| China is currently in a critical period of energy transition.As an important low-carbon energy source,the overall installed capacity of wind turbines continues to climb in the context of carbon peaking and carbon neutrality.At the same time,the research of unit operation and maintenance technology and fault diagnosis methods under large-scale wind power grid connection is of great significance to the safe and stable operation of wind turbines and even power systems.However,the influence of wind speed and equipment parameters during the operation of actual wind turbines can trigger certain uncertainties in the fault characteristics of wind turbines,making the application of existing fault diagnosis methods in engineering practice face great challenges.Based on this,this paper,funded by the National Natural Science Foundation of China and the Shaanxi Provincial Natural Science Basic Research Program,carried out a research on the fault diagnosis method of wind turbine bearings taking into account uncertain fault characteristics,as follows.Firstly,this paper discusses the necessity of wind turbine bearing fault diagnosis on the basis of a brief introduction of wind turbine structure.The analysis focuses on the change of fault characteristics due to the change of ambient wind speed and bearing parameters in complex environment,which lays the foundation for the later research.Secondly,to address the problems of severe operating conditions and noise interference of wind turbines,a wind turbine vibration signal noise reduction method based on the similarity of EMD continuous geometric distribution is proposed by measuring the difference between noise and fault characteristic probability density distribution,which effectively suppresses the interference of noise to wind turbine fault diagnosis methods.Thirdly,for the problem of uncertain frequency of wind turbine fault features,this paper searches for fault features by constructing different spectral interval energies,transfers noise energy to low-frequency fault features by using signal noise resonance,and proposes an adaptive optimal resonance-based wind turbine uncertain fault feature enhancement method to achieve fault diagnosis of wind turbine with uncertain fault features.Finally,the algorithm is verified by applying the fault experimental signals of typical wind turbine rolling bearings.The experiments show that the proposed algorithm can effectively perform noise reduction of wind turbine vibration signals under strong background noise and can achieve feature enhancement of uncertain fault characteristics of wind turbines. |
