Keywords: catenary system, finite element modelling, system identification, Norwegian railways, dynamic response, catenary-pantograph interaction, contact wire, numerical simulations.

In Norway a two level overhead contact line system and pantograph is used to supply the power to the electric railway vehicles. For this power supply to be reliable and uninterrupted, and also for higher speed rail, there must be strict static and dynamic requirements for the contact line characteristics. Due to the Norwegian topography the national railway network is particularly is characterized by numerous and sharp curves, often with a radius well below 1000 m. The dynamic behaviour in the overhead contact system in such tight curves is expected to be different from similar but straight segments. This study looks primary at dynamic sampling by monitoring, and to study its implications in evaluation of the catenary system. It includes the investigation of relevant limiting factors when considering higher speeds, beyond original design speed, while upholding high reliability and low wear. Usually it is assumed that for high speed railway overhead contact lines no significant dynamic response is present below 200 km/h due to the system design. However these effects will, for some existing systems designed for lower speed, appear at much lower speeds. The investigated speed range will thus be from 90 to 200 km/h, depending on the situation. For the sampling to be efficient it is important to establish relevant conditions for the dynamic system, and to verify the existing system behaviour. This study includes several possible sampling points within one or several spans, as well as sampling at the cantilever support. In an extension this may possibly lead to predictability of the state to a passing pantograph. The study investigates a number of possible sampling options by numeric simulations. Such parametric study is an important part of the optimization process to reduce the dynamic response, thereby reducing wear of the contact wire, and to achieve a reliable system. The focus in the present paper will therefore be on the prediction of train velocity by the cross-correlation function, extracting the frequency content of sampled time series, and uplift predictions at different points of the span, data sampled either as displacements or as accelerations are also included.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £65 +P&P)