The author has developed a method to predict vehicle dynamics. This method is based on the system identification theory. Because this method can be applied only in the case that the velocity of the train is constant, the author has developed a new method to predict vertical vehicle acceleration for variable velocity. This method is based on the characteristic of vehicle dynamics that the lower eigenfrequency of the vehicle depends on time frequency and the higher eigenfrequency depends on spatial frequency. To predict the acceleration in the lower eigenfrequency, the author has developed a new algorithm to convert spatial datasets to time datasets and to reconvert time datasets to spatial datasets. The acceleration in the higher eigenfrequency can be predicted with the former methods.

Copyright (c) 2006 Railway Technical Research Institute