Research On Fault Diagnosis Of Electric Pump Sliding Bearing Based On CEEMD

As an important component of large rotating electromechanical equipment,sliding bear ing has the function of supporting the rotating shaft and transmitting torque.Its operation condition will have a crucial impact on the stable and orderly industrial production.Therefore,in order to make various industrial activities smoothly,it is of great practical significance to carry out fault diagnosis research on sliding bearing.At present,the domestic and foreign scholars in the field of equipment fault diagnosis of rolling bearing has invested a large amount of energy,from the traditional method based on equipment vibration signal diagnosis application research to the use of deep learning intelligent algorithm fault classification diagnosis research have achieved fruitful results,however,in the background of engineering practical application,the large rotating equipment sliding bearing fault diagnosis is carried out,the research on the application of fault diagnosis is rare.In this paper,sliding bearing of high-power electric pump is taken as the research object,and the vibration signals generated in the operation process of field equipment are collected to carry out fault diagnosis research on the sliding bearing of electric pump.First of all,because the noise components in the measured vibration signal will have a great impact on the state feature extraction of the equipment,this paper decomposes the acceleration vibration signal of the sliding bearing of the electric pump with noise by using the Complementary Ensemble Empirical Mode Decomposition(CEEMD)algorithm and solves the correlation coefficients of the IMF components generated by the decomposition,and then calculates the correlation coefficients according to the results According to the principle of component optimization,a series of IMF components are screened and the screened components are used to reconstruct the original vibration signal to achieve the effect of noise suppression.The effectiveness of the method is also proved in the subsequent application.Secondly,because the data collected on site is only a single type of acceleration vibration signal,in order to realize the comprehensive diagnosis of multiple types of vibration signals for equipment operation status and better ensure the accuracy of diagnosis,the acceleration vibration signal after noise reduction is integrated in frequency domain after fast Fourier transform and the corresponding speed time is obtained by performing inverse transform on the integration result The effectiveness of the signal conversion method is verified by the relevant data.Finally,on the basis of the above,the correlation analysis software is used to calculate and analyze the processed data signal and obtain the vibration intensity,spectrum diagram and axis trajectory diagram of the equipment.Through feature comparison,the fault diagnosis of the sliding bearing of the electric pump on site is realized.