foram capazes de oferecer proteção antioxidante ao óleo de soja refinado em diferentes níveis, sendo que o tratamento adicionado de extrato de cravo foi o que mais se destacou na inibição da oxidação lipídica do óleo de soja, seguido do extrato adicionado de canela e pimenta (p<0,05). / Lipid deterioration is a major cause of shorter food shelf life, reducing its nutritional and organoleptical quality. The use of antioxidants during the manufacturing process can significantly decrease the extent of lipid oxidation reactions triggered by free radicals. The growing concern about the consumption of healthier foods and free from synthetic additives has received increased attention by the industry, which is currently looking for more natural alternatives, such as the use of antioxidants from herbs and spices, as a promising replacement of synthetic compounds. The objective of this work was to study the conditions for obtaining hydroalcoholic extracts of cloves (Syzygium aromaticum), cinnamon (Cinnamomum verum) and red pepper (Capsicum annuum) and its application in soybean oil refined, stored in glass containers, compared with samples added of synthetic antioxidant TBHQ and samples free antioxidant (control) during 7 months simulating marketing conditions in supermarkets. Extracts were analyzed by DPPH and ABTS methods. The study of the effect of extraction temperature and the degree of hydration of ethanol on the yield of total phenolic compounds of the extracts of these spices was carried out with application of experimental design and response surface methodology. The best solvent concentration for efficient extraction of phenolic compounds from clove, cinnamon and pepper was 49.5% ethanol (v / v). The best temperatures were between 45 and 50°C for cloves and cinnamon, and 60 ° C for pepper. The extracts of clove, cinnamon and red pepper prepared under optimized conditions were added in concentration of 50 mg kg-1 , 100 mg kg-1, 150 mg kg-1 and 200 mg kg-1 to samples of refined soybean oil and subjected to an accelerated storage test at 65°C with samples containing the same concentrations of TBHQ and samples free of antioxidant (control) in order to check the most efficient concentrations to slow the oxidation. The best extracts concentration 100 mg kg-1 from the Schaal oven test was added to oil packed in glass containers against oil with 100 mg kg-1 of TBHQ, and oils free of antioxidants and stored under normal commercialization conditions (fluorescence light exposure for 15 hours / 24 hours and 333,71 Lux). Peroxide, acid values and absorptivity in the ultraviolet range showed that the extracts were able to offer antioxidant protection to soybean oil at different levels. Clove extract, showed the best performance in the inhibition soybean oil oxidation, followed by cinnamon and pepper (p<0,05).

Uncertainty quantification involves sampling the reservoir parameters correctly from a posterior probability function that is conditioned to both static and dynamic data. Rigorous sampling methods like Markov Chain Monte Carlo (MCMC) are known to sample from the distribution but can be computationally prohibitive for high resolution reservoir models. Approximate sampling methods are more efficient but less rigorous for nonlinear inverse problems. There is a need for an efficient and rigorous approach to uncertainty quantification for the nonlinear inverse problems. First, we propose a two-stage MCMC approach using sensitivities for quantifying uncertainty in history matching geological models. In the first stage, we compute the acceptance probability for a proposed change in reservoir parameters based on a linearized approximation to flow simulation in a small neighborhood of the previously computed dynamic data. In the second stage, those proposals that passed a selected criterion of the first stage are assessed by running full flow simulations to assure the rigorousness. Second, we propose a two-stage MCMC approach using response surface models for quantifying uncertainty. The formulation allows us to history match three-phase flow simultaneously. The built response exists independently of expensive flow simulation, and provides efficient samples for the reservoir simulation and MCMC in the second stage. Third, we propose a two-stage MCMC approach using upscaling and non-parametric regressions for quantifying uncertainty. A coarse grid model acts as a surrogate for the fine grid model by flow-based upscaling. The response correction of the coarse-scale model is performed by error modeling via the non-parametric regression to approximate the response of the computationally expensive fine-scale model. Our proposed two-stage sampling approaches are computationally efficient and rigorous with a significantly higher acceptance rate compared to traditional MCMC algorithms. Finally, we developed a coarsening algorithm to determine an optimal reservoir simulation grid by grouping fine scale layers in such a way that the heterogeneity measure of a defined static property is minimized within the layers. The optimal number of layers is then selected based on a statistical analysis. The power and utility of our approaches have been demonstrated using both synthetic and field examples.

