Multi-channel magnetic flux leakage signal anti-interference imaging recognition method for defects of in-service pipeline

Abstract

The application discloses a multi-channel magnetic flux leakage signal anti-interference imaging recognition method for in-service pipeline defects, relates to the pipeline nondestructive testing technical field, and comprises the following steps: collecting magnetic flux leakage signals of multiple circumferential magnetic flux leakage detection channels and establishing a multi-channel magnetic flux leakage signal matrix, extracting cross-channel peak-valley responses, forming a common mode change sequence and a defect contrast sequence according to the peak value and the valley value, and determining a reverse collapse channel interval by utilizing the reverse change relationship of the two; performing midline correction and defect contrast recovery on signals in the interval according to the magnetic flux leakage responses on both sides of the interval, obtaining an anti-elevation multi-channel magnetic flux leakage signal matrix, and generating a pipeline wall magnetic flux leakage image and determining an in-service pipeline defect area according to the anti-elevation multi-channel magnetic flux leakage signal matrix. The application can reduce weakening of dynamic elevation interference on circumferential expansion defect magnetic flux leakage responses, and improve defect imaging continuity and recognition reliability.

Description

Technical Field

[0001] This invention relates to the field of pipeline non-destructive testing technology, and in particular to a multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines. Background Technology

[0002] As the service life of long-distance pipelines for oil and natural gas increases, the pipe walls are prone to corrosion and localized metal loss due to media corrosion, stress, and external environmental influences. Magnetic flux leakage (MFL) detection, by magnetizing ferromagnetic pipelines and using multiple MFL detection channels arranged circumferentially along the pipeline to collect MFL signals generated at the defect locations, can detect pipe wall defects without damaging the pipeline structure. Existing technologies typically arrange MFL signals collected by different MFL detection channels according to the axial and circumferential positions of the pipeline, and obtain pipe wall defect distribution information through signal filtering, anomaly enhancement, and MFL imaging to further determine the location of the defects and their circumferential coverage.

[0003] The actual inspection environment inside in-service pipelines is quite complex. Weld protrusions, deposits, and attachments on the pipeline inner wall may cause dynamic lift-offs of varying degrees but spatial continuity between adjacent magnetic flux leakage (MFL) detection channels and the pipe wall. When the dynamic lift-off area spatially overlaps with circumferentially expanding corrosion or large-area metal loss areas, some MFL detection channels may experience significant overall background shifts. At the same time, the local MFL peak-valley responses corresponding to the actual defects are attenuated, resulting in a state where multiple adjacent channels share anomalies but local defect characteristics are weakened. Since actual circumferentially expanding defects also have the characteristic of continuous distribution across multiple adjacent MFL detection channels, existing methods for anti-interference processing based on the common changes of multiple channels, abnormal amplitudes, or spatial continuity are difficult to effectively distinguish between associated dynamic lift-off interference and actual expanding defects. This can easily lead to the actual defect response being mistakenly weakened or the dynamic lift-off interference being retained, thereby affecting the integrity of MFL imaging of defects in in-service pipelines and the reliability of defect identification. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies in the detection of magnetic flux leakage signals for in-service pipelines, where both associated dynamic lift-off interference and real circumferential expansion defects can form continuous anomalies across adjacent channels. This makes it difficult to accurately distinguish between dynamic lift-off and real defects when they overlap spatially, resulting in the magnetic flux leakage response of real defects being mistakenly weakened or the lift-off interference being mistakenly retained. Therefore, this invention proposes a multi-channel magnetic flux leakage signal anti-interference imaging and identification method for defects in in-service pipelines.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution: A multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines includes: The leakage magnetic signals of multiple leakage magnetic detection channels arranged along the circumference of the pipeline are collected, and a multi-channel leakage magnetic signal matrix is ​​established according to the axial position of the pipeline and the circumferential position of the leakage magnetic detection channels. Peak and valley responses are extracted from the multi-channel magnetic flux leakage signal matrix. Peak and valley responses that overlap in axial range and are continuously adjacent in circumferential direction are associated to form cross-channel response groups, and the peak and valley values ​​of each magnetic flux leakage detection channel are determined. Based on the peak and valley values, the peak-valley midline quantity and local defect comparison quantity of each leakage magnetic field detection channel are determined, and a common mode variation sequence and a defect comparison sequence are formed respectively. The reverse collapse channel interval is determined based on the inverse change corresponding to the increase in the peak-valley midline amount in the common mode variation sequence and the decrease in the local defect comparison amount in the defect comparison sequence. Based on the peak-valley midline quantity and local defect comparison quantity of the magnetic flux leakage detection channels on both sides of the reverse collapse channel interval, the reference midline quantity and reference defect comparison quantity of each magnetic flux leakage detection channel in the interval are determined, and the magnetic flux leakage signal in the interval is corrected to obtain the anti-lift-off multi-channel magnetic flux leakage signal matrix. Based on the anti-lift-off multi-channel magnetic flux leakage signal matrix, a pipe wall magnetic flux leakage image is generated, and the defect area of ​​the in-service pipeline is determined.

