Reinvestigation into railway wheel-track interaction and suspension damage

Abstract

Without considering either velocity or acceleration effects, the current conventional method presented in literature applies the vertical deflection of a wheel centre caused by a flat defect to the Hertzian contact theory. This method has been numerically and theoretically proved to be inappropriate and can incorrectly predict a higher wheel-rail impact force for a low speed than a high speed. Therefore, under a hypothesis of no wheel bouncing and sliding, two new methods, the velocity-based and the acceleration-based have been proposed. The former method takes the wheel centre deflection change in each computational increment from the Hertzian contact theory while the latter applies the wheel centre acceleration caused by the flat in revolutions to the wheel as a force in dynamic simulation, which interprets the speed effects on impacts precisely. A sensitivity study proves that the velocity-based method is unreliable as opposed to the acceleration-based method. A beam/rigid FE model has also been developed to inspect the wheel-track interaction by performing dynamic analysis in the time domain. It has been found out that the impact responses predicted by the FE analysis and the velocity method are similar and the FE results heavily depend on the compute increment, which implies the FE modelling in ABAQUS may be unreliable for this issue with current applied increments. Finally, the results calculated using the acceleration method have been employed to study the suspension/damper torsional stress caused by a wheel flat. This indicates that a wheel flat may lead to potential fatigue damage if without proper maintenance management.