# Msc Advanced Mechanical Engineering 2010-2012 Faure Julien

Methodology:
Modelling the vehicle

Procedure of the complex model:
Mechanical aspect: finite element formulation rigid body dynamics
zc Direction of the passenger vehicle

Obtain a vertical profile of the track for a line grade

Mathematical Modelling of a train coach subjected to geometry faults
Context:
The importance of the dynamic is obviously related to the overall safety and performance for high speed. In railway history, the aim was to improve performance for braking and pulling the series of coaches. This aim was easily fulfilled compared to the dynamic aspect: the study of bogie stability, the understanding of the inner movements and has been a main stake which leads to the development of impressive high speed vehicles . The role of track on the behavior of trains is essential because the tracks represents the excitation force of the vehicle system and a poor quality of the track can trigger fatal accidents. This is the reason why, the overall track quality is assessed with 6 different grades and 4 parameters enables to judge the track:
Name Horizontal Alignement Gauge Vertical Versine Describing The rail has moved from its horizontal axis The distance between two rails has changed The track has a vertical defect which is symmetrical One of the two rails is above or under the other one Scheme

c

Statistics aspect : Irregularities uses PSD functions determined with experiments Acceleration variance follows normal distribution

Numerical Integration
zb2
b2

zb1

b1

zw1

zw2 zw3

zw4

Acceleration of the car body Variance of the acceleration of a 10s time period : Repetition of the mechanical procedure Estimate the distribution of for each line grade Estimate the reliability of the method

Hypothesis:
Vertical Dynamics of the coach considered One rail profile considered and half of the train modelled (symmetry) The track is straight, rigid and the vehicle is at a constant speed The rail/wheel contact is represented with a linear Hertz spring The car body and the bogies have two DOF (rolling & vertical motion) The irregularities are simulated with trigonometric functions (random phase angles) and a amplitude depending on PSD functions.

Simple model of the coach with 2 Degrees of freedom Modelling the track as a Bernouilli beam on discrete supports

Other work:

Horizontal Versine

Results:
Displacements of each component in function of the time:

Conclusion :

Problem statement:
The geometry quality of a railway track under loads can be indirectly measured analysing the dynamic response of a train coach travelling over it. Instruments to measure these reactions have been developed, but since the cost of the equipment of a normal train coach is considerably lower than that required for a track recording machine, usually this indirect methodology is adopted. These measurements can be at best an approximation of the railway track geometry, since the coach dynamic characteristics influence the response.

Objectives:
Experimental recordings with accelerometers in two car bodies (tramway & train) : The present work aims at : Obtaining the dynamic modelling (vertical and horizontal) of a typical train coach subjected to different railway track geometric faults (gauge, super elevation, twist…) Correlating the accelerations inside the car body with the quality of the track and being able to find a trend. Having a reliability study on the results obtained and compare the results with the existing standards. the qualities of the two tracks are to60 tally different. 50 The vertical acceleration inside the 40 car body are also very different. 30 20 —> Correlation between track irregulari10 ties and acceleration exist. 0 The order of magnitude of the maximum acceleration for simulation & Acceleration [mg] experiments are similar. —> Simulations are validated and a reliability study can be done
Percentage [%]

Likelihood of the line grade as a function of the acceleration inside the car body (v=18.5 m/s) of one specific vehicle. The method enables us to obtain tables of likelihood depending on the line grade, the speed of the vehicle and the mechanical parameters of the track. —> The accelerometers enables us to have a cheap way to measure the quality of tracks and fill the initial aim of the thesis.

Future Work :

Comparison of the maximal acceleration (real recording)

Refine the methodology be-3.5 0 0 0 0 3 97 cause it was the first attempt to -3.6 0 0 0 0 2 98 correlate track and acceleration. Define the best speed for track measurement (avoiding natural frequencies and have the largest confidence interval) Model the vehicle considering two deformable rails and make it more complex with lateral and longitudinal motions. Obtain experimental data for the line grades : The Gaussian process (random angles) to obtain the irregularity is questionable.

-1.3 -1.4 -1.5 -1.6 -1.7 -1.8 -1.9 -2 -2.1 -2.2 -2.3 -2.4 -2.5 -2.6 -2.7 -2.8 -2.9 -3 -3.1 -3.2 -3.3 -3.4

1 94 89 81 68 51 32 17 7 2 1 0 0 0 0 0 0 0 0 0 0 0 0

2 6 10 18 29 41 49 49 39 24 11 4 1 0 0 0 0 0 0 0 0 0 0

3 0 1 1 3 8 17 29 41 48 44 32 19 9 3 1 0 0 0 0 0 0 0

4 0 0 0 0 1 2 5 12 23 35 44 46 38 26 15 8 4 2 1 0 0 0

5 0 0 0 0 0 0 0 1 3 9 18 30 41 46 45 38 29 21 14 10 7 5

6 0 0 0 0 0 0 0 0 0 0 2 5 12 24 39 54 68 78 85 90 93 95

0 25

50

75 10 0 12 5 15 0 17 5

0

5

0

5

0

5

0

5

0

5

20

22

25

27

30

32

35

37

40

42

References :
Lei, X. (2002). Analysis of dynamic response of vehicle and track coupling system with random irregularity of track vertical profile. Journal of Sound and Vibration Sun, Y. (2002). A dynamic model for the vertical interaction of the rail track and wagon system. International Journal of Solids and Structures Dong, R. (1994). Vertical dynamics of railway vehicle track system. Romero, A., & Domı, J. (2010). Fully three-dimensional analysis of high-speed train – track – soil-structure dynamic interaction. Journal of Sound and Vibration Wu, J. (2000). The use of finite element techniques for calculating the dynamic response of structures to moving loads. Computers & Structures

Acknowledgments :
Dr Maurizio Collu Supervisor — Cranfield University Mr Alban Leymarie—Leyfa Measurement Mr Jean– Michel Thiry Pr Jean-Pierre Fortin— SNCF Dr Athanasios Kolios — Cranfield University

45

0