Selection of a suitable wear model for implementation in a generic rail damage function

Rails often suffer from surface defects caused by wear or Rolling Contact Fatigue (RCF). Therefore, several scientific and engineering models have been developed to predict the evolution of these surface defects. One such model is the Whole Life Rail Model (WLRM), which has been successfully implemented and is also easy to use by rail infrastructure managers for rail maintenance planning. However, this model is developed from a limited number of regions that cannot be considered as a representation of all possible scenarios. Furthermore, the development of such an empirical model is time-consuming. Therefore, developing a theoretical or numerical model, or a combination of both, is required that can be easily adapted for different operational scenarios, including various types of wheel and rail material. The WLRM model consists of three parts: 1) the RCF dominated region, 2) the wear dominated region, and 3) the region with both failure mechanisms interacting with each other. The study in this paper is focused on the wear dominated region as this mechanism is more straightforward compared with RCF. Typically two approaches are used in the literature to model the wear mechanism: 1) based on Archard's wear law and 2) based on the energy dissipation theory. In addition to these models, the authors previously developed a meta-model to reduce computational effort. Furthermore, the embedded wear toolkit in the multi-body software VI-Rail is also used for rail wear calculation. Finally, the most suitable model for implementation in WLRM is shown to be the classical Archard's model because of the local approach, including the spin effect and wide variety of wear coefficients.