Research On Fault Diagnosis Of Rolling Bearings In Wind Turbines Based On Vibration Monitoring

With the active development of renewable energy, wind energy is one of the more mature renewable energy sources, which has attracted more and more attention from all over the world. However, with the continuous growth of the production of wind turbine capacity, wind turbine operation and maintenance, monitoring and fault diagnosis gradually become wind power industry is badly in need of solving problems, rolling bearing is a key components in wind turbine drive system, its running state monitoring and fault diagnosis to the whole wind turbine diagnosis is of great importance. Therefore, this paper using the wind turbine rolling bearing as the research object, in view of vibration monitoring, a rolling bearing vibration signal’s denoising method based on complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN) and Spectral Kurtosis algorithm is proposed. At the same time, based on the synchrosqueezing transformation(SST) spectral entropy algorithm for fault feature extraction of bearing vibration signal was achieved, using the support vector machine(SVM) classification and identification of the bearing fault state, further, on the basis of LabVIEW to build wind turbine condition monitoring and fault diagnosis system. The specific work of this paper is as follows:First of all, aiming at the bearing vibration signal is easy to be disturbed by the noise, the characteristic is not easy to be extracted, research on the rolling bearing vibration signal denoising method of combined CEEMDAN and spectral kurtosis. The CEEMDAN algorithm used for vibration signal’s decomposition, selected appropriate component reconstruction using cross correlation method, spectral kurtosis algorithm is used to determine the parameters of the optimal filter, through the actual data of bearing of noise reduction method verified.Secondly, a method for quantitative analysis of fault characteristics of rolling bearing vibration signal based on synchrosqueezing transformation transform spectrum entropy is proposed, realize the running state of the bearing classification based on support vector machine intelligent fault diagnosis classification model is presented. The validity of the method is verified by comparing the simulation bearing vibration data and bearing vibration test data.Finally, the actual demand for the project to build wind turbines rolling bearing condition monitoring and fault diagnosis system based on LabVIEW. Complete the overall design of the system on demand based on the overall system, the rolling bearing vibration signal acquisition and storage function, online monitoring and early warning function, offline fault feature extraction and fault diagnosis function.