[0006] Preferably, establishing the multi-channel leakage magnetic signal matrix includes: Acquire the magnetic flux leakage signal sampling data of each magnetic flux leakage detection channel and the axial position of the detection device corresponding to each magnetic flux leakage signal sampling data; Based on the axial position of the detection device, the leakage magnetic signal sampling data of different leakage magnetic detection channels are mapped to a unified pipeline axial position coordinate. The multi-channel magnetic flux leakage signal matrix is ​​obtained by mapping the magnetic flux leakage signal sampling data according to the circumferential arrangement of the magnetic flux leakage detection channels and the axial position coordinates of the pipeline.

[0007] Preferably, the step of extracting peak and valley responses and forming cross-channel response groups includes: Determine the difference between adjacent leakage magnetic signal sampling data in each leakage magnetic detection channel along the axial direction of the pipeline; The location where the difference changes from greater than zero to less than zero is determined as the local maximum location, and the location where the difference changes from less than zero to greater than zero is determined as the local minimum location. Peak-valley response is formed based on the axially adjacent local maximum and local minimum locations. The peak and valley responses of different magnetic flux leakage detection channels that overlap in axial range and are continuously adjacent in circumferential direction are correlated to form the cross-channel response group; The local maxima and local minima in each peak and valley response are determined as the peak and valley values ​​of the corresponding magnetic flux leakage detection channel, respectively.

[0008] Preferably, the peak-valley midline value is the arithmetic mean of the peak and valley values ​​corresponding to the same magnetic flux leakage detection channel, and the local defect comparison value is the absolute difference between the peak and valley values ​​corresponding to the same magnetic flux leakage detection channel.

[0009] Preferably, the formation of the common mode variation sequence and the defect comparison sequence includes: Arrange the midline values ​​of each peak and valley according to the circumferential arrangement order of the leakage magnetic flux detection channels within the cross-channel response group to obtain the common mode variation sequence; The local defect comparison quantities are arranged in the same circumferential arrangement order of the magnetic flux leakage detection channels to obtain the defect comparison sequence.

[0010] Preferably, before determining the reverse collapse channel interval, the process includes: Determine the magnetic flux leakage detection channel corresponding to the maximum peak-valley midline amount in the common mode variation sequence and the magnetic flux leakage detection channel corresponding to the minimum local defect comparison amount in the defect comparison sequence. Determine the Spearman rank correlation coefficient between the common mode variation sequence and the defect comparison sequence; When the maximum peak-valley midline quantity and the minimum local defect comparison quantity correspond to the same magnetic flux leakage detection channel, and the Spearman rank correlation coefficient is less than zero, the magnetic flux leakage detection channel is determined as the reverse change positioning reference.

[0011] Preferably, when the maximum peak-valley midline value in the common-mode variation sequence or the minimum local defect comparison value in the defect comparison sequence corresponds to multiple magnetic flux leakage detection channels, it includes: The first set of magnetic flux leakage detection channels corresponding to the maximum peak-valley midline amount and the second set of magnetic flux leakage detection channels corresponding to the minimum local defect comparison amount are respectively formed; Determine the Spearman rank correlation coefficient between the common mode variation sequence and the defect comparison sequence; When the Spearman rank correlation coefficient is less than zero and the first magnetic flux leakage detection channel set and the second magnetic flux leakage detection channel set have a common magnetic flux leakage detection channel, the continuous channel segment where the common magnetic flux leakage detection channel is located is determined as the reverse change positioning region.

