Abstract

The subject of this thesis is the analysis of the behavior of longitudinally unstiffened and stiffened steel plate girders (I - girders) loaded with local load (concentrated or distributed load on a certain length) on one flange in the plane of the girder web and determination of their ultimate load. This case of loading often appears in practice and examples are the loads of the main girders during the construction of bridges by launching technology, crane track loads, column-beam connections etc. The problem has been studied intensively for several decades, theoretically and experimentally, with the aim of monitoring behavior and determining the ultimate strength. However, no solutions have yet been obtained that would satisfactorily capture the complex impact of all parameters. Recently, numerical analysis have been used often in combination with experimental research. By using software packages based on the finite element method, it is possible to perform more extensive parametric analysis. Thus, the influences of individual parameters can be singled out and quantified in the best way. A similar parametric analysis was conducted in this paper. In this research, the influence of load length and other parameters on the ultimate load capacity of longitudinally unstiffened and stiffened steel I-beams was analyzed. The primary goal of this study is to analyze strengthening effect, which is defined as the relation between the ultimate capacity of longitudinally stiffened and longitudinally unstiffened girders. The strengthening effect for different load lengths depends on the stiffness of the longitudinal stiffeners, the web aspect ratio, web slenderness and the stiffness of the loaded flange. The current European standard EN 1993-1-5 does not take into account the influence of the load length on the strengthening effect. This analysis, however, shows that the strengthening effect depends on the load length. A formula for determining the ultimate load capacity of unstiffened and stiffened steel plate girders, which include the influence of a number of parameters is defined.