With the increase of the spans in the last decades, new concepts of projects and perceptions about bridges around the world have imposed the need of the development of new design techniques. Aerodynamic instabilities, which were not being of concern not long time ago, started to demand special care during all phases of the projects. Among them, coupledflutter instability has gained remarkable importance, due to its catastrophic consequences and impacts on the safety of the structure. Concerning coupled-flutter, aerodynamic derivatives have been regarded as fundamental components for the assessment of the stability of long-span bridges, resulting in that the focus during the proposition of the bridge deck geometry has been put essentially on the control of those empirical indices. The relationships between bridge deck geometric characteristics and the values assumed by those indices are recognized to be non-linear and straightforward approaches for their prompt evaluation are still not available. Because of that, trial and error experimental techniques based on wind tunnel tests have been regarded as main tools in the search of geometric configurations of bridge decks that satisfy the flutter stability requirements from the aerodynamic derivatives point of view. Since aerodynamic derivatives are resultant from the unsteady pressure characteristics developed along the bridge deck, it must be more rational to search for relationships between deck geometry and flutter stabilization from the unsteady pressure characteristics point of view. However, the complex geometric characteristics associated to modern longspan bridges may impose difficulties on the pressure measurements in wind tunnel tests, becoming counter-productive. So, the direct evaluation of the aerodynamic derivatives through alternative methods has been preferred in wind tunnel tests. In this research, the impacts of the deck geometric characteristics on the aerodynamic derivatives and, consequently, on flutter stabilization are assessed from the unsteady pressure characteristics point of view. Rectangular cross-sections arranged in single box and twobox configurations are used as base geometries and the influences of different geometric singularities (geometric modifications as leading edges, vertical plates and slots) on the unsteady pressure characteristics of those cylinders are systematized through the proposition of semi-empirical formulations. The knowledge generated herein may serve as a base for the development of a design framework based on the direct manipulation of the unsteady pressure characteristics of the deck through the use of geometric singularities, aiming the flutter stabilization by controlling indirectly the aerodynamic derivatives. Such a framework is to be used in substitution of some stages of wind tunnel test campaigns, and its feasibility is evaluated along the study. It was concluded that the relationships of geometric singularities with the unsteady pressure characteristics are much more linear and predictable than their relationships with the aerodynamic derivatives, opening a new methodology for the proposition of geometric improvements. Data regarding the geometries studied herein are also provided along the study, as reference for future development. / Kyoto University (京都大学) / 0048 / 新制・課程博士 / 博士(工学) / 甲第14914号 / 工博第3141号 / 新制||工||1471(附属図書館) / 27352 / UT51-2009-M828 / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 白土 博通, 教授 杉浦 邦征, 教授 河井 宏允 / 学位規則第4条第1項該当