Railway Axles Condition Monitoring Technique Based on Wavelet Packet Transform Features and Support Vector Machines

Abstract

Railway axles are critical to the safety of railway vehicles. However, railway axlemaintenance is currently based on scheduled preventive maintenance using Nondestructive Testing.The use of condition monitoring techniques would provide information about the status of the axlebetween periodical inspections, and it would be very valuable in the prevention of catastrophicfailures. Nevertheless, in the literature, there are not many studies focusing on this area and thereis a lack of experimental data. In this work, a reliable real-time condition-monitoring technique forrailway axles is proposed. The technique was validated using vibration measurements obtained atthe axle boxes of a full bogie installed on a rig, where four different cracked railway axles were tested.The technique is based on vibration analysis by means of the Wavelet Packet Transform (WPT) energy,combined with a Support Vector Machine (SVM) diagnosis model. In all cases, it was observed thatthe WPT energy of the vibration signals at the first natural frequency of the axle when the wheelset isfirst installed (the healthy condition) increases when a crack is artificially created. An SVM diagnosismodel based on the WPT energy at this frequency demonstrates good reliability, with a false alarmrate of lower than 10% and defect detection for damage occurring in more than 6.5% of the section inmore than 90% of the cases. The minimum number of wheelsets required to build a general model toavoid mounting effects, among others things, is also discussed.