| Nowadays,salt chemical companies use long-distance pipelines to transport brine.However,brine is corrosive,and the brine pipeline has the characteristics of long distance,large span.Pipelines also cross roads,woodland,cultivated land,ravines,and villages.Once the brine pipeline leaks,it will cause damage to surrounding farmland and villages.Therefore,it is urgent to develop a reliable and safe pipeline leak detection system to detect the operating conditions of the brine pipeline.This paper proposes a new method for distributed detection of leakage in a brine pipeline: According to the analysis of software system functional requirements and technical requirements,a leakage detection terminal based on STM32F7 is designed.The weak signal collected by the sensor is sent to the STM32F7 after being amplified and adjusted by a differential amplifier circuit,AD digital-to-analog conversion circuit.This system uses a "ping-pong" structure in STM32F7,one for reading and one for writing,which can avoid the problem of read-write conflicts.The vibration signal and pressure signal collected by the brine pipeline are used for simple fault analysis through the intelligent terminal.And transmit the suspected failure data to the cloud computing platform through the 4G wireless transmission module.At the same time,GPS is also used to synchronize the timing,and time stamps are added to the collected data for unified management.And use second pulse signal of GPS to realize the function of synchronously collecting signals of each distributed terminal.For the fault signal transmitted to the host computer,this paper uses the One-dimensional convolutional neural network model to analyze the characteristics of the existing fault data,and fully extracts the characteristics of the fault data of the halogen pipeline.In order to judge the pipeline operation,the One-dimensional convolutional neural network model has trained.And the accuracy of this model id more than 90%. |
PDF Full Text Request