A novel methodology to automatically include general track flexibility in railway vehicle dynamic analyses

Abstract

The interaction between the rolling stock and the infrastructure plays a crucial role in railway vehicle dynamics. The standard approach consists of using a multibody formulation to model the railway vehicles running on simplified tracks. The track model can be rigid, if it comprises only a geometric description of the rail; semi-rigid, if it considers an elastic foundation underneath the rail; or a moving track model, if it comprises a track section underneath each wheelset traveling with the same speed of the vehicle. Despite their computational inexpensiveness, these approaches do not provide a complete representation of track flexibility and disregard coupling effects with the vehicle and among the track components. This work proposes a methodology to automatically generate finite element models of railway tracks comprising its relevant flexible components, i.e., rails, pads, fastening systems, sleepers, and ballast or slab. The finite element mesh is generated based on a parametric description of the track that allows an accurate description of its geometry, including curvature, cross-level, grade, and irregularities. The methodology is demonstrated with a case study in which a track with a complex geometry is loaded with two different approaches. The first approach prescribes moving loads, which is a typical approach used to design or analyze the infrastructure. The second approach applies loads retrieved from the dynamic analysis of a complete vehicle. The results show the benefits of this method and reveal that prescribed loading underestimates the forces resulting from the vehicle dynamics, which is an important issue on curved sections.

A finite element methodology to model flexible tracks with arbitrary geometry for railway dynamics applications

“…Cross Level Rail inclination 12 reference frames previously described. A detailed description of the automatic generation of railway tracks with arbitrary geometry is found in [70].…”

Parametric analysis of railway infrastructure for improved performance and lower life-cycle costs using machine learning techniques

“…Other authors have proposed co-simulation methodologies using Finite Element Method (FEM) and multibody formulations in order to study the flexibility of track structures under realistic trainset loads [31][32][33][34]. These developments open up the possibility of integrating more detailed wheel-rail contact models [35][36][37][38][39][40][41][42][43][44], to consider track irregularities [45,46] and other track singularities [10,[47][48][49][50] in the studies aiming to assess track performance and degradation evolution [51][52][53][54][55] under realistic operation conditions.…”

Three-dimensional modelling of slab-track systems based on dynamic experimental tests

“…Sainz-Aja et al studied experimentally the properties of rail pads under different operational conditions [30]. Other authors have proposed co-simulation methodologies between MB and FE formulations in order to study the track structure under realistic trainset loads [31][32][33][34]. These developments open the possibility of integrating complex track geometries [35][36][37], more detailed wheel-rail contact models [38][39][40][41][42][43][44][45][46], to consider track irregularities [47,48] and other track singularities [49][50][51] in the studies aiming to assess the track performance and degradation evolution [52][53][54][55][56][57] in realistic operation conditions.…”