| Nowadays, as the protect consciousness of the environment is strengthened, kinds of vacuum pumps are more and more applied. The bearing is a common part of vacuum pumps. So, the fault diagnosis for bearings is imperative under the situation. The traditional analyzed-spectrum means in fault diagnosis is power spectrum analysis. But when the rate of the frequency differentiate is low, it can not be power spectrum differentiated. Zoom-FFT method is an analyzed-spectrum means which can enhance the rate of frequency differentiate. It is applied to fault diagnosis of the bearing of vacuum pump, which make sense in theory and practice.In this dissertation the vibrant mechanism of bearings are summarized. The common bearing fault diagnosis methods and their applied occasion are introduced detailedly. The emphasis is the means of response-spectrum, self—power spectrum and envelope spectrum.The two methods of time-frequency analyze methods are expatiated. The principle of wavelet transform, short-FFT and FFT are introduced detailedly. And several wavelet functions are expatiated.Several Spectrum zoom methods that are in common use are described exhaustively. To compare wavelet transform Zoom-FFT with complex modulation Zoom-FFT, MATLAB emulational is used, which validates wavelet transform Zoom-FFT is better. The FFT—FS is compared with power spectrum, which validates the former not only has higher rate of the frequency differentiate, but offset it's limitation that as the rate of the frequency differentiate decreasing the length of the data increases.Applying Spectrum zoom method to fault diagnosis is researched particularly. After gathering fault vibrant signal of vacuum pumps with two sensors, the wavelet transform Zoom-FFT and FFT—FS are done zoom spectrum for needed frequency band of vibrant signal, which finds the fault is in inner race of the bearing. The results of test show that this method will have a higher accuracy in fault diagnosis research.
|
PDF Full Text Request