[0012] Preferably, determining the reverse collapse channel interval includes: The nearest local low point in the common mode change sequence is retrieved from both sides of the pipe circumference using the reverse change positioning reference, or the nearest local low point in the common mode change sequence is retrieved from both ends of the reverse change positioning area outward along the pipe circumference. The magnetic flux leakage detection channel whose peak-valley midline value is not greater than the corresponding peak-valley midline value of the adjacent magnetic flux leakage detection channels on both sides of the circumference is determined as the local low point. The range of continuous channels between the leakage magnetic field detection channels corresponding to the nearest local low points on both sides is defined as the reverse collapse channel interval.

[0013] Preferably, the leakage magnetic signal within the correction interval includes: Using the peak-valley midline value of the leakage magnetic flux detection channel at both sides of the reverse collapse channel interval as the boundary value, linear interpolation is performed according to the circumferential arrangement position of the leakage magnetic flux detection channel to obtain the reference midline value of each leakage magnetic flux detection channel inside the interval. Using the local defect comparison amount of the magnetic flux leakage detection channels on both sides as the boundary value, linear interpolation is performed according to the circumferential arrangement position of the magnetic flux leakage detection channels to obtain the reference defect comparison amount of each magnetic flux leakage detection channel within the interval. Determine the ratio of the reference defect comparison value to the corresponding local defect comparison value for each magnetic flux leakage detection channel. When the ratio is greater than one, the ratio is determined as the defect comparison recovery coefficient. When the ratio is not greater than one, one is determined as the defect comparison recovery coefficient. The difference between the original magnetic flux leakage signal and the corresponding peak-valley midline quantity of each magnetic flux leakage detection channel within the interval is recovered by amplitude recovery according to the defect comparison recovery coefficient, and the difference after amplitude recovery is combined with the corresponding reference midline quantity to obtain the corrected magnetic flux leakage signal.

[0014] Preferably, obtaining the anti-lift-off multi-channel leakage magnetic signal matrix and determining the defect region of the in-service pipeline includes: Keeping the leakage magnetic signal outside the reverse collapse channel interval unchanged, the corrected leakage magnetic signal inside the reverse collapse channel interval and the leakage magnetic signal outside the reverse collapse channel interval are recombined according to the circumferential arrangement position of the leakage magnetic detection channel to obtain the anti-lift-off multi-channel leakage magnetic signal matrix. The anti-lift-off multi-channel magnetic flux leakage signal matrix is ​​mapped to the pipe wall magnetic flux leakage image according to the axial position of the pipe and the circumferential arrangement of the magnetic flux leakage detection channels. The recovered peak-valley response is extracted from the anti-lift-off multi-channel magnetic flux leakage signal matrix. The connected mapping region of the recovered peak-valley response with overlapping axial ranges and corresponding magnetic flux leakage detection channels that are continuously adjacent in the circumferential direction in the pipe wall magnetic flux leakage image is determined as the defect region of the in-service pipeline. Based on the axial position of the pipeline corresponding to the defect area in the in-service pipeline and the covered magnetic flux leakage detection channel, the axial position and circumferential coverage of the defect are determined.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention correlates the peak and valley responses in multiple circumferential magnetic flux leakage detection channels and constructs a common-mode variation sequence characterizing the overall magnetic flux leakage signal shift and a defect comparison sequence characterizing the degree of local defect fluctuations. By utilizing the inverse variation relationship between the two in adjacent magnetic flux leakage detection channels, the reverse collapse channel interval is determined. This enables the identification of special working conditions where dynamic lift-off caused by weld protrusions, deposits, and attachments spatially overlaps with actual circumferential expansion defects. This avoids misjudging actual expansion defects as common-mode interference based solely on common anomalies or spatial continuity of multiple channels, and improves the ability to distinguish between correlated dynamic lift-off interference and actual pipe wall defects.

