Vibration Signal Acquisition,processing And Diagnosis System Based On Subspace Decomposition

Despite the endless development of new energy,oil remains the most important and most used resources.Oilfield water injection operation is one of the important oil recovery process and its purpose is to replenish the energy of oil reservoir.Most injection system consists of a centrifugal pump and electrical components.Only the pump working properly,can oil recovery work safely and efficiently.Therefore,more and more people pay attention to the research of equipment condition monitoring.For the shortage that the real-time processing capability of present pump water monitoring system is not strong,this paper constructed a condition detection system based on subspace decomposition.This paper is based on pumps in oilfield water-injection station.Firstly,it builds an entire set of hardware platform which consists of not only data acquisition module based on digital accelerometer ADXL345 and MCU STM32F030 but also data processing module based on MCU STM32F407.The selection of acquisition and communication chips and layout is completed according to engineering demands.Secondly,this paper introduces some algorithms used in this system and runs their simulation.The preprocessing is made using signal smoothing,filtering and spectral convert to change acceleration signal to displacement signal.Then EPSD is used to process the displacement signal to obtain characteristic frequency of the vibration signal.Because different malfunction leads to different vibration characteristic frequency,the working condition can be judged by the results of the analysis.Finally,field tests prove that the circuit in the system works stably and meets the design requirements.It is able to achieve the function of vibration signal acquisition,processing and diagnosis of working condition.