The scientific development has the aim of executing experiments with such quality so that obtained results are accepted and recommended. The quality of an experiment can be described through statistics, ascribing it reliance on the obtained data, being the experimental error one of the parameters the evaluations of such quality. Therefore, the present work aimed at verifying the interference of different positions of seedlings on polystyrene trays on the normality and homogeneity of the errors in experiments carried out in different environments of lettuce production. Six experiments were carried out in the Departamento de Fitotecnia da Universidade Federal de Santa Maria in the summer (2006-2007) and fall winter (2007) with the lettuce crop, cultivar Vera two with seedling production in a greenhouse and four with the production of plants, two in a plastic structure and two in the field, the treatments were defined by 12 different positions of the seedlings on the tray of production. The analyzed variables in the seedling production and after their transplant to the environments were the fresh and dry phytomass of leaves and roots (g) and the fresh and dry phytomass of leaves (g), respectively. Presuppositions of errors normality and homogeneity were tested being compared to the averages of each treatment, as well as the conjoint analysis of the experiments. The different positions of the lettuce seedlings on the production trays did not significantly interfere in the experimental error estimative in the carried out experiments in plastic greenhouse and in the field. The central positions of seedlings was contributive in the final media of the lecttuce plants in greenhouse and in the field. In the summer experiments, the recommendations should be carried out individually to each environment, while in the fall winter experiments they can be generalized to both production environments. / O desenvolvimento da ciência é atingido pela execução de experimentos com qualidade, para que os resultados gerados sejam aceitos e recomendados. A qualidade de um experimento pode ser descrita através de estatísticas, sendo o erro experimental um dos parâmetros de avaliação dessa qualidade. Assim, o objetivo desse trabalho foi verificar a interferência de diferentes posições de mudas em bandeja de poliestireno sobre a normalidade e homogeneidade dos erros em experimentos realizados em dois ambientes de produção de alface. Foram conduzidos seis experimentos no Departamento de Fitotecnia da Universidade Federal de Santa Maria nas estações verão (2006-2007) e outono-inverno (2007) com a cultura da alface, cultivar Vera, dois com produção de mudas em casa de vegetação e quatro com produção de plantas, sendo dois em estufa plástica e dois a campo. Os tratamentos foram 12 diferentes posições das mudas na bandeja de produção. As variáveis analisadas na produção de mudas e após o transplante aos ambientes foram a fitomassa fresca e seca da parte aérea e raiz (g) e fitomassa fresca e seca da parte aérea (g), respectivamente. Foram testadas as pressuposições da homogeneidade dos tratamentos e normalidade dos erros sendo comparadas as médias de cada tratamento, bem como realizada a análise conjunta dos experimentos. Não houve violação das pressuposições de homogeneidade das variâncias dos tratamentos e normalidade dos erros nos experimentos com diferentes posições de mudas na bandeja de produção, nos diferentes ambientes e épocas de cultivo para a cultura do alface. As diferentes posições das mudas de alface nas bandejas de produção não interferiram na estimativa do erro experimental nos experimentos conduzidos em estufa plástica e em campo. As posições centrais das mudas das repetições contribuíram na produtividade final das plantas de alface em estufa plástica e em campo. Em experimentos com a cultura da alface conduzidos no verão, as recomendações das diferentes posições das plântulas para o cultivo devem ser realizadas individualmente em estufa e em campo, enquanto que no outono-inverno as mesmas podem ser generalizadas para os dois ambientes de produção.

