Optimization of rail profile to reduce wear on curved track

Wear is one of the most important factors affecting the structural life of railways. In urban railways, the curved track results in more wear than tangent track. Furthermore, the wear is frequently detected in a sharply curved track. Wear leads to reduction in the durability of railway as well as increase in the risk of derailment. In a recent study, a rail grinding was suggested to enhance the dynamic performance and extend the fatigue life of rails. This paper deals with an optimum design procedure for asymmetric rail head profile wherein the design profiles of high and low rails are simultaneously determined by minimizing the wear on curved tracks. The procedure employs a genetic-algorithm-based optimization method. The dynamic performances of a train such as lateral force, derailment and vertical force are evaluated using a rail vehicle dynamics program. The optimum design is validated by comparing the initial rail profile with the optimum profile in a full-scale wear test.