Due to the influence of random wind shear in the atmospheric phenomenon, the random vibration of the main shaft of the wind turbine generator is generated. This vibration signal will be mixed with the misalignment signal of the high-speed shaft, which will cause interference to the fault diagnosis. Based on the analysis of the phenomenon of wind shear and the fault, the independent component analysis was carried out on the high-speed shaft mixed vibration signals on the basis of using rapid fixed point algorithm based on kurtosis, and the weak fault information is extracted successfully. At the same time, this method was compared with the weak information extraction method based on wavelet denoising, which proved the superiority of the proposed method. The experimental results show that the method has good field applicability and has a good application prospect in the field of weak information extraction for rotating machinery of wind power generation.
Comparison of Support Vector Machine and Neural Network Classification Methods in Land Use Information Extraction through Landsat TM Data
