Assessing the key factors affecting the substructure of ballast-less railway track under moving load using a double-beam model and random forest method

Abstract

In this paper, an approach is presented for analyzing the substructure of a slab track. The approach involves deriving a semi-analytical solution of the double-beam model, which incorporates a visco-elastic connection between two concrete beams and a spring-damping foundation. The model is then subjected to a concentrated moving load with a constant velocity. To more effectively assess the impact of various parameters, a random forest algorithm is utilized. The parameters considered include the spring-dashpot properties of the connection layer and the bedding layer, the thickness of the upper and lower beams, and the velocity of the moving load. The results of the random forest regression show that the stiffness of the intermediate and foundation layers have the greatest influence on the vertical deflection of the upper beam. However, the effect of stiffness of intermediate layer on the deflection and bending moment of the lower beam is found to be marginal. Moreover, for high-speed trains, the velocity of the concentrated moving load substantially changes the extreme responses attributed to the vertical deflection and bending moment. In this context, the damping properties of the interconnection layer and the subgrade course have a greater impact on the responses of the upper and lower beams.