Final Report Summary - RAILCOM (Electromagnetic compatibility between rolling stock and rail-infrastructure encouraging European interoperability)
The objectives of the project were:
- harmonisation of interference limits for train detection systems on TEN-T railway network;
- characterisation of the railway electromagnetic environment for communication systems, with correlation between EM emission and operating conditions of the system;
- proposals for the ongoing standardisation process within CENELEC.
The research activity, including modelling and measurements, has been focused on train detection and communication systems. Harmonised calculation methods have been identified and validated through appropriate test campaigns. Strong effort has been made in order to favour harmonisation of interference limits and methods for determination of limits of train detection systems, overcoming the barriers imposed by national regulations and practice. The EM of the railway systems has been related to its operating conditions, in order to enable to forecast electromagnetic emission on the basis of the characteristics of the systems and to assess interference with communication systems.
Thanks to measurements in the Netherlands and Switzerland a lot of knowledge about resonances existed already. With additional measurement campaigns, more data about the interface between train and power supply has been collected in several European countries. Both line voltage and line current of a test train are measured over several days, with good accuracy and for frequencies up to 20 kHz, using a PC based data-acquisition system coupled to an emc-sturdy measurement system based on the so called D/I-technology.
The measured data were evaluated in the time domain by bandpass filtering, followed by a moving RMS function with defined time window (integration time) for different centre frequencies. A histogram was build for each of the frequency bands from the results obtained from bandpass filtering process. A final output is probability distribution of voltages and currents versus frequency for each test campaign (traction power supply system).
The project has provided a toolbox of fully validated methods for the characterisation of train detection systems, and for the assessment of compatibility between rolling stock and train detection systems.
With respect to communication purposes the EM environment consists of two parts:
1. On top of the train;
2. Under the train.
For the analysis of the GSM-R the environment on top of the train is the most important one, with respect to the Eurobalise the most important one is the environment under the train. For GSM-R a method to determine the characteristics of the EM-environment on top of a train has been developed. The results obtained using this method are dependent on the type of antenna used, indications exist that the distributions observed in experiments are strongly influenced by the antenna characteristics. For the Eurobalise the environment under the train is the most important one. During the EMC ARTS programme this has lead to the development of a 'standardised' antenna, which can be used under a vehicle. Basically the antennas used consist of square loops which are placed under the vehicle. The test campaigns have identified and agreed among the UNISIG partners in the RAILCOM project the laboratory practical test arrangement, dimensions of the inductive loop, and noise characteristics in time and frequency domain.
A general evaluation of the most important theoretical models for electrical lines in frequency and in time domain has been performed; the positive and negative aspects of each representation have been pointed out (i.e. application fields, approximations and requested CPU time for elaboration) and innovative models available for future applications have been presented. Concerning the net-impedance activity, coordinated by NITEL, the AC substation, the autotransformer, the impedance bonds, the track circuit and the traction line models have been presented in detail and then assembled to build a complete model of a railway system. The effects on net-impedance behaviour of each apparatus and sub-system connected to the line have been identified and evaluated; the relationship between line length, number of autotransformers and the net impedance itself has been considered too.
The DC Substation activity has been finalised: the dc substation has been characterised as harmonic voltage source and the maximum envelope of harmonic voltages has been derived, including effects of distortion of primary voltage and of load characteristics. A simplified formula for conservative estimation of voltage harmonics has been proposed.
During the implementation of the project a series of actions aiming at disseminating the project results has been carried out. As a result articles in journals and at conferences have been published. Project results have been written down and are available in the public deliverables as listed in chapter 2 of this document. To round off the project a final dissemination conference has been held April 2009 at the UIC premises in Paris.