| Rolling element bearings is widely used in various industrial field, its running states will affect the whole machine's function. Therefore, it is great important for its early fault diagnosis. For the early failure of the rolling element bearings, the signals contain the fault information which is very weak in general and often submerged in the noise signal, therefore, it is the key to reduce the noise and enhance the weak fault signal.The purpose of this research project is to propose spectral kurtosis combined with EMD (Empirical Mode Decomposition) method applied to the early fault diagnosis of rolling element bearings. Based on others'research, we introduce spectral kurtosis and EMD method to this area. Due to the data acquired from the vibration of the rolling bearing, we carried out the weak signals of early failure enhancement; improve the level and efficiency of fault diagnosis and keep the stability and safety of machine system. Main contents in this dissertation are as follows:First, there is a lack of self-adaptive for the spectral kurtosis and EMD modal aliasing, resulting in low frequency band component issues such as excess IMF, so propose the spectral kurtosis combined with EMD method. Spectral kurtosis is very sensitivity to fault pulse signal which can enhance the fault pulse and reduce noise, it can effectively improve the efficiency and accuracy of EMD decomposition, and further increases fault diagnosis.And then the spectral kurtosis combined with EMD method is applied to simulation of the pulse signal with noise and early rolling bearing fault diagnosis, the experimental results show that the spectral kurtosis combined with EMD method can effectively enhance the weak mechanical fault features, and it is very suitable for feature enhance of early rolling element bearings faults. Therefore, the method has a great significance for early rolling bearing fault diagnosis and it has a good application prospects.
|
PDF Full Text Request