Research On Diagnostic Monitoring System Of Water Flooding Pump Condition

With the development of modern industry, the safe operation of equipment, more and more important, and the safety monitoring of equipment have become an increasing attention. The oilfield centrifugal water flooding pump assemblies are the essential oil production equipment, since its operation condition is directly related to the status of the stable and high yield of crude oil. It become very important to monitoring the conditions of the centrifugal water flooding pump assemblies, so that can detect and forecast the potentially failure, analyze the sorts and causes of the failure and service it in good time, prevent worsen of the operation condition, and extend the service life.In this paper, the vibration information of the water flooding pump assembly become the object of study, based on the vibration signal acquisition analysis theory and virtual instruments, designed the diagnosis monitoring system for monitoring the working condition of the centrifugal flooding pump. It was first to get the water vibration information by the piezoelectric acceleration sensor unit, then made the following disposals, made pre-pre-filtering and amplification by signal conditioning, convert the conditioned signal to digital signal which computer can handle by data acquisition card; got the acceleration signal and displacement signal through first and second integrals to the acceleration signals respectively, the use of LabVIEW software system for collecting information on the implementation of the analysis and treatment, based on the water flooding pump assembly common fault summing up the characteristics of a variety of fault information, realized the water injection unit fault diagnosis system design by making use of software implementation, realized the implementation man-machine interface design by the use of LabVIEW, VB, Access database three software interface programming.In this paper, the vector that represent the signal characteristics of different frequency bands was drawn by decomposing the signal in different frequency band and analysis decomposition coefficient in different frequency band respectively.