INVESTIGATION INTO THE CALCULATION OF SUPERELEVATION DEFECTS ON CONVENTIONAL RAIL LINES

Abstract

Railway curves have influence on train speed on a curve and/or wheel/rail interface. Additional forces that have to be compensated appear in the curves. The purpose of superelevation is to compensate acceleration emerging in the curve thus assuring comfortable passenger transportation and equal wearing of both rails. However, it is very difficult to calculate superelevation when designing and maintaining a railway track, because the estimation of actual train speed on the curves is very complicated. As we know, railway lines can be divided into conventional, high speed and heavy haul ones. As these lines are absolutely different, requirements for the installation and maintenance of the track may also differ. Conventional rail lines are the object of research discussed in this article. The speed of freight and passenger trains is different on conventional rail lines, which is an essential factor in determining superelevation. On the ground of scientific researches, the article analyzes and evaluates the factors influencing wheel/rail interface on the curves. The paper also deals with railway line curves, superelevation and uncompensated lateral acceleration. The article presents the method used in Lithuania for calculating superelevation in the railway curves and analyzes calculation defects. For research purposes, analytical and statistical methods have been used. The obtained results have shown that actual superelevation in the researched curves does not match the calculated one. The calculations and obtained results of superelevation depend on how average train speed in the curves is estimated and used for calculations. As most of the results show that even small variations in the curve have a great influence on track/vehicle behaviour, it is necessary to find more precise methods for calculating superelevation, evaluating actual train speed and considering permissible uncompensated lateral acceleration in the curves.