Feasibility study of a rail vehicle damper fault detection by artificial neural Indexed by: networks

The aim of the study was to investigate rail vehicle dynamics under primary suspension dampers faults and explore possibility its detection by means artificial neural networks. For these purposes two types analysis were carried out: preliminary 1 DOF model a second one - passenger coach benchmark tested in multibody simulation software MSC.Adams with use VI-Rail package. Acceleration signals obtained from latter served as an input data into network (ANN). ANNs different number hidden layers capable detecting for trained fault cases, however, achieved accuracy below 63% at best. These results can be considered satisfactory considering complexity dynamic phenomena occurring vibration system vehicle.