Research On Fault Diagnosis Method Of Motor Bearing

Motor is an important driving force in the modern industrial process,and the bearing is one of the most basic components of the motor,it is the most vulnerable components,and the bearing running directly affect the motor is good or bad.Therefore,in order to make the motor can work properly,the bearing status monitoring and diagnosis is very necessary.Based on the analysis of the commonly used motor bearing fault damage identification method,it is found that the vibration signal not only has very strong anti-interference ability,but also can detect the minor fault.So the use of vibration detection technology for motor bearing fault diagnosis has become the mainstream trend.Firstly,the main failure modes and their causes of the motor bearing are analyzed.The derivation process of the characteristic frequency of the motor bearing fault is described in detail,and the characteristics of the typical fault s ignal of the motor bearing are summarized.Second,the noise will interfere with the collection of the signal data,thus affecting the validity and accuracy of the data.The form of the noise and the degree of influence on the acquisition process are close ly related to the type of the fault.The types of the fault are different,The noise generated by them will also be distributed in different frequency bands.The noise of these different frequency bands also generates different degrees of superposition or reduction of the target sound signals collected by the system,resulting in the change of the target signal enthalpy and the distortion of the collected data.Considering the principle and process of the influence of noise on the collected data,this thesis studies a more advanced feature extraction method,which is based on the wave-odd entropy with relatively low correlation.Finally,a clustering algorithm is introduced in the process of fault diagnosis.Using improved clustering algorithm for damage id entification.Experiments show that this method can effectively identify the motor bearing failure,and compared with the traditional K-means method,fault diagnosis speed and accuracy are significantly improved.