Thesis (Ph. D.)--University of Texas at Austin, 2004. / Vita. Includes bibliographical references (leaves 275-293). Also available in an electronic version.

Thesis (Ph. D.)--University of Texas at Austin, 2004. / Supervisor: John L. Tassoulas. Vita. Includes bibliographical references. Also available from UMI.

Predictions of the environmental loading and induced motional and structural responses are among the most important aspects in the overall design process of offshore structures and ships. In this thesis, attention is focused on the wave loads and excited bodily motion responses of large offshore structures and special vessels. With the aim of improving the existing theoretical methods to provide techniques of theoretical effectiveness, computational efficiency, and engineering practicality in marine and offshore applications, the thesis concentrates upon describing fundamental and essential aspects in the physical phenomenon associated with wave-structure interactions and deriving new methods and techniques to analyse offshore structures and unconventional ships of practical interest. The total wave force arising from such a wave-structural interaction is assumed to be a simple superposition of the potential and the viscous flow force components. The linear potential forces are solved by the Green function integral equation whilst the viscous forces are estimated based on the Morison's damping formula. Forms of the Green function integral equation and the associated Green function are given systematically for various practical cases. The relevant two-dimensional versions are then derived by a transformation procedure. Techniques are developed to solve the integral equation numerically including the interior integral formulation and, in particular, to tackle the mathematical difficulties at irregular frequencies. In applying the integral equations to solve problems with various offshore structures and special vessels, some modified, improved or simplified methods are proposed. At first, simplified method is derived for predictions of the surge, sway and yaw motions of elongated bodies of full sectional geometry or structures with shallow draft. Then, a new shallow draft theory is described for both three- and two-dimensional cases with inclusion of the finite draft effect. Furthermore, a three-dimensional strip method is formulated where the end effects of the body are fully taken into account. Finally, an approximation to the horizontal mean drift forces of multi-column offshore structures are presented. Some new findings are also discussed including the multiple resonances occurring in the motions of multi-hulled marine structures due to the wave-body interaction, the mutual cancellation effect of the diffraction and the radiation forces arising from a full shaped slender body, and so on. Further to those verification studies for individual methods developed, more comprehensive example investigations are given related to two industrial applications. One is a derrick barge semi-submersible with zero forward speed; and the other, a SWATH ship with considerable speed. By correlation of all the proposed approaches with available analytical, numerical and experimental data, the thesis tries to demonstrate a principle that as long as principal physical aspects in the wave-structure interaction problem are properly treated, an appropriately modified or simplified method works, performs well and, sometimes, even better.

Este trabalho apresenta um modelo para a verificação do trem-tipo de projeto de pontes e viadutos da norma brasileira a partir do tráfego real nas estradas. Inicialmente, apresenta-se a comparação da carga móvel da norma brasileira com o Eurocode e com as normas da AASHTO. Em seguida, apresenta-se o estado da arte no Brasil e no mundo em pesquisas na área de cargas móveis em pontes. Propõe-se um modelo de verificação do trem-tipo baseado em simulações das passagens e cargas de veículos, através do método de Monte Carlo. São utilizados dados de pesagem e de quantidades de veículos provenientes de uma rodovia concedida à iniciativa privada no interior paulista. São efetuadas análises de pontes celulares com diversos vãos, de 30 a 200 metros, sendo estudados os momentos fletores no meio do vão e forças cortantes adjacentes aos apoios, em obras isostáticas, e momentos fletores no apoio central de pontes contínuas de dois vãos. É dada maior ênfase às situações de congestionamento sobre a ponte, e são efetuadas extrapolações estatísticas para a obtenção dos esforços característicos, levando-se em conta as diferentes situações de tráfego possíveis. Por fim, é feita uma extensão ao estudo das grelhas, utilizando-se linhas de influência transversais em ponte com vão de 30 metros, sem transversinas intermediárias. / This work presents a model to verify the design traffic load model of the Brazilian code, based on real traffic at road bridges. First, it compares Brazilian load model with Eurocode 1 and AASHTO Standard and LRFD Specifications. Then, recent development and research on bridge traffic load models are presented, in Brazil and other countries. A verification model of the traffic load is proposed, based on quantity and weight vehicle simulations, using Monte Carlo method. Real traffic data of trucks, used in this work, have been collected at highways of Sao Paulo state. Box-girder bridges from 30 to 200 m spans are investigated, studying the bending moment at midspan and the shear forces near the supports, on simply supported bridges, and the bending moment at center support, on two-span continuous bridges. Traffic jam situations on the bridge are emphasyzed, and statistical extrapolations to the load effects carachteristic values are done. Finally, it presents an extension of this study to multi-beam cross sections, using transverse influence lines, in a 30 m span bridge, without transverse beams.

Este trabalho apresenta um modelo para a verificação do trem-tipo de projeto de pontes e viadutos da norma brasileira a partir do tráfego real nas estradas. Inicialmente, apresenta-se a comparação da carga móvel da norma brasileira com o Eurocode e com as normas da AASHTO. Em seguida, apresenta-se o estado da arte no Brasil e no mundo em pesquisas na área de cargas móveis em pontes. Propõe-se um modelo de verificação do trem-tipo baseado em simulações das passagens e cargas de veículos, através do método de Monte Carlo. São utilizados dados de pesagem e de quantidades de veículos provenientes de uma rodovia concedida à iniciativa privada no interior paulista. São efetuadas análises de pontes celulares com diversos vãos, de 30 a 200 metros, sendo estudados os momentos fletores no meio do vão e forças cortantes adjacentes aos apoios, em obras isostáticas, e momentos fletores no apoio central de pontes contínuas de dois vãos. É dada maior ênfase às situações de congestionamento sobre a ponte, e são efetuadas extrapolações estatísticas para a obtenção dos esforços característicos, levando-se em conta as diferentes situações de tráfego possíveis. Por fim, é feita uma extensão ao estudo das grelhas, utilizando-se linhas de influência transversais em ponte com vão de 30 metros, sem transversinas intermediárias. / This work presents a model to verify the design traffic load model of the Brazilian code, based on real traffic at road bridges. First, it compares Brazilian load model with Eurocode 1 and AASHTO Standard and LRFD Specifications. Then, recent development and research on bridge traffic load models are presented, in Brazil and other countries. A verification model of the traffic load is proposed, based on quantity and weight vehicle simulations, using Monte Carlo method. Real traffic data of trucks, used in this work, have been collected at highways of Sao Paulo state. Box-girder bridges from 30 to 200 m spans are investigated, studying the bending moment at midspan and the shear forces near the supports, on simply supported bridges, and the bending moment at center support, on two-span continuous bridges. Traffic jam situations on the bridge are emphasyzed, and statistical extrapolations to the load effects carachteristic values are done. Finally, it presents an extension of this study to multi-beam cross sections, using transverse influence lines, in a 30 m span bridge, without transverse beams.