Numerical Simulation of Wear in Railway Wheel Profiles

The work describes a method to predict the evolution of the wheel profile of a railway vehicle, depending on the load history acting on the wheelset. The method is based on the determination of the wear on the contact area, which is divided into finite elements according to the strip theory. For each element, in presence of slip, the amount of material loss is evaluated depending on the local value of tangential force and creepage (the meaning of creepage is assumed according to the definition given in [14], [15], [16] as the ratio between the sliding velocity and the tangential rolling velocity). The empirical relation is evaluated according to results of experimental test obtained from literature. The wear is calculated for the entire contact area superimposing the contribution of each element. The motion of the wheelset in lateral direction causes a motion of the contact patch along the profile. Sequentially, the contact area will acquire a different contact shape and stress distribution. The shape of the worn profile depends on both the load condition and the motion of the wheelset with respect to the track. This profile can be obtained from the new one by subtracting at each time step the material removed from the contact area. This procedure is simple, but requires variable profiles for each time step, and is not efficient in computational terms. The strategy proposed here by the authors, is to consider finite periods obtained superimposing several revolution of the wheelset. The worn profile is evaluated in a single step from the cumulative of damage of an entire period. The limitation of this method consists in the different behavior of a wheelset with worn profile respect to a wheelset with new ones, and therefore produces different wear. It is necessary to determine an optimal value for the period to be used to re-evaluate the profile shape, in order to minimize the difference in the predicted shape itself. The method is applied to a suspended wheelset, running on a simulated test track, with S1002/UIC60 profiles. Different periods of re-evaluation of the profiles are considered in order to demonstrate the influence of this parameter.