Affordable Access

Publisher Website

Simulation of the dynamic response of high-speed line structures composed of granular or bituminous sub-ballast layers and comparison with in situ measurements from embedded instrumentation

Authors
  • Khairallah, Diana
  • Chupin, Olivier
  • Blanc, Juliette
  • Hornych, Pierre
  • Piau, Jean Michel
Publication Date
Jan 01, 2022
Identifiers
DOI: 10.1016/j.trgeo.2022.100767
OAI: oai:HAL:hal-03678319v1
Source
HAL-Descartes
Keywords
Language
English
License
Unknown
External links

Abstract

The "Brittany-Loire" high-speed line (HSL BPL), connecting the cities of Rennes and Le Mans is the first large-scale application, in France, of ballasted tracks with bituminous sublayer and varied subgrade conditions. To evaluate the behaviour of this new type of track structure, and validate the design assumptions, it was decided to set in place a monitoring system, to monitor the mechanical response of the track. For that purpose, three track sections were instrumented during construction using accelerometers, anchored deflection sensors, strain gauges and temperature probes. The objective of this paper is to analyse the response of three instrumented track sections of the HSL BPL, either with a bituminous or a granular sublayer, and to model their response using a dynamic model, considering a viscoelastic multilayer structure subjected to moving loads, implemented in the ViscoRail software. From the numerical point of view, the objective was to verify the capacity of ViscoRail to simulate the response of the three BPL sections, using as much as possible a unique set of mechanical parameters, while considering the specificities of each sections. The comparison between the modelling results and the in situ measurements indicates that the model is able to reproduce the main features of the dynamic response of railway tracks. In particular, the role of the asphalt concrete layer, which tends to reduce vertical accelerations in the ballast layer is well reproduced by ViscoRail.

Report this publication

Statistics

Seen <100 times