This project investigates the use of a truss network, a structural mechanics model, as a metaphor for adapting a computational fluid dynamics (CFD) mesh. The objective of such adaptation is to increase computational effi- ciency by reducing the numerical error. To drive the adaptation, or to give the scheme an understanding of accuracy, computational errors are translated into forces at mesh vertices via a so-called monitor function. The ball-vertex truss network method is employed as it offers robustness and is applicable to problems in both two and three dimensions. In support of establishing a state-of-the-art adaptive meshing tool, boundary vertices are allowed to slide along geometric boundaries in an automated manner. This is achieved via feature identification followed by the construction of 3rd order bezier surface patches over boundary faces. To investigate the ability of the scheme, three numerical test cases were investigated. The first comprised an analytical case, with the aim of qualitatively assessing the ability to cluster vertices according to gradient. The developed scheme proved successful in doing this. Next, compressible transonic flow cases were considered in 2D and 3D. In both cases, the computed coefficient of lift and moment were investigated on the unrefined and refined meshes and then compared for error reduction. Improvements in accuracy of at least 60% were guaranteed, even on coarse meshes. This is viewed as a marked achievement in the sphere of robust and industrially viable r-refinement schemes.

Geodetic measurements with a gyro theodolite are an important technique for establishing control networks in tunnelling projects. The observations are used as a separate control point outside the main network while checking the angle of the tunnels direction as it is difficult to connect the underground network to a superior network outside the tunnel.  This study aims to investigate the usefulness of gyro surveys in tunnel networks and evaluate how much they improve the positional accuracy of underground control networks. The study compares a total of four network simulations (with and without gyro observations) that have been divided into 12 sections. The sections are then compared to each other to determine the influence of gyro observations on the positional uncertainty in the network. Gyro observations were found to contribute to significant improvements in measurement data. Without them the tunnel would have an uncertainty greater than five centimetres within approximately 1000 m. The results provide valuable insights into the impact of gyro measurements on geodetic control network in tunnels and how they can be used to improve measurement data. / Gyromätningar och gyroteodoliter har idag en viktig roll på grund av de höga krav på noggrannhet som ställs inom många av dagens tunnelprojekt vid stomnätsmätning i tunnlar. Gyroobservationerna används för att få en utomstående kontroll av stomnätet då detta på grund av tunnelns utformning är problematiskt att koppla mot ett överordnat nät. Det finns dock inte mycket skrivet om hur dessa bör användas samt hur stor effekt mätningarna har på ett vanligt tunnelbygge idag. Studiens syfte är att undersöka gyromätningar användbarhet i tunnelnät samt utvärdera hur mycket dessa förbättrar noggrannheten på kortare stomnät i tunnelmiljö. I rapporten har sammanlagt fyra nätsimuleringar för en tunnelbana i Stockholm jämförts med varandra där två tillhör byggnätet och två bruksnätet (med och utan gyroobservationer). Dessa har delats upp i totalt 12 sektioner varav 6 stycken innehåller gyroobservationer och 6 är utan gyroobservationer. Sektionerna har sedan jämförts med varandra för att ta reda på hur stor inverkan gyroobservationerna har på osäkerheten i näten samt om det är möjligt att nå upp till en teoretisk osäkerhet i nätet på max fem centimeter. Resultatet visar att den radiella medelosäkerheten minskade från 1,69 cm till 0,93 cm i sektion 1, 1,24 cm till 0,64 cm i sektion 2 och 7,13 cm till 1,64 cm i sektion 3. Vid genomslaget mellan sektion 1 och 2 minskade osäkerheten från 5,10 cm till 1,14 cm i sektion 1 och i sektion 2 från 4,41 cm till 1,48 cm. Sträckan på fackverksnätet innan osäkerheten översteg fem centimeter blev 794m i sektion 1 och 846 m i sektion 3. I sektion 2 översteg aldrig osäkerheten fem centimeter trots att längden på fackverket var 971 m. Sektion 1 och 2 i bruksnätet visade ingen betydande förändring beroende på om gyroobservationer användes eller inte. I sektion 3 minskade dock den radiella medelosäkerheten med 2,29 cm vilket kan förklara genom att stomnätet inte har något anslutningsnät i slutet på denna sektion. Generellt visades gyroobservationerna bidra till stora förbättringar av mätdata i byggnätet och utan dessa skulle tunneln inom ca. 1000 m ha en osäkerhet större än fem centimeter. Resultatet ger en bra inblick i hur stor påverkan gyromätningar har på stomnät i tunnlar samt hur dessa kan användas för att förbättra mätvärden.

