Research On Weak Fault Diagnosis Of One-way Valve Of High Pressure Diaphragm Pump

The check valve is the main component of the high pressure diaphragm pump,its working state is normal or not plays a vital role in the safety and stability of the high pressure diaphragm pump,which affects the economic benefits of the whole enterprise.Therefore,the research on the fault diagnosis and the monitoring of the running state of the check valve have important research significance and engineering application prospects.In practical engineering,the working environment of the high-pressure diaphragm pump is affected by the harsh environment,variable working conditions,complicated equipment structure and the inter-correlation between the components of the diaphragm pump,so that the fault signal of the check valve is not only weak but also accompanied by powerful background noise.Therefore,the fault signal exhibits non-stationary,nonlinear,multi-level,coupling and complexity,which causes great interference to the fault feature extraction and fault diagnosis of the check valve.For this issue,the research content is as follows:(1)To address the problem that the weak fault is difficult to diagnosis under the strong noise and non-stationary and nonlinearity of the check valve,a weak fault detection method of check valve based on Variational Mode Decomposition(VMD)and Wigner-Ville is proposed.Firstly,the VMD method with adaptability and anti-noise is applied to process the lower SNR fault signal of check valve to achieve more fault information.The Wigner-Ville time-frequency method with excellent time-frequency resolution is used to character the energy distribution of the fault signal of check valve.The experimental of the check valve shows that the proposed method can effectively analyze the weak fault characteristics and the qualitative diagnosis the weak fault of the check valve.(2)Aiming at the problem that the multi-level,coupling and complexity of the signal under strong noise and the method proposed in the previous chapter can only qualitatively characterize the weak faults of the check valve,the weak fault feature extraction and diagnosis method based on VMD-SVD(Singular Value Decomposition,SVD)is proposed.Firstly,using excellent VMD method to extract the singular value features that characterize the intrinsic properties of the signal constitutes the feature vector to characterize the operating state of the check valve.In order to highlight the advantages of singular value features,using the single singular value boundary method to quantitative analysis the singular value.Finally,(Variable Predictive Based Class Discriminate,VPMCD)method that can make full use of the mutual intrinsic relationship between each feature can better characterize the essential characteristics of the signal.The experimental analysis proves that the method can avoid the drawbacks of the traditional time-frequency analysis method,and can effectively extract the weak fault characteristics and realize the weak fault diagnosis of the check valve.(3)For the previous chapter,the VMD method reduces the useful signal,at the same time,the signal-to-noise ratio of the fault signal is low.The VPMCD method can only select a single regression model,resulting in low generalization ability of the check valve weak fault and the recognition result has the problem of accidental one-sidedness,and a weak fault diagnosis method for check valve based on adaptive stochastic resonance and(Multi-mode Fusion Variable Predictive Based Class Discriminate,MFVPMCD)is proposed.The statistical probability method is employ to optimize the VPMCD method to realize the modeling of complex mapping relationship between feature vectors.Compared with other pattern recognition methods,MFVPMCD method has stronger ability to mine the essential characteristics of signals.The experimental shows that the proposed method can effectively and accurately realize the weak fault diagnosis of the check valve,and can overcome the problem of low generalization ability and accidental one-sidedness of the recognition result.The paper mainly focuses on the problem that the weak fault characteristics of the check valve are difficult to extract and diagnose.A series of fault feature extraction and diagnosis methods are proposed to realize the effective diagnosis of the weak fault of the high pressure diaphragm pump check valve.