A lot of technologically relevant incompressible fluids exhibit a substantial variaton of viscosity with the pressure;a falling cylinder viscometer is frequently used for the measurements, determining the viscosity indirectly from the time it takes the sinker to fall a given distance. The relation between the sinker fall velocity and the fluid viscosity is, however, derived under the constant viscosity assumption. The objective of the present thesis is to perform a numerical simulation of the viscometric experiment, assuming an explicit form of the pressure-viscosity dependence and realistic parameter values and to quantitatively assess the difference in body motion to the Navier- Stokes model. The computational method proposed, handling both the nonlinear constitutive relation and the body motion, was tested on simple problems with analytical solutions. The semianalytical relation for the Navier-Stokes model, also re-derived here, is compared to the computational results.The validity of the assumptions used in the theoretical derivation, based on the results of the numerical simulation, is discussed regarding the geometry of the viscometer. Powered by TCPDF (www.tcpdf.org)

Made available in DSpace on 2016-12-08T16:44:40Z (GMT). No. of bitstreams: 1 PGPV11MA091.pdf: 445481 bytes, checksum: 675ce53ec70abe958734cda2b68bc41b (MD5) Previous issue date: 2011-05-20 / Considering the gradient of nutrients in soil depth and water availability is very important the presence of a wide and well distributed root system in soil layers. Thus, the genotypes creation with well developed root system is interest to bean breeding programs. The lower cooking time for common bean is other characteristic desired and relevant to breeding programs, because it is very important to acceptance of a cultivar. However, availability of genetic variability is required for the action of breeder, which may be generated by induced mutation. This study aimed: i) to evaluate the effect of induced mutation in the root distribution of common bean in the soil profile; ii) to identify the correlation among the character root distribution and the others agronomic traits in mutant common bean populations; iii) to verify the existence of variation in the method used to determine the bean cooking time. Two experiments were conducted at Universidade do Estado de Santa Catarina - UDESC (Santa Catarina State University - Brazil). In the first experiment, the mutant populations Pérola, IAPAR 81, IPR Chopim and IPR Uirapuru in M1, M3 and M4 generations in level of 0, 100 and 200 Gy, were conducted in field in a randomized block design with two replications. The root system was assessed by profile method at 0-10, 10-20 and 20-30 cm soil layers. Plant height, height of first pod, stem diameter, number of pods per plant, number of grains per pod, weight of 100 grains and grain yield were analyzed to end of the crop cycle. In the second experiment, the grains from thirty six mutant common bean populations were subjected to the cooking test using the Mattson cooker apparatus, in a randomized block design with two blocks and two replications each experimental unit. The data were analyzed using three statistics models with information related to experimental error, to the sampling error and considering the replications as a variation factor. Three types of error were used in the statistics analysis, being the total error, experimental error and sampling error. The induced mutation provided variability and increased the root distribution of bean populations. The trait root distribution showed positive correlation with grain yield. Different results were obtained with use of three types of error for the cooking time data. The technique used to determination of the bean cooking time showed intrinsic variation to the method, being necessary to estimate the sampling error to purify the experimental error. Besides the need of criterion in the specification of statistics models and choice of appropriate error for testing the hypotheses / Considerando o gradiente de nutrientes em profundidade do solo e a disponibilidade hídrica, é de suma importância a presença de um amplo e bem distribuído sistema radicular nas camadas do solo. Deste modo, a criação de genótipos com sistema radicular bem desenvolvido é de interesse aos programas de melhoramento de feijão. Outra característica almejada e de relevância aos programas de melhoramento é o reduzido tempo de cocção dos grãos de feijão, pois a mesma é crucial para a aceitação de uma cultivar. No entanto, para a atuação do melhorista é necessário disponibilidade de variabilidade genética, podendo esta ser gerada por meio da mutação induzida. Neste sentido, os objetivos deste trabalho foram: i) avaliar o efeito da mutação induzida na distribuição do sistema radicular de feijão no perfil do solo; ii) identificar a existência de correlação entre o caráter distribuição radicular e as demais características de interesse agronômico em populações mutantes de feijão; iii) verificar a ocorrência de variação no método empregado para a determinação do tempo de cocção de feijão. Para tanto, foram realizados dois experimentos na Universidade do Estado de Santa Catarina (UDESC), no Centro de Ciências Agroveterinárias (CAV). No primeiro experimento, as populações mutantes Pérola, IAPAR 81, IPR Chopim e IPR Uirapuru nas gerações M1, M3 e M4 e nas doses 0, 100 e 200 Gy foram conduzidas a