| Railroad vehicles are subjected to excessive dynamic loads and are required to operate at higher speeds. Accurate numerical modeling of railroad vehicle systems is vital to increasing railroad safety and efficiency. This thesis aims to contribute to this area through the use of the methods of nonlinear computational multibody system dynamics.;Most algorithms employed by the railroad industry are based on two-dimensional wheel/rail contact formulations, in which the coupling between certain geometric parameters is neglected. This coupling is accounted for in three-dimensional contact formulations. Simplified two-dimensional models can lead to an inaccurate prediction of the location of the contact points. The results obtained in this thesis show that small errors in predicting the wheel/rail contact locations can lead to significant errors in the calculation of forces. Based on this preliminary wheel/rail contact investigation, a three-dimensional wheel/rail contact model is used to develop a new formulation that accounts for the track structural flexibility. In this formulation, the geometry of the rails is described using isoparametric finite element representation, while the rail deformation is described using the finite element floating frame of reference formulation. This new formulation, which is highly nonlinear, is validated by comparison with simple linear models (moving load on a Winkler foundation). Also, the relationship between the track frame widely used in railroad vehicle dynamics and the Frenet frame used in the theory of curves in classical differential geometry is developed in this thesis. |
