An Experimental Study Of The Metal Pipeline Defect Detection Based On FSM Technology

With the rapid development of industry,the safety and efficiency of energy transportation have higher requirement.Therefore,pipeline transportation,which has the advantages of safety,occupies an increasingly important position in the field of industrial energy transportation in China.However,with the deteriorating operating environment of pipelines,man-made damage and the increase of pipe operation time,pipeline accidents frequently occur.After the pipeline accident,it not only causes tremendous economic losses,but also pollutes the environment and even seriously affects personal safety.Due to the more complex manufacturing process of elbow pipes and the more stress corrosion during operation,a non-invasive,high-efficiency and high-precision pipe defect monitoring technique is needed.Based on the in-depth analysis of the principle of FSM,the hardware of the FSM system is designed.The hardware mainly includes the FSM power supply,the cable between the FSM power supply and the measured object,the reference board,the electrode,the electrode welding equipment and the data acquisition equipment.Among them,the power supply uses WWL-LSX precision linear DC high-power supply,electrode welding equipment using STP105-type welder,and explore the best welding conditions M4,M6 electrode,feed currents are 20 A,25A,30 A,respectively,data acquisition equipment using Fluke 8808 A high-precision digital multimeter.The dimensional parameters of the defects(circular defects,rectangular defects,trapezoidal defects)were designed by orthogonal experiments,and the machining defects were calculated by using a 3060 VMC CNC milling machine.The four bent tubes were measured,the primary and secondary sequence analysis was performed to determine the depth factor.It is a significant factor affecting the defect;the influence of significant factors on the FC value is analyzed;the effect of the defect on the voltage on the surrounding electrodes is analyzed.Aiming at the problem of unidentifiable small corrosion pits found in the experiment,the voltage method of primary and secondary is proposed,and the feasibility of this method is analyzed theoretically.Combined with BP neural network,a small corrosion pit depth recognition based on BP neural network is esTablelished.Neural network has been trained and verified.The simulation experiment shows that the relative deviation of defect depth is within 4.5%.The created BP neural network model of defect depth recognition has higher accuracy.Therefore,the esTablelished BP neural network Network model has a certain value,and it can better identify the depth of pipeline defects.