The Extraction Of Vibration Signal Based On Multivariate Statistical Analysis And Its Application In The Fault Diagnosis Of Crucial Components In Transmission Systems

Localized fault in machine components including bearings and gears tend to result inshocks and arouse transient impulse response in vibration signal which is the dynamicforms of the machine, and the characteristic waveform changed with the evolution of fault.Thus, the extraction of signal transient feature which shows the localized fault in machinecomponents has always been the most crucial problem in machinery fault diagnosis. Thisresearch is supported financially by the Natural Science Foundation of China (No.51375322). With the aim of machinery fault diagnosis, and with the research target ofbearings and gears, which are the crucial components of transmission system, thisdissertation proposes Independent Component Analysis (ICA) transient componentextraction method based on dimensionality reduction to extract the signal transient feature.The theoretical research and the application research are studied in depth, respectively.Firstly, the failure mechanism and the characteristics of vibration signal of bearingsand gears in the transmission system are analyzed respectively, which provides thetheoretical support of rationality and necessity about the extraction of signal feature. Toensure that the theoretical research works are validated by the experiment analysis, thebearings and gears are tested under the localized fault condition and the vibration signalsare collected.Secondly, the theory, definition, calculation method and properties of PrincipalComponent Analysis (PCA) and its dimensionality reduction theory in signal featuredetection are introduced systematically, and the effectiveness of PCA dimensionalityreduction is verified by analyzing simulation signal and vibration signals of bears andgears. Then, the theory, objective function and iterative algorithms of ICA are introducedsystematically, and the characteristics of ICA are explained by simulation signal analysis. Based on the characteristics of ICA and combined with the characteristics of the transientfeature extraction, ICA transient component extraction method is studied deeply, and itscharacteristics and shortcomings are analyzed by its application in simulation signal andvibration signals of bears and gears.Thirdly, this dissertation proposes an ICA transient component extraction methodbased on dimensionality reduction, in view of the shortcomings of ICA transientcomponent extraction method without dimensionality reduction and according to thecharacteristics of PCA. The proposed method is to extract the transient signal based onICA, after reducing the dimensionality based on PCA and obtaining the main feature ofobserved signal. The validity of this method is verified by comparing the extraction resultsunder different dimension of dimensionality reduction, and the superiority is verified bycomparing this method with the mean filtering method and the wavelet thresholdingmethod.Finally, by considering the practical application in machinery fault diagnosis, theproposed technique is applied in the fault diagnosis of crucial components in transmissionsystem. For the detection of bearings fault, gearbox fault and automobile transmission fault,the proposed technique is applied in the extraction of the transients impulse response thatshow the localized fault of bearing outer race, inner race, rolling and the gears, whichfurther proves the correctness in theory and the effectiveness, applicability in practical.In this dissertation, it is confirmed that the technique of ICA transient componentextraction method based on dimensionality reduction is effective for extracting the signaltransient feature through the research on the extraction of characteristic signal based onmultivariate statistical analysis, which is of certain theoretical and practical value formachinery fault diagnosis in crucial components in transmission systems.