Made available in DSpace on 2017-07-10T18:13:23Z (GMT). No. of bitstreams: 1 Adilson Rodrigues.pdf: 576092 bytes, checksum: 4540b2c17f817db56489da230508a7d5 (MD5) Previous issue date: 2010-07-30 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The aim was to detect morphometric differences in the stages of larval development Hemisorubim platyrhynchos and Pseudoplatystoma corruscans and allometric growth in these two fish species of ecological and economic relevance in South America Larvae were separated into six stages of development ontogenetic larval (yolk sac larval, preflexion, initial of flexion, flexion, end of flexion and post flexion). That allowed Landmarks grouping them in three series of geometric distances. Was applied to PCA for each series of distances between different stages of larval development of both species to examine the morphometric patterns (similar or not). The factorial analysis of variance (species and stage of development) was used above retained axes of PCA for interpretation of data and test the possible differences. To verify the existence of allometric pattern in species, the scores generated by the first axis of PCA was used to estimate the multivariate allometric coefficients. The series I from PCA segregated (H. platyrhynchos and P. corruscans) the species and stages (larval yolk and pre-flexion). In the series II H. platyrhynchos showed no differences between the larval stages, but for P. corruscans was marked development from the flexion stage to the late flexion allowing a separation of H. platyrhynchos. In post-flexion stage (series III) morphometric discrimination between species was not evident. In relation to the allometric growth, the increase regions-median cephalic and caudal was accentuated in the early stages of both species in question, tending to isometry as the species developed. This work was possible morphometric differentiation of species in larval stages of development, aside from a post-flexion. / O objetivo deste trabalho foi detectar diferenças morfométricas entre as larvas de Hemisorubim platyrhynchos e Pseudoplatystoma corruscans em diferentes estágios de desenvolvimento larval, como também no crescimento alométrico dessas duas espécies de peixes de relevância econômica e ecológica na América do Sul. As larvas foram separadas em seis estágios de desenvolvimento larval ontogenético (larval vitelino, pré-flexão, início de flexão, flexão, final de flexão e pós-flexão), determinando marcos anatômicos que possibilitaram agrupá-las em três séries de distâncias geométricas. Para encontrar possíveis padrões morfométricos (similar ou não) entre as larvas das duas espécies nos diferentes estágios de desenvolvimento foi aplicada a Análise de Componentes Principais (ACP) para cada série de distâncias. A análise de variância bifatorial (espécies e estágio de desenvolvimento) foi usada sobre os eixos retidos da ACP para a interpretação dos dados e testar as possíveis diferenças. Para verificar a existência de padrão alométrico nas espécies, os escores gerados pelo eixo 1 da ACP foram utilizados para estimar os coeficientes alométricos multivariados. A ACP para a série I separou as espécies (H. platyrhynchos e P. corruscans) e estágios (larval vitelino e pré-flexão). Na série II H. platyrhynchos não mostrou diferença entres os estágios larvais, porém para P. corruscans ocorreu desenvolvimento acentuado a partir do estágio de flexão ao final de flexão, possibilitando a separação de H. platyrhynchos. No estágio de pós-flexão (série III), a separação morfométrica entre as espécies não foi evidente. As espécies são distintas morfometricamente especialmente nos estágios iniciais de desenvolvimento larval, diferenças atribuídas principalmente ao formato e volume do saco vitelino à eclosão e ao tamanho das larvas, apesar de algumas similaridades nos padrões de crescimento de diferentes regiões do corpo. O crescimento alométrico das larvas correspondeu a eventos do desenvolvimento e comportamentais observados na ontogenia inicial das espécies. Este estudo sobre morfometria geométrica e crescimento alométrico das larvas de H. platyrhynchos e P. corruscans poderá auxiliar as pesquisas em biologia pesqueira, subsidiando as estratégias de manejo e conservação das populações naturais.