Research On Composite Fault Diagnosis Method For Rolling Bearings Of Wind Turbines

In recent years,wind energy has received widespread attention as a renewable energy source worldwide.However,with the large-scale development of wind power generation,the fault problem of wind turbines has become increasingly prominent.Rolling bearings are one of the indispensable components in wind turbines,and are prone to failure due to the harsh operating environment.And during the operation of the actual wind turbine,due to the characteristics of its own structure and the harshness of the working environment in which the bearing is located,different parts in the bearing may be locally damaged at the same time,which is a composite failure state.The compound fault signal is not a simple superposition of a single fault signal,but the fault characteristics of different components overlap each other and interfere with each other,which makes the vibration signal more complicated,so it brings great difficulty to the complex fault feature extraction and fault diagnosis of rolling bearing of the wind turbine.In view of the above problems,this article uses rolling bearings of wind turbines as research objects to carry out fault diagnosis research on bearings.The main research contents are as follows:1.Analyze the vibration characteristics of local faults of each component of rolling bearing,and mainly study the situation that when there are compound faults in rolling bearings,different faults will stimulate the same or different resonance frequency bands,and establish a mathematical model to simulate the simulation.The rolling bearing compound failure provides a theoretical basis.2.Research on composite fault diagnosis of rolling bearing of wind turbine based on optimal Morlet wavelet filtering.The band-pass filtering characteristics of Morlet wavelet filtering are studied.The exponential attenuation component can accurately match the typical fault pulse component in the bearing vibration signal,so it is used to extract the features of rolling bearing faults.However,Morlet wavelet filtering is used to band-pass filter the fault signal.The center frequency and bandwidth need to be determined.Therefore,the artificial fish swarm optimization algorithm(AFSA)is used to optimize the two parameters with the correlation kurtosis as the objective function to obtain the optimal Morlet.Wavelet filter,Then the optimal Morlet wavelet filter is used to band-pass filter the bearing composite fault vibration signal.At the same time,the 1.5-dimensional Teager kurtosis spectrum has good demodulation and noise suppression performance,so it is used as a subsequent processing method for the filtered signal.The fault characteristic frequency is extracted to diagnose the bearing fault.3.Composite fault diagnosis of rolling bearing of wind turbine based on adaptive multi-point optimally adjusted minimum entropy deconvolution.In response to bearing compound failure,different fault features are staggered,overlapped and interfere with each other,and the type of bearing failure is difficult to judge.The MOMEDA algorithm is used to separate composite faults,but the algorithm is severely affected by the two parameters of filter length L and fault period T.In order to solve this problem,this chapter extracts the adaptive MOMEDA composite fault feature extraction method for wind turbine bearings.The multi-point kurtosis of the envelope of the vibration signal is used as the objective function in the artificial fish swarm algorithm(AFSA).The adaptive optimization of the filter length L and the failure period T is realized.The simulation and engineering example analysis results show that the method can effectively and accurately realize the separation of composite faults,so as to accurately identify the composite fault status of the wind turbine bearing.