logo

You are here

Warning message

Attention! This event has already passed.

The reduction of vibration in railway tracks through the application of non-linear modelling

Thursday, 27 April, 2006 - 16:00
Campus: Brussels Humanities, Sciences & Engineering campus
K
auditorium P. Janssens
Jurgen Maes
phd defence

The impact of trains running through congested areas at higher speeds and with higher
axle loads is perceived as more environmentally intrusive. At the same time, railway
companies are forced to reduce the life-cycle cost of the rail infrastructure. This work tries
to contribute to possible solutions through the study of non-linear material characteristics
on the dynamic behaviour of the track. A new resin-bonded rubber material, a new railpad
model and a new non-linear numerical track model presented in this work are the next step
towards the further development of new railway track components such as the Composite
Sleeper and the Double Tuned Rail Damper for the reduction of noise emission, vibration
nuisance and maintenance requirements.

Resin-bonded rubber (RR) is a shredded rubber material, bonded with a polymer resin. The
rubber can be recycled from used tyres. This composite material has rubber-like behaviour
and can be characterised as a non-linear visco-elastic material. In this work, a test set-up is
developed to measure the dynamic properties of RR elastic pads as a function of preload
and frequency, based on the ISO 10846 direct method which is extended to high
frequencies.

The non-linear dynamic railpad behaviour, translated in a Wiener model, is incorporated in
a FE railway track model that is solved in space and time. Not only can this model be used
to determine the significance of the non-linear behaviour of railpads on the dynamics of the
railway track, the model can also be used as an analysis tool for the development of new
track components.

A scale model railway track is built as a benchmark. It offers the advantage of flexibility in
construction, manageability of important parameters and availability compared to
measurements in situ. Accelerance measurements are done at various fastening preloads
and static loading levels for a range of railpad types. It is found that the numerical results
fit the experimental data well under the presented conditions.

Numerical analysis shows that the track response is more accurately predicted throughout
the 40Hz to 3500Hz frequency range when the load and frequency dependence of the
railpad stiffness is taken into account. However, a well considered constant railpad
stiffness gives satisfactory results in a limited frequency range. It is also shown that the
reciprocity principle should be used cautiously when an unsprung mass is attached to the
rail. The finite track length is found to be dependent on the railpad characteristics. Soft
railpads require a longer track length or the use of special boundary conditions to prevent
the reflection of transverse waves at the rail ends. And finally, it is demonstrated that it is
theoretically possible to improve the track response by enhancing the non-linear
characteristics of the railpads.