Research On Transfer Diagnosis Method Of Rolling Bearing Of Mining Machine Under Dynamic Assistance

Rolling bearings are the "joints" of the transmission chain of mining electromechanical equipment,which directly affect the reliability of high-value and advanced electromechanical equipment operation in mines and open-pit mines.In the actual work process,due to the sudden changes in the thickness of the stripping layer and time-varying transportation volume,the rolling bearing is long-term serviced in the harsh working conditions of large load,time-varying load,frequent start and stop,and extreme cold and heat environments such as high temperature,high humidity and dust.Which is prone to failure,light shutdown affects continuous production resulting in economic losses,and heavy casualties lead to major safety accidents and adverse social impacts.In the actual work process,its long-term service in harsh conditions of large load,time-varying change,frequent start and stop and extreme cold and hot environment such as high temperature,high humidity and dust.Which is prone to failure,light shutdown affects continuous production resulting in economic losses,and heavy casualties lead to major safety accidents and adverse social impacts.Therefore,it is of great theoretical significance and practical engineering value to carry out the health monitoring and early warning of the rolling bearing of the transmission chain of mine electromechanical equipment.In order to ensure the safe,efficient and stable operation of mine electromechanical equipment,vibration and other physical information are collected for real-time monitoring and early warning of its service status.But the existing fault diagnosis methods based on signal analysis and intelligent algorithm still have some problems in this process.On the one hand,most research on the state identification of mining machine rolling bearings directly focuses on the extraction of vibration signal features,without clarifying the dynamic characteristics of the hidden failure forms and excitation mechanisms of the bearings mentioned above.And the influence of the excitation induced by the fracture of the conductor bar on its state identification is not considered.On the other hand,as the implementation carrier of the "smart mine",the electromechanical equipment is developing toward unmanned and intelligent,and the "massive" multi-physical perception information reflecting the operation status of its transmission chain during the service period is collected.However,the abovementioned "vast data" has the problem of rich health status data,few or even missing fault data samples.In the actual field,it is difficult to obtain "complete" and "universal" rolling bearing fault samples,which makes the state identification method based on label classification low diagnostic accuracy and poor effect.Therefore,it is an urgent need to conduct in-depth research on the vibration characteristics of rolling bearings in mining machine transmission chains and their accurate and fast state identification in the case of small samples of faults.In this regard,research is carried out in four aspects: vibration characteristics analysis of hidden faults of mining rolling bearings,extraction of effective components of early hidden fault information from massive data,transfer of fault sample data,and accurate and efficient status identification,specifically including the following:(1)Research on dynamic characteristics of mining rolling bearings considering time-varying displacement excitation.Starting from the morphological characteristics of hidden faults such as pitting corrosion zones and small peeling pits in mining rolling bearings,the basic characterization model of the impact waveform induced by the above faults is derived,and the excitation mechanism of time-varying displacement is analyzed;Then,based on the above analysis results,constructing a dynamic model of rolling bearing pitting corrosion like spot zone and incipient tiny peeling pit faults based on time-varying displacement excitation using polynomial functions,exploring the vibration characteristics of bearings with pitting corrosion zones and tiny peeling pits considering time-varying displacement excitation(2)Research on the dynamic characteristics of mining rolling bearings considering the excitation of guide bar fracture fault.On the basis of basing on dynamic model of the bearing pitting corrosion zones and incipient tiny spalling pits of rolling bearings with time-varying displacement excitation considering which is constructed above,the excitation mechanism of the "bearing-rotor" system caused by guide bar fracture fault was explored;Considering the above excitation,a fault coupled dynamic model of bearings in different states is constructed to characterize the dynamic response of mining rolling bearings under the excitation of guide bar fracture fault.(3)Research on the diagnosis method of mining machine transmission chain bearing based on multi domain correlation energy fluctuation evaluation coefficient.Taking the dynamic response characteristics of vibration signals generated by different types of faults in mining motor guide bars and rolling bearings as the gripper,this paper explores the fault sensitive indicators in their vibration signals under strong background noise,and integrates multiple indicators to construct multi domain correlation energy fluctuation coefficients.Based on this,a new method of extracting effective components from vibration data of mining machine transmission chains is proposed to improve the quality of sample data required for subsequent intelligent fault diagnosis of mining machine bearings.(4)Research on the transfer diagnosis method of mining machine rolling bearings based on digital analog drive.Considering the real on-site situation where there are few fault samples and some types are missing,combined with the bearing fault dynamics model constructed in preceding,a mining machine transmission chain rolling bearing transfer diagnosis method based on digital analog drive is proposed,which realizes the recognition of the complete state types of the bearings mentioned above in the case of small samples or even missing fault types of actual measured data,And the effectiveness of the transfer diagnosis method for rolling bearings in mining machine transmission chain based on digital analog drive was verified through bench tests and field test.The research results provide basic theoretical and methodological support for the health monitoring of rolling bearings that a key basic component of the transmission chain of mining electromechanical equipment during their service life;The predictive maintenance and condition-based maintenance of high value and high-end electromechanical equipment transmission chains in underground mines and open-pit mines have important theoretical value and practical significance.