[0016] 2. After determining the reverse collapse channel interval, this invention uses the actual magnetic flux leakage response on both sides of the interval as a local continuous reference. It corrects the peak-valley centerline shift and local defect contrast attenuation within the interval, while keeping the magnetic flux leakage signal that does not meet the reverse collapse condition unchanged. This suppresses the influence of dynamic lift-off while preserving the cross-channel continuous magnetic flux leakage of the real circumferentially expanding defect. Furthermore, by combining the upper limit of the background peaks and valleys obtained from the defect-free calibration to form an effective defect area, it can reduce false identification caused by ordinary background fluctuations and improve the continuity, integrity, and reliability of identifying defect location and circumferential coverage in in-service pipeline magnetic flux leakage imaging. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the multi-channel magnetic flux leakage signal anti-interference imaging and identification method for defects in in-service pipelines provided by the present invention. Figure 2 A schematic diagram of local defect comparison and reverse collapse provided by the present invention; Figure 3 A schematic diagram of the magnetic flux leakage image of the pipe wall before correction when the dynamic lift-off overlaps with the actual defect, as provided by the present invention. Figure 4 This is a schematic diagram of the magnetic flux leakage image of the pipe wall and the determination of the defect area after local relative continuous recovery provided by the present invention. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] This invention is applicable to magnetic flux leakage detection of ferromagnetic pipelines in service, focusing on detection scenarios where weld protrusions, deposits, attachments, or radial movement of the magnetic flux leakage detection device cause multiple adjacent magnetic flux leakage detection channels to be dynamically separated to varying degrees but spatially continuous, and where the dynamically separated area spatially overlaps with the actual pipe wall defect area. In this embodiment, dynamic lifting refers to the temporary increase in the distance between the magnetic sensing position of the magnetic flux leakage detection channel and the corresponding pipe wall relative to the normal operating distance during the movement of the magnetic flux leakage detection device along the in-service pipeline. This is caused by the influence of weld protrusions, deposits, attachments on the inner wall of the pipeline, or the radial movement of the detection unit. The dynamic lifting describes the change in the relative distance between the magnetic flux leakage detection channel and the pipe wall during the detection process, and does not limit the specific mechanical structure that causes the change in the relative distance. For actual circumferential corrosion or large-area metal loss, continuous anomalies originate from the spatial extension of the defect itself in the circumferential direction of the pipe; for dynamic lift-off caused by weld protrusions or deposits, continuous anomalies originate from the continuous change in the distance between adjacent magnetic flux leakage detection channels and the pipe wall; therefore, it is not possible to reliably distinguish between the actual defect response and dynamic lift-off interference simply by whether multiple adjacent magnetic flux leakage detection channels show anomalies at the same time. Furthermore, when the dynamic lift-off area and the actual defect area spatially overlap, the magnetic flux leakage detection channel, which is strongly affected by the dynamic lift-off, may produce a significant overall magnetic flux leakage signal shift. At the same time, due to the increased distance between the magnetic sensing position and the actual pipe wall defect, the local peak-valley response corresponding to the actual defect is attenuated. This creates a special spatial state between consecutive adjacent magnetic flux leakage detection channels where the overall level of the peak-valley response is enhanced while the local peak-valley amplitude difference is weakened. Within the same cross-channel response group, as the relative degree of dynamic lift-off influence on consecutive adjacent magnetic flux leakage detection channels increases, the overall magnetic flux leakage signal level of its peak-valley response tends to increase, while the local amplitude difference between the peak and valley values ​​in the same peak-valley response tends to decrease. This creates a spatial state between consecutive circumferential channels of the pipeline where the overall background change and the local defect fluctuation change are opposite in direction.

