Research On Pipeline Defect Identification Method Based On Magnetic Flux Leakage Data Imaging

Oil and natural gas are one of the main energy sources in people's daily life.They account for more than 50% of the total primary energy consumption in the world.Pipeline transportation is the most important transportation mode of oil and natural gas.Nondestructive testing is a subject to detect the internal or surface damage and material properties of oil and gas pipelines without damaging materials or components.With the progress of modern science and technology,magnetic flux leakage nondestructive testing(mfl-ndt)has many advantages,such as simple principle,easy engineering application and low requirements for the surface of the tested workpiece,so it has become one of the main testing methods for oil and gas pipelines.However,at present,the analysis and identification of pipeline magnetic flux leakage data still adopts the method of manual judgement,which has the disadvantages of low efficiency and high cost.Therefore,a more intelligent pipeline defect identification method needs to be studied.In this paper,a pipeline defect recognition method based on magnetic flux leakage data imaging is proposed: the validity of magnetic flux leakage data is judged,the qualified data is preprocessed by abnormal data elimination and mileage correction,and the magnetic flux leakage data is transformed into gray value and imaged.According to the characteristics of gray image,image optimization methods such as wavelet de-noising,interpolation processing and channel base value calibration are used,It can effectively improve the imaging effect and lay the foundation for subsequent image processing.The edge extraction method based on improved Canny operator and the line detection method based on improved Hough transform are used,which can accurately identify the line of pipeline girth weld,and determine the location of the mileage point of pipeline girth weld according to the line parameters.On this basis,the data of girth weld position is intercepted separately to generate a two-dimensional array.The mean value of each column of data is calculated to generate a one-dimensional array.The cross-correlation between each channel's data and the one-dimensional array is calculated and normalized to obtain the cross-correlation coefficient.The cross-correlation coefficient is used as the basis for judging whether the channel's data is abnormal data,So as to judge whether there are defects on the pipe girth weld,and calculate the channel number of abnormal data.The threshold segmentation method is used to transform the image into binary image,the morphological algorithm is used to separate the defect and extract the edge of the defect,and the contour detection algorithm is used to detect the contour of the defect and calculate the center of gravity coordinates of the contour,so as to determine the location of the mileage point of the pipe defect.In this paper,a 1219 pipeline magnetic flux leakage data is taken as the experimental data.Based on the magnetic flux leakage imaging of pipeline data,the cross-correlation algorithm and image processing algorithm based on fast Fourier transform are used to realize the identification of abnormal girth weld and corrosion defects of pipeline.The results show that this method can effectively identify the mileage points of all 29 girth welds in 1219 pipeline,and the deviation is within 1%;It can be judged that 2 of 29 girth welds are abnormal girth welds,and the channel number of abnormal data can be calculated,which are consistent with the actual data;The results show that the location of corrosion defects with large area can be identified,which is consistent with the actual data.Compared with the manual judgement method,this method is more efficient and intelligent.