EEMD And Hilbert Square Demodulation Based Research On Fault Diagnosis Of MW Wind Turbine Gearbox

There is a growing interest in renewable energy systems with the shortage of fossil fuels and serious environment problems.As a clearly and renewable resource,wind energy is developed rapidly worldwide for its owns advantages.By the end of 2014,the global wind power total installed capacity is 370 GW.Along with the rapid development of wind power,how to ensure the stable operation of the wind generator and resolve the failure of the wind turbine are problems that must be resolve immediately.Gearbox,as the core component of wind turbine,is responsible for smooth operation of wind turbines.Moreover,the availability of the whole system depends on the serviceability of the gearbox.Vibration signal,as a carrier of fault feature of drive train,can effectively reflect most of the fault information of the wind turbine gearbox.In order to ensure safety and operation reliability of the wind turbine,wind turbine gearbox fault diagnosis is researched based on vibration analysis in this thesis.The main contents are as follows:(1)This thesis describes the research status of wind turbine gearbox fault diagnosis,and summarizes the characteristic and deficiency of gearbox fault diagnosis technology.(2)The common fault type and failure forms of gear box especially the highest failure rate components such as tooth and bearing are analyzed.Then the fault characteristic parameters of the high failure rate elements are calculated to provide theoretical foundation for the studied of signal processing.(3)Dynamics simulation of the wind turbine gearbox transmission systems has been carried on in Adams and ANSYS Workbench software in order to better understand the characteristic of the fault vibration signal,and make an analysis on the setting of related parameters and the method of hexahedral mesh generation.(4)According to the feature of vibration signal,fault diagnosis method based on EEMD and Hilbert square demodulation has been proposed and a principle of selecting proper sampling frequency for raw signal re-sampling is developed in order to increase the accuracy of ensemble empirical mode decomposition.Finally,vibration signal collected from our test and Case Western Reserve University and laboratory test stand of ZhongZiQingAn new energy company is used to estimate the fault diagnosis effect of this method,and the results demonstrate that the fault feature can be extracted by the proposed method...