Numerical and Experimental Investigation of Heavy Freight Train Dynamics

Freight trains all over European Countries are equipped with mechanical couplers, which are using two buffers and a screw coupler in the centre. This configuration needs an extensive study in case of pulled mass exceeding 1600 t: actually National Association would limit train mass to 2000/2400 tons, in order to avoid excessive stresses on couplers. This challenging operative condition should become really severe in presence of switches or sharp radius curve, especially considering that freight trains are able to sustain deceleration in emergency braking condition over 1 m/s2. As well known, in order to investigate the safety issue of heavy freight trains under severe braking/traction conditions (i.e. emergency braking) also negotiating turnouts or sharp curves, an experimental approach can be performed, using a suitably instrumented trainset. Unfortunately, this approach is usually very expensive and does not provide a full understanding of the problem: the information gained with experimental tests would apply only to the particular trainset composition and to the specific track considered. On the other hand, the adoption of numerical simulations, in connection with the experimental tests, can be a useful approach to extend obtained results to a wider range of conditions, allowing an easier variation of the different test parameters. The paper will deal with the investigation performed by Politecnico di Milano together with R.F.I. (Rete Ferroviaria Italiana, the Italian Railway Network operator) on the heavy freight train dynamics. An experimental approach has been used in order to investigate the typical operative condition of a freight train; a freight wagon has been equipped with load cells, displacement transducers on the buffers and tri-axial accelerometers on the wagon frame. Moreover the traction/braking torque applied by the locomotives have been measured. The experimental trainset was composed by two heading locomotives, a series of freight wagons (pulled mass 1600 t), and then another locomotive at the end of the train. The results of the test allowed a better comprehension of the behaviour of the complete trainset, on a steep line, especially during sharp curve negotiation (R = 200 – 300 m): particular attention was paid on the buffer behaviour, because of its fundamental importance for the running safety of the wagons. It was highlighted that the operative condition typical of a sharp radius curve negotiation leads to a stiffer buffer: the increased stiffness of the buffer cannot be neglected for the investigation of the running condition. These tests were used to update an existing numerical software for the analysis of the longitudinal dynamics of heavy freight train, named T.S.Dyn. (Trainset Dynamic Simulator): its numerical model is able to reproduce forces/displacement in the coupling between two adjacent vehicles (buffers and draw gear) of a trainset. Moreover the model is able to consider the dynamic behaviour of the pneumatic braking system of the entire trainset and for this reason it can find proper application even for the simulation of severe braking condition (i.e. emergency braking). The numerical model has been updated taking the advantage of the experimental activity performed and it was implemented in T.S.Dyn code. A comparison between numerical and experimental results will be described in the full paper.