Fault Diagnosis Of Locomotive Bearing Based On Mode Decomposition

In recent years,China's high-speed railway has been developing rapidly,and its operating mileage is increasing,and its operation speed has been constantly broken.Therefore,safety issues in the operation of high-speed railway have become the focus of attention.Rolling bearing is an important component of the locomotive,which is the most easily damaged parts.The damage fault of rolling bearing,such as peeling,pitting,defect and scratch,can rapidly increase,cause bearing failure,and have the greatest harm to the bearing system,which is directly related to the safe operation of the train and the safety of the passengers.Therefore,monitoring the state of rolling bearings and finding faults in time is a great significance.Because of the low Signal-to-Noise Ratio of the early vibration signal of the rolling bearing,the fault signal is often drowned by the strong noise,which is difficult to be extracted from the noise component.Therefore,the diagnosis of early damage of rolling bearings is always a difficult problem in this field.At present,many methods have been applied in the field of fault diagnosis of rolling bearings.One of the common diagnostic methods is to extract some characteristic parameters that contain the fault information from the vibration signal to reveal the fault type,fault location and fault level.Therefore,how to effectively extract the characteristics of the fault is particularly important.The fault characteristic parameters are divided into time domain parameters and frequency domain parameters.The common methods of extraction are Empirical Mode Decomposition(EMD)and Variational Mode Decomposition(VMD),which have been applied more and more widely in recent years.In this thesis,the research status and research methods of rolling bearings are introduced,and the vibration mechanism of rolling bearings is analyzed.By analyzing the frequency characteristics of the rolling bearing,the impulse signal of the fault defect of the rolling bearing is simulated,and a certain amount of Gauss white noise is added to make the impulsive signal completely submerged in it.The two decomposition methods of EMD and VMD are used to decompose the rolling bearing signals respectively.Then,the kurtosis values of each Intrinsic Mode Function,(IMF)obtained from the decomposition.Since each IMF contains rolling bearing failure information,if the value of the IMF's kurtosis is larger,the more information it contains.Therefore,a IMF with the maximum kurtosis is selected for Hilbert transform and Teager Energy Operator(TEO)demodulation.The fault information of the rolling bearing is extracted by the obtained envelope spectrum and spectrum diagram,and the advantages of the four methods are obtained by comparing the envelope spectrum and the Peak Signal to Noise Ratio(PSNR)of the spectrum diagram.The method is applied to deal with the data of rolling bearing in Case Western Reserve University(CWRU),so as to illustrate the effectiveness and feasibility of the method used.Because the rolling bearing experimental platform built by Case Western Reserve University tends to be idealized.In order to further verify the method of simulation closer to actual working condition of the runtime rolling bearing,set up the experimental platform to collect,analysis and processing the experimental data.The effectiveness and superiority of the method are explained.