Research On Automatic Fault Monitoring Of Oil Pumping Station Based On Noise Analysis

The oil pump station unit is composed of an electric motor and a centrifugal pump,which is the core equipment of the oil production process.In order to avoid a production safety accident caused by the equipment failure,establishing oil transfer pump unit working state real time monitoring and fault diagnosis system has very important significance.Oil transfer pump unit operation in a state of vibration and noise,and the specific operation noise contains rich device status information,can be used as an objective evaluation of equipment running status and the basis of fault diagnosis.Currently applied acoustics diagnosis for equipment fault diagnosis technology has become a new research hotspot,although people and centrifugal pump for motor operation mechanism of noise generation and typical failure causes and basic clear variation mechanism of the noise,but how to accurately identify all kinds of typical faults in the whole running noise still needs further research.This topic is proposed based on noise analysis of oil pump station automatic monitoring technology research for fault,looking forward to technically improve the oil transfer pump unit operation state evaluation of reliability and precision of fault diagnosis.This topic fusion technology,virtual instrument technology,acoustic detection and analysis to the composition of oil transfer pump unit structure and noise characteristic parameters as the basis,from the oil pump station to collect the unit operation of noise samples for analysis,through the combination of theory and test system research.This paper focuses on the variation of noise spectrum induced by cavitation failure,so as to determine the detection method of cavitation failure and the initial discriminant algorithm of cavitation.Specific experimental study on the process of oil transfer pump fault on-line real time monitoring in the process of precision discriminant cavitation state in the early years of the fault,needs to be based on specific turbine cavitation noise characteristics to determine the cavitation noise than early failure analysis of samples.In this paper,the synthetic method of initial noise sampling noise was proposed based on characteristic frequency analysis.This topic on the basis of oil transfer pump cavitation failure variation mechanism of the noise in the oil gathering and transferring production from FS100 field-65-200 unit running noise samples capture the cavitation noise spectrum,early failure according to extract multiple single frequency signal amplitude decreasing way,as a benchmark based on LabVIEW synthesis realization of cavitation noise early failure sample.Three steps are realized: the characteristic frequency selection and calibration,the corresponding amplitude setting,the initial characteristic noise multi-frequency information samples are synthesized.Experiment proves that the synthesis for 1 Hz frequency given value adjustment of step length,a given frequency and reference frequency deviation 0.32 Hz,the biggest synthetic signal and reference signal amplitude corresponding maximum relative error of 1.90%,according to different types of oil transfer pump unit,this method possesses repeatability and practical operability.