Dynamic Feature Extraction And Recognition Of Tooth Crack In Planetary Transmission System

Because of it is compact structure and large transmission ratio,planetary system is employed in the cutting part and the traction part of the shearer.Shearer is used in low-speed,alternating,heavy load conditions,coupled with insufficient lubrication and other problems,which often lead to the failure of planetary gear train in the cutting part,which directly affects the efficient mining.Serious faults of planetary gear system for shearer are all developed from the early root cracks,so the key to ensure the reliable operation of the planetary gear system is to accurately identify the fault of tooth crack.However,due to the fact that the planetary gear system of shearer is always in low-speed and non-stationary operation condition,its vibration signal has the characteristics of low frequency,weak energy,strong background interference and so on,which makes the fault diagnosis of planetary gear crack face great challenge.In order to solve the problem of fault feature extraction of planetary gear system in shearer cutting part,the influence of tooth cracks of different sizes on the dynamic response of planetary gear system is studied in depth,and a tooth crack model is established which extends along the depth and width of the crack.It is difficult for the pure rigid body model of the planetary gear system to accurately reflect the fault characteristics of the planetary gear system in the simulation process.Considering the calculation efficiency of the model,the sun wheel with crack fault in the planetary gear system is treated as flexible,and the rigid flexible coupling model of the planetary gear system is constructed.Solving the dynamic response of a planetary gear system,an efficient theoretical analysis method is provided for the rapid diagnosis of gear tooth crack.Aiming at the problem of weak early crack and non-stationary vibration signal of planetary gear system for shearer,a new fault feature extraction method of FRFT-VMD is proposed by using the non recursive characteristics of Variable Mode Decomposition(VMD)and the advantages of non-stationary signal analysis of Fractional Fourier Transform(FRET).The FRFT is utilized to filter the initial vibration signal and extract the effective band signal.Then the Intrinsic Mode Function(IMF)is obtained by the VMD decomposition of the signal.Finally,spectral analysis of each IMF is carried out to extract the effective eigenvalue.The simulated signal and tested signal are utilized to study study the algorithm.Based on the research of fault mechanism and feature extraction algorithm,the online monitoring and fault diagnosis system of the planetary gear system for shearer is developed by Lab VIEW and MATLAB.Through the real-time analysis of the data gathered by the vibration sensor,the feature parameters were extracted to complete the fault diagnosis.At the same time,the system also integrates FFT,envelope spectrum analysis,cepstrum analysis and other conventional analysis methods as the preliminary diagnosis strategy of the state of the planetary gear train of the shearer,to achieve multi-level monitoring and diagnosis oriented to multiple needs.The test platform of a planetary gear train is established,and the tooth cracks of different scales are made.The design test is carried out under the same test conditions as the simulation conditions,which verify the reliability of the developed online monitoring and fault diagnosis system,as well as the effectiveness of the simulation model and feature extraction algorithm.