Research On Fault Diagnosis System Of Wind Turbine Gearbox Based On DSP

As a green and renewable energy,wind energy can solve the problem of environmental pollution and energy shortage effectively.Therefore,our nation attaches great importance to the development of wind energy and wind power techniques.Wind turbine is the key equipment gaining wind energy.Once wind turbine breaks down,it will cause vast economic loss.The bad operation conditions can impose alternating load to the core parts of wind turbine,thus damaging wind turbines.Therefore,it is very important to conduct diagnosis and maintenance of wind turbine gearbox.The diagnosis and maintenance of wind turbine gearbox mainly embodies at the below two aspects:one is to monitor the condition of wind turbine gearbox and monitoring signal features is one of the key techniques.Aimed at the drawback of the traditional method on setting the thresholds of signal features,this dissertation adopts the gold section method to optimize the thresholds;The other one is to conduct fault diagnosis of wind turbine gearbox by utilizing those commonly used diagnosis methods,such as time domain methods,frequency domain methods and time-frequency domain methods.On the basis of the existing research outcomes of fault diagnosis techniques,the dissertation has used Digital Signal Processor(DSP)and combines the software platform of Microsoft to design and develop a fault diagnosis system for wind turbine gearbox.The main research contents of the dissertation are listed as follows:(1)The dissertation has analyzed the structure and operating principle of wind turbines,investigates fault modes and fault mechanism of gearboxes and commonly used fault diagnosis methods,thus providing theoretical foundation for the design and development of the system.In addition,the requirement and function analysis of the fault diagnosis system has been carried out,thereby establishing the whole frame of the system.(2)In consideration of the drawback of the traditional method on setting the thresholds of signal features,the dissertation has used the gold section method to optimize the thresholds in the threshold range set by Beta distribution and normal distribution,thus obtaining optimal thresholds.The analysis results of the experimental data show that the leaking alarm and false alarm could be reduced to a large extent by using the optimal thresholds.Then the proposed method has been applied to the fault diagnosis system in order to monitor and maintain the wind turbine gearboxes well.(3)Based on the DSP(type:TMS320C6748)produced by TI(Texas Instruments)Corporation,the dissertation has developed the hardware system comprised data acquisition,digital signal processing and data transmission and finished the testing and validation of the algorithms used in the system.(4)A software system with abundant functions,high practicability which includes condition monitoring and fault diagnosis has been developed on the Microsoft platform,and the system has also been used to diagnose the faults of roller bearings of wind turbine gearboxes in some wind farms and good results has been obtained.