Research On Fault Diagnosis Of Urban Rail Train Wheelset Based On Sparse Morphology

With the development of urban rail transit network,rail transit such as subway has brought great convenience to people,so its safety issues are also attracting more and more attention.Rotating machinery in mechanical parts is the most prone to failure,and it is also the most likely to cause major accidents.In the case of casualties,in the rotating mechanical parts that are the focus of attention,the wheelsets occupy an important position.The health status of the wheelsets is directly related to the safety and stability of the vehicle operation.Therefore,the establishment of the urban rail transit vehicle wheelsets The detection and diagnosis system is very important.The core of the system should be the classification and feature extraction of wheelset faults,and build a complete fault diagnosis process.(1)This paper first introduces the research status of wheelset fault diagnosis,introduces the time-frequency analysis method commonly used in diagnosis,and then focuses on the research status of sparse decomposition method and its application in the field of mechanical fault diagnosis,and determines it as the The main fault diagnosis method,finally introduces the vibration mechanism and fault characteristics of wheel-pair scratches and polygons.(2)The early faults of the wheelset are multi-coupling,the signal-to-noise ratio is low,and the useful fault signals are mostly submerged in a large number of unwanted noise signals.According to the sparse decomposition algorithm,the signal can be effectively denoised and improved.The sparse decomposition denoising algorithm is used to verify the effectiveness of the algorithm,and the wheel-to-fault test bench is established to perform sparse decomposition and denoising on the wheel-scratch signal.After reconstruction,the signal-to-noise ratio of the fault signal can be greatly improved.Can prove the validity of this algorithm.(3)In view of the fact that the fault impact of the signal is easily submerged,the mathematical morphology method is introduced to enhance the fault feature energy of the weak fault signal.The mathematical form method combines the open-close and closed-open morphological filters,and uses the kurtosis value to Adapting to the length of the selected structural elements,the experiment proves that the fault characteristic impact signal can be effectively enhanced,and the feature extraction and recognition of the weak fault signal is enhanced.(4)Propose Sparse Morphology Analysis(SMA)for wheelset fault diagnosis.The original signal is firstly subjected to sparse decomposition and reconstruction denoising,and the shortcomings of complex time duration are calculated for sparse decomposition.The adaptive immune particle swarm optimization algorithm is used to optimize the optimization process.Then the adaptive mathematical morphology method is used to enhance the fault characteristic signal.The frequency domain transform of the sparse morphological reconstruction signal can extract the fault characteristic frequency,and the effectiveness of the simulated signal and the wheelset scratch signal is verified.Finally,the SVM is used to diagnose and classify the wheel scratch signal.(5)According to the above research content,construct the urban rail train wheel fault diagnosis system,which can carry out data collection and data analysis on the wheelset,and the experiment can prove its effectiveness.This paper explores the wheelset of the vehicle and studies the fault diagnosis algorithm for the wheel fault mechanism and signal characteristics.It can detect and diagnose relevant faults in the early faults of urban rail vehicles and other urban rail vehicles,and enhance the stability of urban rail transit.Sex and safety,reducing the cost of troubleshooting.