Research On The Operation Monitoring System Of Oilfield Water Injection Motor Based On Embedded System

As the key equipment in the oil field,the water injection motor is directly related to whether the oil extraction task can be completed on schedule whether its operating state is stable or not.Once a failure occurs,it needs to be shut down for maintenance,and it may cause a huge safety accident.Therefore,the safety inspection of oilfield water injection motors has always been a very important task.Traditional safety inspections require staff to hold expensive and cumbersome off-line testing instruments to inspect and inspect water injection motor units scattered in different places one by one,which is time-consuming and laborious and requires high levels of staff.Based on the above background,in order to monitor the running state of the water injection motor during work,discover and eliminate potential safety hazards in time,this paper studies an oilfield water injection motor operation monitoring system based on an embedded system.The system takes Raspberry Pi as the core,and is equipped with self-developed and designed visual operation monitoring system software,which can analyze and process the working status signals of the water injection motor collected by the sensor in real time,including the vibration,speed and temperature signals of the water injection motor bearing,so as to automatically judge the current situation.The running status of the water injection motor.The biggest features of this system are: low overall cost,simple operation interface,and embedded intelligent fault diagnosis algorithm model,which can automatically identify the abnormal operation status of the water injection motor,and judge the current fault type,and realize the operation monitoring of the water injection motor.The main work of this paper includes the following three parts:First,based on the understanding of the current status of fault diagnosis research at home and abroad,after a large number of simulation results comparison,it is determined to use 2D-CNN-LSTM as the fault diagnosis model of this research.The one-dimensional time-series signal of the vibration acceleration of the water-injected motor bearing is converted into a two-dimensional grayscale image,and then the convolutional neural network is used for feature extraction,and the long-short-time neural network is input for fault identification and classification;secondly,according to the specific needs and engineering background,through hardware selection Model and software development,based on the Qt/Embedded framework,an oilfield water injection motor operation monitoring system based on an embedded system was designed;finally,the on-site test proved that all functional modules of the system can be used normally,and the overall operation is stable,reaching the expected goal.