Font Size: a A A

Research On Early Fault Diagnosis Of High-pressure Diaphragm Pump Check Valve Based On Chaos And Stochastic Resonance

Posted on:2021-11-05Degree:MasterType:Thesis
Country:ChinaCandidate:Y J ChenFull Text:PDF
GTID:2512306200953239Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
As an important power equipment in the project,the high-pressure diaphragm pump is widely used in industrial production fields such as petroleum mining and slurry transportation.Check valve is one of the core mechanical parts of the high-pressure diaphragm pump.Because it works in an environment with complex structure and changing working conditions,the incidence of failure is much higher than that of other parts of the pump.If the fault location and fault type can be found in the early stage of the fault in time,the major consequences caused by the damage of the check valve can be effectively contained,so it is of great significance to carry out early fault diagnosis research of the high-pressure diaphragm pump check valve.The early vibration signal of the check valve is not only disturbed by strong background noise,but also affected by attenuation in signal transmission,loss of signal acquisition equipment and other factors,resulting in the component of the vibration signal that can reflect the fault characteristics become weak.This brings great troubles and challenges to the early fault diagnosis of the check valve.In response to this problem,the specific research contents of this article are as follows:(1)Aiming at the problem that the early fault characteristic information is weak and difficult to detect when submerged in strong background noise,an early fault detection method for check valves based on Stochastic Resonance(SR)and Variational Mode Decomposition(VMD)of parameter optimization is proposed.First,Duffing Stochastic Resonance is used to enhance the characteristics of fault vibration signals,and Chaos Particle Swarm Optimization(CPSO)is used to adaptively determine the stochastic resonance parameters to achieve the optimal stochastic resonance effect.Then use CPSO again to obtain the optimal VMD parameter combination,decompose the output signal of the SR system with VMD,select the effective IMF component signal for reconstruction by the envelope entropy value,and finally perform the envelope spectrum analysis.Experimental analysis proves the method can effectively detect the early fault feature information.(2)Aiming at the problem that the SR method can effectively detect early failure signals,but cannot estimate the signal amplitude to determine the magnitude of the failure,a method for early failure detection of the check valve based on the parameter optimized VMD and improved chaotic Duffing oscillator is proposed.First,the CPSO is used to optimize the parameters of the VMD,and then the fault vibration signal is VMD decomposed,and the effective component signal is selected to be input into the improved Duffing oscillator,and the weak characteristic signal is detected by combining the change of the phase trajectory and the maximum Lyapunov index.Experimental analysis proves that this method can effectively detect early failure signals and estimate the degree of failure.(3)In view of the fact that the aforementioned method cannot diagnose various types of early failures of the check valve well,a method for early fault diagnosis of the check valve based on parameter-optimized stochastic resonance and Hidden Markov Model(HMM)is proposed.First,CPSO-optimized stochastic resonance processing of the check valve vibration signals under normal conditions,early stuck valve failures and early wear failures,then VMD decomposition of the output signal of the stochastic resonance system to obtain multiple IMF components,extract each component Multiple feature parameters are used to construct a feature vector set,and the feature vector set is subjected to vector quantization.Finally,the feature vector is input to the HMM to implement fault classification.The experimental results show that this method can improve the diagnosis rate of the early fault type of the check valve.The thesis focuses on the high-pressure diaphragm pump check valve as the research object.Through the signal noise reduction method and fault diagnosis method proposed in this paper,the problem of early check valve failure is difficult to be diagnosed.
Keywords/Search Tags:check valve, early fault diagnosis, stochastic resonance, chaos, hidden markov
PDF Full Text Request
Related items