A mechanistic-empirical method for characterisation of railway track formation

Abstract

The demand on South African railway freight lines is growing constantly with respect to train speeds and axle loading, thus increasing the forces and stresses experienced by the track structure. This may lead to rapid track deterioration as a result of a loss in track geometry caused by the poor and nonuniform underlying support. The railway engineering industry has moved towards gaining a better understanding of track formation behaviour for development of practical and effective deflection measurement techniques that allow accurate determination of formation structural capacity, identification of track problem areas and evaluation of track condition. Thus, there is a need to develop a deflection measurement method that can accurately evaluate the track formation condition through analysis of the full deflection basin under train passage, in order to quantify the formation structural capacity based on the relevant in-service conditions. In this research study, a mechanistic-empirical method (referred to as the inverse method) is proposed with the main objective of developing an alternative method for evaluation of formation layer structural capacity through the inverse estimation of formation layers’ elastic moduli. The method was founded on surface deflection theory from falling weight deflectometer (FWD) analysis. Finite element analysis and field data from railway substructure deflections under train transient loading obtained through multi-depth deflectometers (MDDs) were used to assess the validity of the inverse method by comparing measured and modelled railway substructure responses. The results showed that substructure deflections and stresses are affected by the complex superposition of different bogie loading configurations on a particular superstructure. The increase in axle loading was found to be directly proportional to the increase in formation peak deflection. The effect of travelling speed was however insignificant for speeds less than 80 km/h. Furthermore, the load distribution in the railway substructure did not follow a 45o influence line as commonly assumed in surface deflection theory. On the contrary, railway equilibrium influence lines (EILs) were significantly influenced by the elastic moduli of the formation layers and in-situ subgrade, therefore governed by the structural capacity of the substructure layers. The research therefore concluded that the formation layers are expected to gradually deteriorate and experience increased deflections over time. However, the top of formation may vary as it seems to be highly influenced by the superstructure load distribution. The inverse method strongly agreed with the long-term formation peak strains measured with MDDs. Furthermore, the method was determined suitable for evaluation of formation structural capacity, as good agreement was found between the measured and estimated formation layers’ elastic moduli.