Abstract
Imperfections in the wheel–rail contact are one of the main sources of generation of railway vibrations. Consequently, it is essential to take expensive corrective maintenance measures, the results of which may be unknown. In order to assess the effectiveness of these measures, this paper develops a vehicle–track interaction model in the time domain of a curved track with presence of rail corrugation on the inner rail. To characterize the behavior of the track, a numerical finite element model is developed using ANSYS software, while the behavior of the vehicle is characterized by a unidirectional model of two masses developed with VAMPIRE PRO software. The overloads obtained with the dynamic model are applied to the numerical model and then, the vibrational response of the track is obtained. Results are validated with real data and used to assess the effectiveness of rail grinding in the reduction of wheel–rail forces and the vibration generation phenomenon.
Highlights
Rail corrugation has become one of the major problems in the field of railway engineering
In order to assess the effectiveness of these measures, this paper develops a vehicle–track interaction model in the time domain of a curved track with presence of rail corrugation on the inner rail
The proposed method, which consists of a time domain feedback between a multi-body model of the vehicle and a finite element numerical model of the track, is an interesting tool to study railway vibrations in time domain if some of the ground properties are unknown
Summary
Rail corrugation has become one of the major problems in the field of railway engineering. One of the inputs necessary for its development is the equivalent stiffness of the track (Keqn ), which cannot be calculated because some of the main properties of materials of the tack are undefined (Table 1) To solve this problem, the following calculation procedure is proposed (Fig. 3): Firstly, as the dynamic overloads caused by the rail corrugation are unknown, a simplified numerical model that disregards the presence of such pathology is developed (model 1a). A more realistic numerical model is developed (model 1b) and the dynamic overloads vector Fn is applied From this point on, a new calibration process in both rails begins, and a new stiffness value (Keqn ) is obtained. The model will be calibrated and validated for a Keqn stiffness and a Fn dynamic forces vector
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
