Skip to Main content Skip to Navigation
Theses

Développement de traverses instrumentées pour l'étude du comportement des voies ferrées

Abstract : Smart Sleepers, instrumented sleepers deformation sensors (FBG) allow to obtain information on the behavior of the cross on track and especially on its level of deformation during the passage of trains. As a result, more detailed knowledge of the adequacy of the design of the cross and the stresses it undergoes in the process is possible. Moreover, it portends that the way the cross rests on the ballast will affect the deformation of passing trains. The settlement of the ballast is likely to modify the support of the cross on it, we can therefore conclude that the Smart Sleeper technology should provide information on the evolution of the ballast layer. Optimizing maintenance operations could be considered accordingly. An assessment of the rolling stock in terms of the process of recognition, counting, or detection of faulty equipment could be explored. The thesis is centered on the operation and interpretation of the data provided by the Smart Sleepers. For this work will be divided into four main stages. 1) Data collection and analysis: Amounts of statistically representative traffic data should be collected from the Smart Sleepers lanes automatically. In this regard, several channels are currently equipped instrumented sleepers. The sites equipped correspond to different types of channels (high speed lines, conventional lines, secondary network) to study the various cases encountered in service. Participation in the development of new sites, in France or abroad, will be considered. Automated collection of data (big data) is underway and will be part of the thesis. This particular part of the work will include the development of a scalable database architecture adapted to the needs of provision of different data depending on the intended application and user. This work includes the development of technology, the development of a method for the collection and processing of data (filtering Fourier transform). 2) Modeling of the entire train / track: To interpret the data obtained, it is necessary to develop a model of the whole process more channels. The objective is to predict the information measured in the Smart Sleepers from the knowledge of the mechanical characteristics of rolling stock and track. The model will be first validated results from laboratory tests. Then a recalibration will be considered on a small number of tests on representative site behavior of this system. 3) Study and parametric identification: Once developed and validated model, a parametric study will be conducted in order to estimate the influence of the main mechanical parameters of the train / track on the signal obtained in the Smart Sleeper. The goal is, in a reverse process to be able to go back to rolling stock characteristics or channel from the information measured in the smart sleeper. Reflection on improving the design of the cross will be carried out (number of sensors, their type, their positions). 4) Definition of indicators: The objective is to demonstrate the technology's ability to provide information recoverable for the actors in the railway world and to improve equipment maintenance and route. The ability to extract relevant indicators concerning the behavior of the cross on the way, the overall evolution of the track structure possibly in relation to infrastructure properties, and some rolling stock characteristics will be the final goal. data translation will be developed by relevant indicators for users
Complete list of metadata

https://pastel.archives-ouvertes.fr/tel-03235557
Contributor : Abes Star :  Contact
Submitted on : Tuesday, May 25, 2021 - 7:08:09 PM
Last modification on : Wednesday, May 26, 2021 - 3:41:18 AM
Long-term archiving on: : Thursday, August 26, 2021 - 8:54:24 PM

File

TH2020PESC1011.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03235557, version 1

Collections

Citation

Le Hung Tran. Développement de traverses instrumentées pour l'étude du comportement des voies ferrées. Génie mécanique [physics.class-ph]. Université Paris-Est, 2020. Français. ⟨NNT : 2020PESC1011⟩. ⟨tel-03235557⟩

Share

Metrics

Record views

116

Files downloads

46