[0020] Example 1, as Figure 1 As shown, this embodiment provides a multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines, which specifically includes the following steps; S1. Acquisition of multi-channel leakage magnetic signal and spatial unification The in-service pipeline to be inspected is magnetized, and multiple magnetic flux leakage detection channels arranged along the circumference of the pipeline are used to continuously collect magnetic flux leakage signals as the magnetic flux leakage detection device moves along the axial direction of the pipeline. In the same testing task, each leakage magnetic field detection channel uses leakage magnetic field signals with the same detection direction, and maintains consistency in magnetization direction, magnetic sensing element installation direction, and signal polarity definition. Before formally performing in-service pipeline testing, the polarity of the leakage magnetic field signals used by the current leakage magnetic field detection device is calibrated. Using a calibration pipe section corresponding to the material, outer diameter, and wall thickness of the pipeline to be tested, and under the condition that the magnetization conditions and the working state of the leakage magnetic field detection channel are consistent, leakage magnetic field signals are obtained at the normal working distance of the magnetic sensing position and after increasing the lift-off distance, and the direction of peak-valley midline quantity change corresponding to the two states is determined. When the peak-to-valley midline value increases after increasing the lift-off distance, the polarity of the current magnetic flux leakage signal remains unchanged; when the peak-to-valley midline value decreases after increasing the lift-off distance, the corresponding magnetic flux leakage signals in the current detection task are uniformly reversed in polarity, so that the peak-to-valley midline value increases according to the unified polarity definition after increasing the lift-off distance; the polarity reversal only changes the positive and negative directions of the magnetic flux leakage signal value, and does not change the absolute amplitude difference between different sampling positions, the axial position of the pipe where the local extreme value is located, or the spatial correspondence between each magnetic flux leakage detection channel; Further, a standard defect calibration pipe section with a known metal loss area is used to collect the leakage magnetic field signal of the same known metal loss area under two conditions: normal working distance and increased lift-off distance. When the polarity is unified, the peak-valley midline value increases when the lift-off distance is increased, and the peak-valley amplitude difference of the known metal loss area decreases. This confirms that the current leakage magnetic field signal meets the lift-off response direction on which the subsequent reverse collapse identification of this invention is based. This invention does not require any leakage magnetic signal component to naturally exhibit the same lift-off response direction on any leakage magnetic detection device. Instead, it performs subsequent reverse collapse identification only after the current detection device and the current leakage magnetic signal have been confirmed to have the corresponding response direction through the above calibration. During the magnetic flux leakage detection process of in-service pipelines, the axial position of the magnetic flux leakage detection device corresponding to each magnetic flux leakage signal sampling data is recorded simultaneously; based on the axial position of the magnetic flux leakage detection device, the magnetic flux leakage signal sampling data of different magnetic flux leakage detection channels are mapped to a unified pipeline axial position coordinate. Let the unified axial position coordinates of the pipeline be as follows: in, Indicates the first The axial position coordinates of the pipe The serial number indicating the axial position coordinate of the pipeline. This indicates the total number of axial position coordinates of the pipe included in the current detection data; Let the first One magnetic flux leakage detection channel is located at the axial position of the pipeline. The corresponding leakage magnetic signal at this location is: in, This only indicates the leakage magnetic field detection channel number. Only indicates the first One magnetic flux leakage detection channel is located at the axial position of the pipeline. The original leakage magnetic signal at the location; When the original sampling positions of different magnetic flux leakage detection channels correspond to the same axial position of the pipe, they are directly arranged according to the corresponding axial position of the pipe. When the original axial sampling positions of different magnetic flux leakage detection channels are not completely consistent, adjacent axial sampling positions are interpolated according to the axial position of the magnetic flux leakage detection device to make each magnetic flux leakage detection channel obtain a unified axial position coordinate of the pipe. The axial position interpolation only completes the spatial alignment between different magnetic flux leakage detection channels and does not change the technical principle of subsequent cross-channel reverse change recognition.

[0021] After unifying the axial positions, the magnetic flux leakage signals are arranged using the circumferential position of the pipe in the magnetic flux leakage detection channel as one dimension and the axial position of the pipe as another dimension, forming a multi-channel magnetic flux leakage signal matrix; any magnetic flux leakage signal in the multi-channel magnetic flux leakage signal matrix can be identified by the magnetic flux leakage detection channel number. and the axial position of the pipe This unique determination allows for subsequent circumferential comparison of the magnetic flux leakage responses of different magnetic flux leakage detection channels at the same axial position in the pipeline.

