In recent years, significant studies have focused on monitoring the track geometry irregularities through measurements of vehicle dynamics acquired onboard. Most of these studies analyse the vertical irregularity and the vertical vehicle dynamics since the lateral direction is much more challenging due to the non-linearities caused by the contact between the wheels and the rails. In the present work, a machine learning-based fault classifier for the condition monitoring of track irregularities in the lateral direction is proposed. The classifiers are trained with a dataset composed of numerical simulation results and validated with a dataset of measurements acquired by a diagnostic vehicle on the straight track sections of a high-speed line (300 km/h). Classifiers based on decision tree, linear and Gaussian support vector machine algorithms are developed and compared in terms of performance: good results are achieved with the three algorithms, especially with the Gaussian support vector machine. Even though classifiers are data driven, they retain the essence of lateral dynamics.

No abstract

No abstract

No abstract

Application of active suspensions in high-speed passenger trains is gradually getting more and more common. Active suspensions are primarily aimed at improving ride comfort, wear or stability. Failure of these systems may not only just deteriorate the performance but it may also put vehicle safety at risk. There are not many studies that explain how a thorough study proving safety of active suspension should be performed. Therefore, initiating this type of study is necessary for not only preventing incidences but also for assuring acceptance of active suspension by rail vehicle operators and authorities. This study proposes a flowchart for systematic studies of active suspension failures in rail vehicles. The flowchart steps are solidified by using failure mode and effects analysis and fault tree analysis techniques and also acceptance criteria from the EN14363 standard. Furthermore, six failure modes are introduced which are very general and their use can be extended to other studies of active suspension failure. In the last section of the paper, the proposed flowchart is put into practice through four failure examples of active vertical suspension.