Research On Circular Domain Resampling And Fault Feature Analysis Of Damage Signal Of Low-speed Swing Bearing

Low-speed swing bearings are widely used in space launch tower rotary system,wind turbine rotor and yaw system,tunnel boring machine and tower crane jib,etc.Considering that the failure-free operation of swing bearings is essential to ensure the safe operation of equipment,it is necessary to focus on the fault diagnosis of swing bearings.The special low-speed reciprocating motion mode destroys the periodicity of the fault signal of swing bearings and reduces the number of damage impact in a certain period of time,resulting in weak fault signal and difficult diagnosis.Therefore,based on the simulation model of swing bearing fault,the circular domain resampling and fault feature analysis methods of damage signal of swing bearing are studied in this thesis.The main research contents include:(1)The impact phase characteristics are analyzed according to the motion law of swing bearings,and the impact amplitude modulation characteristics are analyzed in combination with the load distribution law of swing bearings.On this basis,a numerical simulation model of local damage of swing bearings is constructed,and the simulation analysis of local fault signals is carried out,laying a data verification foundation for subsequent studies.(2)The circular domain resampling and the circular domain segmentation method of vibration signal are studied for the change of swing bearing speed.Firstly,the time-domain signal is envelope demodulated,and then the equal-time-interval sampled signal is converted into equal-angle interval circular domain signal based on the confidence signal.Short signals in equal length circular domain of single swing are obtained by segmenting signals in unequal length circular domain using circular domain segmentation method.It aims to eliminate phase error and lay a foundation for the next step of autocorrelation population average.(3)A fault signal enhancement method of swing bearings based on piecewise accumulative approximation and autocorrelation total average is proposed to further eliminate phase errors caused by reversing and enhance weak damage signal characteristics.The simulation results show that this method can effectively extract the fault features of swing bearings and quickly identify the fault signature signal of swing bearings.(4)The circular domain resampling and fault feature extraction method proposed in this thesis are verified by the actual test data in the laboratory environment.The analysis and processing results of simulation signals and experimental signals show that the proposed method can improve the signal-to-noise ratio and effectively enhance the fault signature signals of swing bearings under special motion modes.The above research work can provide a technical basis for the state monitoring and fault diagnosis of swing bearings.