Two basic forces act on a railway track: vertical and lateral forces arising from trainmotion. As a result of those forces, the edges of ballast grains are crushed and it causes a decrease in internal friction but also it is a reason for ballast layer consolidation. In addition, the chemical and biochemical factors cause changes of the mechanical parameters. Moreover, the pollution of ballast grains by sand particles causes changes in the ballast parameters. When constructinga track model a problem is how to replace the different factors, mentioned previously, causing energy dissipation by a relatively simple structure. We have decided to include dry friction in the model. We have decided to substitute other factors that cause energy loss by an increase of the dry friction force.

The model of the process of railway track settlement is restricted to examining the track deterioration in the form of vertical irregularity changes that consist of three submodels. The first sub-model, is a multi-degree of freedom system. The values of the settlement obtained from it were a basis to determine parameters for simpler models. We have used the first simple model for simulation purposes, and the second one for assessing track settlement.

Using those models, look-up tables were produced for assessing track settlement for various loads per axle. We have assumed that irregularity in the welded rail joint zone was the initial factor causing expansion of vertical irregularities of a track. The measured geometry of rail joints was introduced to the simulation program. We have performed a series of simulations during which the predicted irregularities obtained from the previous simulation were introduced as the excitations for the next simulation. We took a passenger car as the object of the simulations. Simulations were performed for a thousand runs and the velocity of the simulated vehicle was equal to 120 km/h. The simulations results were compared with measured irregularities in the welded rail joint zone. It was found, that based on the analysis of the power spectrum density of the vertical irregularity signals that the length of the dominant irregularity waves obtained from simulations is shifted to longer waves. On the other hand, they have amplitudes of the same order and the standard deviation of irregularity is similar (0.0014 - for measurements, 0.0012 - for prediction).

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 £110 +P&P)