Research On Early Fault Diagnosis Of High Pressure Diaphragm Pump Check Valve Based On ICEEMD And CMFDE

Pipeline transportation has become the fifth largest mode of transportation in my country.There are dozens of long-distance slurry pipelines that have been built and put into use in China.As the most important equipment for pipeline transportation of slurry,reciprocating high-pressure diaphragm pumps have very high manufacturing costs and high-pressure diaphragm pumps fail.It will lead to safety accidents,and enterprises also face huge economic losses.As the key mechanical component of the high-pressure diaphragm pump,the check valve also works very frequently.Therefore,the operation of the check valve directly affects the performance and efficiency of the high-pressure diaphragm pump.The slurry conveyed by the high-pressure diaphragm pump is in a state of solid-liquid two-phase flow.During the operation of the check valve,it is affected by the size of the solid particles in the slurry,changes in the slurry flow,and the working condition of the pump itself.To cause a check valve failure to occur suddenly or simultaneously,and the collected signal is affected by multiple excitation sources,and the signal has the characteristics of non-stationarity and non-linearity.The working environment of the check valve is noisy,and the collected signal contains a lot of background noise.The early failure of the check valve is weak,the characteristics are not obvious,and it is easily covered by noise,so the early failure of the check valve is difficult to detect and diagnose.Successfully diagnosed the early failure of the check valve under the background of noise,which can prevent the high-pressure diaphragm pump from being unable to work normally due to the wear and breakdown of the check valve and causing economic losses and safety accidents.Aiming at the early fault diagnosis of the check valve under the background of a lot of noise,the research contents are as follows:(1)Aiming at the early failure of the check valve of the reciprocating high-pressure diaphragm pump,the vibration signal contains a large amount of background noise,causing the feature information to be covered by noise,a removing noise method of improved complete ensemble empirical mode decomposition(ICEEMD)and Hausdorff distance(HD)in check valve early fault signal is proposed.Use ICEEMD to decompose the collected signal into multiple Intrinsic Mode Functions(IMF)containing different information,and the noise in the original signal is decomposed into IMF components to varying degrees.Use HD to locate the noise components in the IMF components,calculating the Hausdorff distance of the probability density function of each IMF component and the original signal,and separate noisy IMF components from IMF components decomposed by ICEEMD.The kurtosis as an indicator,was used to select and reconstruct some of the IMF components with larger kurtosis among the remaining IMF components.Hilbert envelope was used to demodulate the reconstructed signal,and conduct comparative tests to analyze the noise reduction effect.The simulation results show that HD can effectively locate and separate the noise IMF component of the IMF component from the ICEEMD decomposition,and the method can effectively extract the feature frequency of the signal under noise background.The experimental results of the measured data show that the proposed method can effectively extract the fundamental frequency and its multiplied frequency of check valve submerged by noise,and has a good noise reduction effect.(2)Aiming at the problem of weak early failure of the check valve of the reciprocating high pressure diaphragm pump,unobvious features and a large amount of noise interference,a method for early weak fault diagnosis of check valve based on composite multiscale fluctuation dispersion entropy(CMFDE)is proposed.Replace the normal cumulative distribution function mapping in the CMFDE method with tangent sigmoid to improve the noise resistance of CMFDE.Calculate the composite multiscale fluctuation dispersion entropy of the acquired signal,construct the feature matrix,and input it into the support vector machine(SVM)classifier to establish the SVM check valve early fault recognition model.The actual engineering data of the early failure of the check valve is input into the model for the early failure diagnosis of the check valve to verify the method,and comparative experiments are performed.The verification experiment and the comparison experiment show that it is not necessary to reduce the noise of the original signal of check valve,which simplifies the diagnosis process.The composite multiscale fluctuation dispersion entropy feature can accurately reflect the different signal characteristics of check valve,which improves the identification rate of the early weak fault diagnosis of check valve,and the fault diagnosis result is less affected by classifiers,with the identification accuracy reaching 96.667%.The thesis focuses on the problems that the collected high-pressure diaphragm pump check valve signal is seriously interfered by noise,the signal characteristics are easily covered by noise,and the early failure of the check valve is weak and the characteristics are not obvious.Corresponding signal noise reduction and early fault diagnosis are proposed.Method and successfully achieved effective problem solving results.