Research On Fault Diagnosis Technology Of Planetary Gearbox Under Complex Conditions

As the global wind power industry developed rapidly,the installed capacity and the monomer size of wind turbines increases rapidly,as well as their wind load capacity.Under complex working conditions,the planetary gearbox in the wind turbines is relative prone to failure,costly to maintain.So the requirement of real-time and accuracy for the fault diagnosis of the wind turbine is more and more high.The research on the fault diagnosis technology of the planetary gearbox of the wind turbine is of great significance to guarantee the reliability and safety,and to reduce the maintenance cost.In this paper,the fault diagnosis technology of the planetary gearbox based on vibration signal analysis is studied:(1)In order to solve the problem that the frequency spectrum of the vibration signal in the variable speed condition is ambiguity,a fault diagnosis method based on the NLSTFT instantaneous frequency estimation for tacholess order tracking and adaptive VMD with noise reduction is presented.Firstly,in order to eliminate the effect of speed fluctuation on the vibration signal,NLSTFT time-frequency analysis method of the iterative estimation of the instantaneous frequency is used.Then the instantaneous frequency curve is transformed into the instantaneous phase curve.Thus the tracking of tacholess order is conducted.And the VMD method for decomposing frequency components is used to diagnose the faults without using the mode superposition.In order to solve the problem that the parameters of VMD are difficult to be determined manually and are sensitive to strong noise,PSO algorithm is used to optimize the VMD parameters.The OGS algorithm is used to compress and reduce the noise signal before the VMD decomposition so as to overcome the influence of noise.The simulation and the actual experiment results show that the proposed algorithm can effectively realize the planetary gearbox fault diagnosis under the condition of varying speed.(2)In order to solve the problem of low fault diagnosis rate in planetary gearbox under complex conditions,a fault diagnosis method based on diversity multi-criteria fault feature selection of DMCFS and heterogeneous model ensemble learning model of HE-BS is proposed.Firstly,the fault features are extracted based on the collected vibration signals.Considering a number of feature related factors,4 kinds of feature evaluation criteria are defined,and the mathematical model of multi criteria fault feature selection is established;Then,the MOEA/D optimization algorithm is used to solve the model,and a set of multiple fault feature subsets is obtained;Finally,based on multi-group diversity fault feature subset and heterogeneous model ensemble learning to achieve information fusion fault diagnosis.Experimental results on the UCI standard data test set and experiment signal show that the proposed algorithm can significantly improve the accuracy of fault diagnosis and robustness to noise.