[0022] S2. Effective peak-valley response extraction and cross-channel response group establishment; Based on the multi-channel magnetic flux leakage signal matrix, the magnetic flux leakage signals of each magnetic flux leakage detection channel are analyzed along the axial direction of the pipeline. For the One magnetic flux leakage detection channel was obtained in sequence according to the axial position of the pipeline: Calculate the difference in magnetic flux leakage signal between the axial positions of adjacent pipes: in, Only indicates the first Each magnetic flux leakage detection channel is located at the axial position of the adjacent pipe. and The difference in leakage magnetic signal between them; When the difference in magnetic flux leakage signal before a certain axial position of a pipe is greater than zero, and the difference in magnetic flux leakage signal after that position is less than zero, the corresponding position is determined as a local maximum position; when the difference in magnetic flux leakage signal before a certain axial position of a pipe is less than zero, and the difference in magnetic flux leakage signal after that position is greater than zero, the corresponding position is determined as a local minimum position. When the same leakage magnetic field signal value appears at consecutive sampling positions, making the leakage magnetic field signal difference equal to zero, the zero difference position is not directly used as the local extremum. Instead, the nearest non-zero leakage magnetic field signal difference in the consecutive identical value segment is searched along both sides of the pipe axis. When the nearest non-zero difference before the consecutive identical value segment is greater than zero and the nearest non-zero difference after the segment is less than zero, the consecutive identical value segment is determined as a local maximum region, and the pipe axis position corresponding to the middle of the consecutive identical value segment is used as the local maximum position. When the nearest non-zero difference before the consecutive identical value segment is less than zero and the nearest non-zero difference after the segment is greater than zero, the consecutive identical value segment is determined as a local minimum region, and the pipe axis position corresponding to the middle of the consecutive identical value segment is used as the local minimum position. Along the axial direction of the pipeline, a peak-valley response is formed by the occurrence of an adjacent local maximum and a local minimum, as well as the leakage magnetic field signal between them. In this embodiment, the peak-valley response refers to the local signal variation segment defined by an axially adjacent local maximum, a local minimum, and the leakage magnetic field signal between them in the leakage magnetic field signal continuously collected along the pipeline axis by the same leakage magnetic field detection channel. The peak-valley response does not require the local maximum to appear before the local minimum, and its axial range is determined by the interval between the two pipeline axial positions corresponding to the local maximum and the local minimum. For the The peak-valley response of a single magnetic flux leakage detection channel determines the magnetic flux leakage signal value corresponding to the local maximum as the peak value. The leakage magnetic signal value corresponding to the local minimum is determined as the valley value. To avoid directly using minor background fluctuations on defect-free pipe walls for subsequent cross-channel analysis, before formal testing, background benchmarks were collected on defect-free calibration pipe sections corresponding to the material, outer diameter, and wall thickness of the pipe to be tested using the same magnetic flux leakage detection device. Under the same magnetization and signal acquisition conditions as in-service testing, the first... The peak and valley values ​​of all background peak-valley responses of each leakage magnetic flux detection channel are determined, and the absolute difference between the peak and valley values ​​of each background peak-valley response is calculated; the first... The maximum value among all absolute differences of background peak-valley responses obtained by a single magnetic flux leakage detection channel in a defect-free calibrated pipe section is determined as the upper limit of the background peak-valley for that magnetic flux leakage detection channel. In this implementation, the upper limit of the background peak and valley refers to the peak value of the first peak and valley under the same magnetic flux leakage detection device, a defect-free calibration pipe section of the corresponding specification, and the same magnetization and signal acquisition conditions as in-service testing. The maximum value of the absolute difference between the peak and valley values ​​of the background peak-valley response appearing in the first leakage magnetic flux detection channel is used to characterize the first... The upper limit of background peaks and valleys that may be generated by a single magnetic flux leakage detection channel under defect-free conditions; For the first in-service pipeline inspection data One magnetic flux leakage detection channel, when the current peak-valley response satisfies: When the peak-valley response is valid, it is determined as a valid peak-valley response. When the current peak-valley response does not satisfy the above relationship, it is treated as background fluctuation and does not participate in the establishment of subsequent cross-channel response groups. In this implementation, a valid peak-valley response refers to a peak-valley response in the in-service pipeline leakage magnetic flux signal where the absolute difference between the peak value and the valley value is greater than the upper limit of the background peak-valley of the corresponding leakage magnetic flux detection channel. The upper limit of the background peak-valley comes from the actual background data under the same detection device, corresponding specification defect-free pipe section, and the same magnetization and signal acquisition conditions. After obtaining effective peak and valley responses from each leakage magnetic flux detection channel, the effective peak and valley responses between different leakage magnetic flux detection channels are correlated. According to the arrangement order of the leakage magnetic flux detection channels in the circumferential direction of the pipeline, when the axial range of the effective peak and valley responses in two adjacent circumferential leakage magnetic flux detection channels has a common axial position in the pipeline, it is determined that the two effective peak and valley responses have an overlapping relationship in the axial position. When two or more continuously adjacent leakage magnetic flux detection channels along the circumference of the pipeline have effective peak-valley responses that overlap in their axial range, these effective peak-valley responses are associated to form a cross-channel response group. In this embodiment, a cross-channel response group refers to a set of effective peak-valley responses composed of two or more continuously adjacent leakage magnetic flux detection channels along the circumference of the pipeline, and the effective peak-valley responses have an overlapping relationship in the axial position of the pipeline. A cross-channel response group is used to characterize the corresponding leakage magnetic flux response of the same local pipe wall axial region in multiple circumferentially adjacent leakage magnetic flux detection channels. Effective peak-valley responses that exist only in a single leakage magnetic flux detection channel are not included in the reverse collapse analysis in this embodiment.

