Quantitative Calculation And Characteristic Analysis Of Internal Detection Signal Of Pipeline Defect Magnetic Flux Leakage

Long-distance oil and gas pipelines are widely used to transport oil,gas and other energy needed for people’s production and life,and it is of great economic and social importance to ensure the safe operation of long-distance pipelines.However,if a pipeline leaks or bursts,it can cause huge economic losses and endanger people’s lives.With the advantages of strong anti-interference capability,fast signal acquisition and no coupling agent required,pipeline leak detection technology has become one of the most effective means of pipeline safety maintenance recognised in the international non-destructive testing field.It is important to establish a mathematical model of the defect detection signal and to analyse the component signals to accurately assess the defects for leak detection.However,the existing theoretical analysis model of leakage magnetism is only a single analysis for defects such as corrosion,cracks and round holes,usually in-service pipeline work will be subject to internal pressure,there will be stress concentrations around the defect,the stress present at the tip of the defect is one of the factors affecting the leakage magnetism signal of the defect.At the same time the traditional magnetic charge model assumes that the magnetic charge is uniformly distributed on the defect sidewalls,whereas in the actual inspection process the pipe is magnetised and the magnetic lines of force are affected by the shape of the defect and leak in the air at different angles and with different degrees of thinness,resulting in the magnetic charge at the defect not being distributed on the defect sidewalls in a uniform situation.Therefore the traditional uniform magnetic charge model leads to large errors in the calculation of the leakage signal and the quantitative analysis of the inspection results.This paper establishes an improved three-dimensional magnetic charge mathematical model based on J-A theory,and analyses the influence of the magnetomechanical effect of pipeline defects on the leakage signal under the action of internal pressure;introduces the charge density parameter to establish a non-uniform magnetic charge mathematical model for internal detection of pipeline leakage,and calculates the non-uniform distribution of magnetic charge on the defect sidewalls;analyses the influence of different defect sizes and lift-off values on the leakage signal;studies the optimal segmentation accuracy and its influencing factors in the calculation of the non-uniform magnetic charge model;and designs systematic experiments to verify the correctness of the theoretical model.The results of the study show that: the radial and axial components of the leakage signal show a non-linear decrease with increasing internal pressure;the magnetic charge distribution curve is symmetrically shaped at the centre of the defect,with a trough at the centre and a peak at the defect prism;the magnetic charge density value increases sharply near the defect surface and prism as the defect depth and width increase;different defect sizes and the value of lifting off affect the defect leakage signal.The signal eigenvolume in the contour plot can visually identify the signal characteristics of the defect,responding to the location and size;the first-order derivative of the leakage signal eigenvalue decreases with cell size and then rises,with the location of the minimal value of the derivative curve being the best segmentation accuracy.