Research On Fault Diagnosis Of Rolling Bearing Based On Notch Denoising And Empirical Wavelet Transform

With the rapid development of modern industry,the fields such as aerospace,mining,coal,metallurgy,etc.,are affecting the national economy.And its requirements for the reliability of mechanical equipment are getting higher and higher.Once the equipment fails,it may cause irreparable consequences.As a key element in mechanical equipment,the state of the rolling bearing largely determines whether the equipment can operate normally.Therefore,it has great significance for improving the reliability of mechanical equipment and equipment maintenance efficiency that carrying out the fault diagnosis research of rolling bearing.The main research contents and achievements of this dissertation are given as follows:(1)Firstly,from the structure and type of rolling bearing,the research on the failure causes and failure modes of its is made.Then,the vibration signal is chosen to do research for the fault diagnosis.Secondly,two noise reduction methods are proposed to deal with the fault signal of rolling bearing.Finally,the proposed methods are combined with Recurrence Quantification Analysis(RQA)and Support Vector Machine(SVM)to complete the fault identification and classification.(2)In order to solve the problem that the fault signal of rolling bearing has working noise and harmonic interference,which can lead to distortion or submersion of the useful impulse signal,a noise reduction method for rolling bearing fault signal of an adaptive notch filter based on parameter optimization is proposed.The experimental results of simulation and measured signal show that the proposed method is effective to remove noise and suppress harmonic interference of rolling bearing fault signals.(3)The application of Empirical Wavelet Transform(EWT)in fault signal processing of rolling bearing is studied.Affected by working noise,when EWT is applied,the spectrum will be excessively divided.Aiming this problem,a method of signal noise reduction and decomposition is proposed.This method is based on the improved EWT.The experimental results of simulation and measured signal show that the proposed method can solve the problem effectively.And it can suppress the interference of working noise within a contain extent.In addition to this,it can extract the impact feature component accurately and completely.(4)According to the above research results,combining the proposed methods with RQA and SVM,the fault identification and classification are completed.The experimental results show that proposed method combined with RQA performs well and the accuracy of fault diagnosis can be achieved more than 90% by using the proposed two methods alone and in combination.This suggests that the fault type of rolling bearing can be identified more accurately.