Digital twin study to investigate the impact of heavy haul train driving strategies on rail wear

Train driving strategies are dictated by the need for meeting timetabling and energy-efficient control. The magnitude and duration of traction and braking efforts can vary significantly and is strongly dependent on heavy haul train operating parameters. There is a lack of tools to estimate rail wear for a complete network, and considering full train operation, including the contribution of different types of rollingstock. 

Rail wear simulation studies require high-fidelity multibody models which make the simulation work computationally expensive, and the obtained results often just describe a particular individual vehicle-track interaction case. With the advent of high-performance computing facilities, a full train co-simulation that consists of multiple rail vehicle physics-based (multibody) models can now be performed using parallel simulations in a single run. In this paper, a digital twin method has been proposed to investigate the impact of train driving strategies on rail wear. The method has been demonstrated with a case study. The limitations of the proposed method have been discussed.