Research On Gear Typical Fault Diagnosis Based On GHM Multiwavelet Transform And MCKD

As a key component of rotating machinery,gears are mainly used to transfer torque and change the direction of movement.The working state of gears directly affects the safety and stability of machinery.The gear has a high failure rate under complex working conditions or long-term operation.Therefore,it is of great significance for ensuring the safe operation of machinery and improving production efficiency to monitor the running status of the gear and diagnose the gear faults online.Typical faults of gears mainly include:tooth breakage,pitting corrosion and wear.When gears fail,they usually cause periodic vibration of equipment.Conventional signal processing methods such as time domain and frequency domain cannot accurately extract fault information from non-linear and instantaneous abrupt non-stationary vibration signals containing more noise components,which increases the difficulty of gear fault diagnosis.This paper presents a gear fault diagnosis method based on GHM multiwavelet transform and MCKD.First of all,according to the vibration mechanism of the gear,to establish a simplified vibration physical model of the gear transmission system and use it as a model basis.Analyze the modulation phenomenon generated by the fault signal with large impact energy,and establish the modulation model of the fault signal,and analysis the spectral characteristics of typical fault vibration signals of gears with modulation phenomena.Secondly,the basic theory of maximum correlation kurtosis deconvolution(MCKD)is studied and a gear fault diagnosis method combining GHM multiwavelet transform with MCKD is proposed.Taking gear pitting fault as an example to simulate the gear vibration signal,and the simulation signal is used to verify the effectiveness of the fault diagnosis method.Then,a typical gear fault simulation experiment platform and a data acquisition system are set up to collect vibration signals of non-fault working conditions,gear pitting faults and partial broken teeth faults,and GHM multi-wavelet decomposition and reconstruction methodsare used to process the signals,then MCKD algorithm is used to process the signals,and finally envelope spectrum is analyzed to extract the fault frequency.Experimental results show that the combination of GHM multiwavelet transform and MCKD algorithm can accurately extract the fault frequency of gears from vibration signals with multiple double-sideband and multiple frequency interferences.This method is beneficial to realize on-line diagnosis of typical gear faults and provides an effective method for the research of real-time monitoring of mechanical faults and online diagnosis.