The Fault Monitoring And Diagnosis Of Wind Turbines’ Key Components Based On Stator Current

With the rapid development of wind power industry as well as the stability of the wind turbine system, ease of maintenance and other aspects of the requirements, condition monitoring and fault diagnosis technology of wind turbine system cause the extensive concern of the academia and industry. Gearbox and generator were used as key components of wind turbine, there has a realistic significance to make a condition monitoring and fault diagnosis to them. Currently, the vibration signal analysis is the main method of wind turbine fault monitoring and diagnosis, but this method has a high equipment costs, the installation of the sensor is not convenient and the interference is larger. In contrast, stator current signal analysis method with susceptible to interference, easy to collect signals, high signal-to-noise ratio, and can realize online monitoring, has been used as an important means by experts and scholars at home and abroad for fault diagnosis. Therefore, this paper uses the stator current signal analysis method for wind turbine gearbox and rotor fault diagnosis. The specific research ways are as follows:Firstly, the system configuration and common faults of the wind turbine are introduced in detail, and the generator rotor broken bar fault mechanism and wind turbine gear fault vibration characteristics and stator current detection principle are analyzed in detail, in order to provide the basis for wind turbine rotor broken bar fault detection and gear pitting fault detection.Secondly, according to the failure mechanism of the broken rotor bar, the spectral subtraction is introduced into the stator current spectrum analysis with the analytic wavelet analysis combines to realize the broken rotor bar fault feature extraction and fault detection under load mutation. By numerical simulation signal and model simulation signal to verify the validity of the proposed method. Further to introduce fault level factor to quantify the broken rotor bar fault severity degree.Finally, according to stator current signal characteristics of the wind turbine gear fault, this paper combines fault feature extraction based on empirical mode decomposition(EMD) and fast independent component analysis(Fast ICA) with a sample entropy algorithm for gear pitting detection to effectively quantify the gear fault feature vector. Through the simulated signal to verify the validity of fault feature extraction algorithm based on EMD and Fast ICA method, and to verify the validity of the sample entropy algorithm used to quantify the time sequence complexity. The proposed method is vertified by experiments on a real gearbox under different motor rotating speeds.