The Research Of Rail Vehicle Faulty Bearing Simulation And Sparse Fault Diagnosis Method

In recent years, the rapid development of the rail transportation, especially the highspeed railway and subway, has brought much convenience to our daily life. Meanwhile, it is very important to maintain the vehicle to ensure its safety operation. As the key components and the main fault ones of the vehicles, rolling bearings have been widely used in rail vehicles. It is of great significance to develop the techniques to detect and diagnose the rolling bearings’ status to ensure the safety operation of the vehicles.This research is supported financially by the Natural Science Foundation of China "Signal Transients Extraction under the Frame of Sparsity and Its Application in Rotating Machine Fault Diagnosis"(No. 51375322). With the aim of rail vehicle rolling bearing fault diagnosis, this thesis builds up the rail vehicle faulty bearing simulation and fault detection system, and proposes a sparse decomposition method by using double-TQWT to extract the transient features in the vibration signal. The theoretical research and the application research are studied in the paper.Effective simulation of the faulty bearings is vital to learn the fault mechanism, and is also the fundamental to verify the effectiveness of the fault diagnosis method. The rail vehicle faulty bearing simulation and fault detection system is built up according to the rail vehicle structure, so that the fault diagnosis methods can be tested on the system. The simulation system can conduct bearing single-fault or multi-faults simulation experiments under different speeds or different loads.To overcome the disadvantages of the traditional linear methods and the frequencybased signal decomposition methods, a sparse decomposition method using double-TQWT is proposed in this thesis. By using the double-TQWT, the transient features can be decomposed into low resonance component as it has a low Q-factor. Applications in extracting fault features of bearing fault signals show that the proposed method outperforms the average filtering method, the wavelet threshold algorithm, and the EMD.Finally, the bearings with the localized fault on the outer race, the rolling and the inner race are tested in the rail vehicle faulty bearing simulation and fault detection system, and the vibration signals are collected accordingly. Then, the proposed sparse decomposition method is applied to extract the transient features of the vibration signals. The effectiveness and the feasibility of the proposed method are verified by the applications.This thesis focuses on the study of the rail vehicle rolling bearing fault diagnosis by combining the simulation test and sparse fault diagnosis method. Results show that the rail vehicle faulty bearing simulation and fault detection system can successfully simulate the operation of the rail vehicle with faulty bearings. And the proposed sparse decomposition method by using double-TQWT is effective to extract the transient features, which has been verified by the simulation system. It is of great value to study the rail vehicle rolling bearing fault diagnosis method based on sparse decomposition.