|dc.description.abstract||The behavior of railway systems, in particular the overhead contact line (OCL), is considered
in this thesis. The modelling of such systems with the use of the finite element method are
important to carry out, both to determine the systems static and dynamic behavior. The
dynamic behavior is primarily related to the relationship between the overhead contact line
and the pantograph, which should ensure stable currents to the train.
The aim for this study was to make a numerical model of a simplified, laboratory version
of an OCL. The laboratory model consisted of two spans approximately 15m long, with a
support in the middle. The finite element modelling was carried out both by the use of
Python scripting, and by the use of Abaqus/CAE version 6.14 as numerical software.
Laboratory measurements were conducted, primarily to validate the numerical model,
but also to compare the behavior of the laboratory model in relation to existing OCLs. The results of the measurements indicated that the laboratory model was approximately 3-4 stiffer
in the vertical direction, compared to existing OCLs.
The numerical model was compared with the measurements conducted in the laboratory,
which comparison primarily was related to natural frequencies and the variation of
elasticity along the span. Comparison between the results obtained in the laboratory and
the numerical results showed the same trend both due to natural frequency and elasticity,
but the numerical model was created both less stiff and more dense than the laboratory
model. Despite the latter, the numerical model was regarded as plausible in comparison to
the laboratory model, and it was further used in two parameter studies.
The parameter studies performed were related to:
1. The development of the first vertical natural frequency, for varied tension in the contact
2. The effect of removing droppers from the system, in relation to the first and the second
vertical natural frequency.
The results obtained in these parameter studies showed first of all that the development of
the first natural frequency was dependent on whether it was the contact or messenger wire
that was applied the highest magnitude of tension.
Further, the removal of a dropper in the model seemed to have more effect on the first
natural frequency, than the second natural frequency.
In future work, it is recommended that natural frequencies of higher modes than the first
and the second are investigated. The development of the first natural frequency of existing
OCLs should be compared with the results obtained in the parameter study made in this
thesis, to further validate the conclusions made.||