| Reciprocating high-pressure diaphragm pump is power source which conveying slurry solid-liquid two-phase flow pipeline with the long distance、high lift,is used in many widely fields of industrial production,and check valve as the high-pressure diaphragm pump core part that exhibiting local fault will lead to unplanned shutdown in the operation,the unplanned shutdown will directly affect the production of enterprises,even cause serious accidents.so establishing the model of fault diagnosis is a research direction with value of applications for the high-pressure diaphragm pump,extracting fault feature of the vibration signal in the happening fault of the high pressure diaphragm pump check valve,distinguishing the fault type and timely correcting to eliminate the trouble.The paper focuses on the monitoring state and fault diagnosis problems to study of the check valve of the high-pressure diaphragm pump,and solves the key problems and technical difficulties of feature extraction,fault status recognition and so on,the details are as follows:(1)A fault diagnosis method based on intrinsic time scale decomposition(ITD)-AR model,and combined with support vector data description(SVDD)was proposed for check valve of the high-pressure diaphragm pump in this paper aiming the problem that failing to get enough fault samples in the actual operation,making full use of the normal running state information to establish SVDD model in normal state.This proposed method extract the vibration signal characteristics by ITD-AR and establish SVDD model in the normal state,and judge the check valve running status by the degree of vibration signal feature vectors deviates from the SVDD model.so to realize that monitoring the state for check valve of the high-pressure diaphragm pump.(2)A fault diagnosis method based on variation mode decomposition(VMD)energy entropy and variable predictive model based class discriminate(VPMCD)was proposed for check valve of the high-pressure diaphragm pump in this paper aiming to solve the problem that accuracy identification of fault types,the method uses the VMDmethod to extract the characteristics of the vibration signal,obtains the energy entropy of the vibration signal,and introduces the VPMCD multi-classification model to the fault diagnosis,and establishes the VPM model which embodies the intrinsic relationship between the eigenvalues.and the eigenvalues of the tested samples are predicted on this basis,and the prediction results are taken as the basis of their fault classification,and the fault types of the check valves of the high-pressure diaphragm pumps are detected.(3)Long-distance monitoring system for check valve of high-pressure diaphragm pump was designed in which mainly includes two parts of the client hardware acquisition and PC software analysis,hardware acquisition is to realize signal acquisition based on MPU6050,STM32 as the main control chip,ENC28J60 as signal transmission module,and PC software analysis interface by Visual Studio development analysis of receiving signal,and signal analysis processing by calling the embedded Matlab algorithm,monitoring the running state,diagnose the fault type.The high pressure diaphragm pump check valve as the paper research object,and solved the key problems and technical difficulties abut extracting signal feature 、recognition fault condition for check valve of the high-pressure diaphragm pump.this study has certain practical application value. |