Technology scaling has been the most obvious choice of designers and chip manufacturing companies to improve the performance of analog and digital circuits. With the ever shrinking technological node, process variations can no longer be ignored and play a significant role in determining the performance of nanoscaled devices. By choosing a worst case design methodology, circuit designers have been very munificent with the design parameters chosen, often manifesting in pessimistic designs with significant area overheads. Significant work has been done in estimating the impact of intra-die process variations on circuit performance, pertinently, noise margin and standby leakage power, for fixed transistor channel dimensions. However, for an optimal, high yield, SRAM cell design, it is absolutely imperative to analyze the impact of process variations at every design point, especially, since the distribution of process variations is a statistically varying parameter and has an inverse correlation with the area of the MOS transistor. Furthermore, the first order analytical models used for optimization of SRAM memories are not as accurate and the impact of voltage and its inclusion as an input, along with other design parameters, is often ignored. In this thesis, the performance parameters of a nano-scaled 6-T SRAM cell are modeled as an accurate, yield aware, empirical polynomial predictor, in the presence of intra-die process variations. The estimated empirical models are used in a constrained non-linear, robust optimization framework to design an SRAM cell, for a 45 nm CMOS technology, having optimal performance, according to bounds specified for the circuit performance parameters, with the objective of minimizing on-chip area. This statistically aware technique provides a more realistic design methodology to study the trade off between performance parameters of the SRAM. Furthermore, a dual optimization approach is followed by considering SRAM power supply and wordline voltages as additional input parameters, to simultaneously tune the design parameters, ensuring a high yield and considerable area reduction. In addition, the cell level optimization framework is extended to the system level optimization of caches, under both cell level and system level performance constraints.

In industrial processes decreasing variation is very important while achieving the targets. For manufacturers, finding out optimal settings of product and process parameters that are capable of producing desired results under great conditions is crucial. In most cases, the quality response is measured on a continuous scale. However, in some cases, the desired quality response may be qualitative (categorical). There are many effective methods to design robust products/process through industrial experimentation when the response variable is continuous. But methods proposed so far in the literature for robust design with categorical response variables have various limitations. This study offers a simple and effective method for the analysis of categorical response data for robust product or process design. This method handles both location and dispersion effects to explore robust settings in an effective way. The method is illustrated on two cases: A foam molding process design and an iron-casting process design.

In this study, the effects of bioprocess parameters on glucose isomerase roduction by Bacillus thermoantarcticus (DSMZ 9572) were investigated. For his purpose, firstly, in laboratory scale shake-bioreactors, a defined medium was designed in terms of its carbon and nitrogen sources, to achieve the highest glucose isomerase activity. Optimum concentrations of medium components were determined by a statistical approach, namely Response Surface Methodology (RSM). RSM defined the relationship between the response, glucose isomerase activity and the independent variables, medium components. The highest glucose isomerase volumetric activity was obtained as 1630 U L-1 in the optimized medium containing 10.64 kg m-3 xylan, 5.66 kg m-3 yeast extract, 5.92 kg m-3 (NH4)2SO4, 0.25 kg m-3 MgSO4.7H2O, 0.001 kg m-3 FeSO4.7H2O, 0.001 kg m-3 ZnSO4.7H2O , 0.000075 kg m-3 MnSO4. H2O, and 0.00001 kg m-3 CuSO4.5H2O at conditions: pH0= 6.0, T=55&deg / C, N=200 min-1. Then by using the designed medium, pH and oxygen transfer conditions of the bioprocess were investigated in 3.0 dm3 pilot scale bioreactor. The effect of pH was analyzed at pH=6 uncontrolled and controlled operations with the following conditions: air inlet rate of Q0/VR=0.5 vvm and agitation rate of N=500 min-1. The effects of oxygen transfer parameters were examined, at pH=6 controlled condition, at an air inlet rate of Q0/VR=0.5 vvm and the agitation rates of N=300,500,750 min-1. The variations in glucose isomerase activity, cell, amino acid and organic acid concentrations with the cell cultivation time, specific cell growth rate, the oxygen uptake rate, the liquid phase coefficient by using the dynamic method, maintenance coefficient for oxygen and yield coefficients were determined. The highest glucose isomerase volumetric activity and cell concentration, 1838 U L-1 and 2.26 kg m-3 at the 24th of the cell cultivation were attained at 0.5 vvm, 500 min 1 and pH=6 uncontrolled operation.

The flapping path of a single airfoil and dual airfoils in a biplane configuration is optimized for maximum thrust and/or propulsive efficiency. Unsteady, low speed viscous flows are computed using a Navier-Stokes solver in a parallel computing environment. A gradient based algorithm and Response Surface Methodology (RSM) are employed for optimization. The evaluation of gradient vector components and the design of experiments for RSM, which require unsteady solutions, are also carried out in parallel. Parallel computations are performed using Parallel Virtual Machine (PVM) library. First, a single airfoil undergoing a combined sinusoidal or non-sinusoidal pitching and plunging motion is studied. The non-sinusoidal flapping motion is described using an elliptic curve or Non-Uniform Rational B-Splines (NURBS). It is shown that the thrust generation may significantly be increased in comparison to the sinusoidal flapping motion. For a high thrust, the airfoil stays at high effective angle of attack values during the upstroke and the downstroke, and the effective pitching occurs at minimum and maximum plunge positions. Secondly, the optimization of sinusoidal and non-sinusoidal flapping paths of dual airfoils is considered. Moving and deforming overset grids are used for computations. The deforming overset grids remove the restrictions on the flapping motion, and improve the optimization results obtained earlier. At low flapping frequencies, an airfoil in a biplane configuration produces more thrust than a single airfoil. Yet, at high frequencies the airfoil in biplane configuration produces less thrust at a significantly lower efficiency than the single airfoil.