campo em delineamento de blocos casualizados com duas repetições. A avaliação do sistema radicular foi realizada pelo método do perfil nas camadas de 0-10, 10-20 e 20-30 cm do solo. Os demais caracteres foram mensurados ao término do ciclo da cultura, sendo eles: estatura de planta, altura de inserção do primeiro legume, diâmetro do caule, número de legumes por planta, número de grãos por legume, massa de 100 grãos e rendimento de grãos. No segundo experimento foi realizado o teste de cocção de grãos advindos de 36 populações mutantes de feijão por meio do aparelho cozedor de Mattson, em delineamento de blocos casualizados, com dois blocos e duas repetições por unidade experimental. Os dados foram analisados considerando no modelo estatístico informações sobre o erro experimental, o erro de amostragem e as repetições como um fator de variação. Três diferentes resíduos foram empregados nas análises, sendo, o erro total, o erro experimental e o erro de amostragem. A mutação induzida foi eficiente em proporcionar variabilidade e incremento na distribuição radicular das populações de feijão. O caráter distribuição radicular e rendimentos de grãos apresentaram correlação positiva. Resultados discrepantes foram obtidos com o uso dos diferentes resíduos para os dados referentes ao tempo de cocção. A técnica utilizada para a determinação do tempo de cocção apresentou variação intrínseca ao método, evidenciando a necessidade de estimar o erro de amostragem para evitar que o mesmo componha o erro experimental, além da necessidade de rigor nas especificações dos modelos estatísticos e na escolha do resíduo apropriado para testar as hipóteses

<p>In computational science it is common to describe dynamic systems by mathematical models in forms of differential or integral equations. These models may contain parameters that have to be computed for the model to be complete. For the special type of ordinary differential equations studied in this thesis, the resulting parameter estimation problem is a separable nonlinear least squares problem with equality constraints. This problem can be solved by iteration, but due to complicated computations of derivatives and the existence of several local minima, so called short-cut methods may be an alternative. These methods are based on simplified versions of the original problem. An algorithm, called the modified Kaufman algorithm, is proposed and it takes the separability into account. Moreover, different kinds of discretizations and formulations of the optimization problem are discussed as well as the effect of ill-conditioning.</p><p>Computation of parameters often includes as a part solution of linear system of equations <i>Ax = b</i>. The corresponding pseudoinverse solution depends on the properties of the matrix <i>A</i> and vector <i>b</i>. The singular value decomposition of <i>A</i> can then be used to construct error propagation matrices and by use of these it is possible to investigate how changes in the input data affect the solution <i>x</i>. Theoretical error bounds based on condition numbers indicate the worst case but the use of experimental error analysis makes it possible to also have information about the effect of a more limited amount of perturbations and in that sense be more realistic. It is shown how the effect of perturbations can be analyzed by a semi-experimental analysis. The analysis combines the theory of the error propagation matrices with an experimental error analysis based on randomly generated perturbations that takes the structure of <i>A</i> into account</p>

In computational science it is common to describe dynamic systems by mathematical models in forms of differential or integral equations. These models may contain parameters that have to be computed for the model to be complete. For the special type of ordinary differential equations studied in this thesis, the resulting parameter estimation problem is a separable nonlinear least squares problem with equality constraints. This problem can be solved by iteration, but due to complicated computations of derivatives and the existence of several local minima, so called short-cut methods may be an alternative. These methods are based on simplified versions of the original problem. An algorithm, called the modified Kaufman algorithm, is proposed and it takes the separability into account. Moreover, different kinds of discretizations and formulations of the optimization problem are discussed as well as the effect of ill-conditioning. Computation of parameters often includes as a part solution of linear system of equations Ax = b. The corresponding pseudoinverse solution depends on the properties of the matrix A and vector b. The singular value decomposition of A can then be used to construct error propagation matrices and by use of these it is possible to investigate how changes in the input data affect the solution x. Theoretical error bounds based on condition numbers indicate the worst case but the use of experimental error analysis makes it possible to also have information about the effect of a more limited amount of perturbations and in that sense be more realistic. It is shown how the effect of perturbations can be analyzed by a semi-experimental analysis. The analysis combines the theory of the error propagation matrices with an experimental error analysis based on randomly generated perturbations that takes the structure of A into account