1. Time domain fault diagnosis method
Time domain statistical characteristics are the most commonly used monitoring parameters in vibration monitoring of wind turbine main bearings, which can effectively capture wind turbine bearing faults and avoid vicious accidents. . When the main bearing of a wind turbine fails, the vibration amplitude of the bearing will increase significantly and a corresponding impact signal will be generated. The vibration amplitude change trend is represented by the mean value Resulting in irregular vibration, this parameter is not sensitive to early bearing failure. The kurtosis xq, peak value xP-p and pulse index I are very sensitive to the tiny impact present in the vibration signal, which improves fault identification. The peak index xP-p is usually used to detect impact vibration caused by bearing peeling, cracks, etc., while the kurtosis index xq is used for the earliest fault diagnosis of the bearing. The waveform indicator K is often used to detect mechanical failures caused by discrete defects such as pits, scratches, peelings and scratches in various bearing components. This type of failure does not have excessive total waveform energy, but has high peak values. The larger the waveform index value is, the more serious the bearing failure is.
2. Frequency domain fault diagnosis method
In the process of fault diagnosis of wind turbine main bearings, the fault characteristics of rolling bearings are usually modulation phenomena, and the time domain waveform of the vibration signal is relatively complex and cannot be Intuitively express fault signal characteristics. The vibration frequency signal is objective and can better reflect the basic characteristics of the vibration signal. The time domain vibration signal is converted into a frequency domain vibration signal through Fourier transform, and the spectrum can intuitively reflect the energy size, frequency composition and phase of the vibration signal. However, the frequency domain fault diagnosis method is only suitable for stationary signals. Since the Fourier transform method is a global transformation, the sampling frequency of the system is directly related to the resolution. It is impossible to obtain the specific frequency corresponding to a specific time and the corresponding occurrence time of a specific frequency. Therefore, the frequency domain The fault diagnosis method is not representative for the analysis of non-stationary vibration signals.
3. Time-frequency fault diagnosis method
Both the time domain fault diagnosis method and the frequency domain fault diagnosis method have certain limitations, which cannot make the vibration signal comprehensive and local. The performance was well reflected, so a new diagnostic method was proposed, which is the time-frequency fault diagnosis method. The time-frequency fault diagnosis method organically combines the time domain and frequency domain into a video phase plane to obtain vibration signal spectrum diagrams at different times. Currently, the time-frequency fault diagnosis methods that are widely used include Hilbert-Huang transform (HHT) and envelope mediation method. The envelope demodulation method uses envelope detection and spectrum analysis of the envelope signal, and then diagnoses and identifies faults based on the demodulated spectrum peaks. The envelope demodulation method is particularly suitable for high-frequency impact vibrations. So far, the envelope demodulation method is still the only effective and important analysis method for high-frequency impact vibrations. The envelope demodulation method is mainly used for high-precision fault diagnosis of the main bearings of wind turbines. It can not only diagnose the fault location, but also determine the fault type.