Abstract: Diagnosis and analysis of bearing creep faults in wind turbines by extracting time-domain waveform and spectral features. Firstly, establish a comparative analysis standard based on the time-domain waveform and spectral characteristics under normal conditions. Secondly, a comparative analysis will be conducted on the creep faults of three types of wind turbine bearings. Research has found that the vibration signals of the high-speed shaft rotation frequency and its three times the vibration frequency of the gearbox are the main fault characteristics. When a fault occurs, vibration characteristic indicators such as effective value, kurtosis, and peak value show significant sudden trend cracking, exceeding the mean range of normal data. Although the characteristics of this type of fault are similar to those of gear box highspeed shaft small tooth breakage faults, there are significant differences in time domain and frequency spectrum. Through in-depth analysis of the characteristics of bearing creep faults in wind turbines, the aim is to provide useful references for timely diagnosis and effective prevention of such faults.