Abstract
The axial force of a continuous welded rail(CWR) causing buckling and break of the rail should be controlled appropriately and adequately. In order to measure the axial force of the CWR quantitatively and in a simple manner, a method of presuming the axial force based on detection of the change of the natural frequency according to the axial force is developed. However, the accuracy of this method is insufficient because not only the axial force but the variations of track conditions such as the state of the rail and the rail fastening devices affect the natural frequency of the rail. In this paper, to investigate the influence of the variations in track conditions and its correction method for accurate measurement, we develop a finite element track model. Numerical experiments show that the wear of the rail head and the variation of the sleeper bays in the range of approximately 30 meters around the measuring point affect the natural frequency. Furthermore, we measured the axial force and the natural frequency of the CWR under various temperature conditions in an actual track and compared them with the numerical results. Through these analyses, it was confirmed that the axial force is in proportion to the natural frequency. Moreover, the consideration of the wear of the rail head and the variation of the sleeper bays in the track makes it possible to reproduce the difference of the absolute value of the natural frequency between two measurement points. By contrast, the sensitivity of the natural frequency to the axial force in the experiment is higher than that of the numerical results. Furthermore, the sensitivity of the experimental results in winter is higher than that in summer.