[0023] S3, common mode variation sequence and defect comparison sequence establishment; For the current cross-channel response group, the first One magnetic flux leakage detection channel, based on the corresponding peak value Valley value Determine the midline of peak and valley measurements : And determine the local defect comparison quantity : in, Only indicates the first The peak-valley midline value of the current peak-valley response of each leakage magnetic flux detection channel; Only indicates the first Comparison of local defects in the current peak-valley response of each magnetic flux leakage detection channel; In this embodiment, the peak-valley midline quantity refers to the signal level characterization quantity obtained by the arithmetic mean of the peak and valley values ​​of the same peak-valley response. It is used to describe the offset state of the overall peak-valley response relative to the current leakage magnetic signal amplitude coordinate. It does not represent the actual pipe wall defect depth, nor does it represent the absolute lift-off distance between the leakage magnetic detection channel and the pipe wall. In this embodiment, the local defect comparison quantity refers to the absolute amplitude difference between the peak and valley values ​​of the same peak-valley response. It is used to describe the degree of local leakage magnetic fluctuation within the peak-valley response. It does not directly represent the actual defect size or actual defect depth. The dynamic lift-off impact is judged by the relative spatial variation of peak-valley centerline quantity and local defect comparison quantity within the same cross-channel response group, without determining the dynamic lift-off distance or actual defect size solely based on the absolute value of any peak-valley centerline quantity or local defect comparison quantity. To illustrate the variation characteristics of the peak-valley midline quantity and the local defect contrast quantity, let's assume that the peak and valley values ​​of a certain peak-valley response are subject to the same background offset. The effect is that the changed peak and trough values ​​are expressed as follows: The corresponding midline volume at this point is: The corresponding local defect comparison quantity is: Therefore, when the peak and trough values ​​change in the same direction as the whole, the midline value of the peak and trough can reflect the overall change, while the amplitude difference between the peak and trough values ​​will not change due to the completely identical overall offset. On the other hand, when the lift-off distance is increased and the leakage magnetic response of the local defect is attenuated, the amplitude change of the peak and valley values ​​corresponding to the real defect relative to the local centerline is reduced, thereby reducing the contrast of the local defect. The above relationship is used to explain the physical meaning of the peak-valley centerline quantity and the local defect comparison quantity used in this invention, and does not require that the local defect comparison quantity in the actual in-service pipeline decreases strictly according to a fixed ratio with the lift distance.

[0024] This embodiment uses the aforementioned polarity calibration and lift-off response calibration results as applicable conditions for subsequent analysis. Subsequent reverse collapse identification is only performed when the flux leakage signal used by the current flux leakage detection device exhibits an increase in peak-valley midline quantity under a unified polarity direction after increasing the lift-off distance, and the local defect contrast quantity in the known defect area decreases. In this invention, the peak-valley midline quantity is used to characterize the relative degree to which each flux leakage detection channel within the same cross-channel response group is affected by common-mode offset; the local defect contrast quantity is used to characterize the relative degree to which local flux leakage fluctuations within the same cross-channel response group are attenuated by dynamic lift-off. Arrange the peak and valley midline quantities sequentially according to the circumferential arrangement of the leakage magnetic flux detection channels in the current cross-channel response group. This forms a common-mode variation sequence; the comparison values ​​of each local defect are arranged sequentially according to the identical circumferential arrangement order of the leakage magnetic flux detection channels. A defect comparison sequence is formed. In this embodiment, the common mode variation sequence refers to the ordered data sequence formed by arranging the peak and valley midline quantities of each leakage magnetic flux detection channel in the circumferential arrangement order of the pipeline within the same cross-channel response group. In this embodiment, the defect comparison sequence refers to the ordered data sequence formed by arranging the local defect comparison quantities of each leakage magnetic flux detection channel in the circumferential arrangement order of the pipeline with the same common mode variation sequence as the common mode variation sequence. The same sequence position in the common mode variation sequence and the defect comparison sequence uniquely corresponds to the same leakage magnetic flux detection channel.

[0025] S4. Determination of the reverse collapse channel section; Determine the maximum peak-valley midline quantity and the corresponding magnetic flux leakage detection channel in the common mode variation sequence, and determine the minimum local defect comparison quantity and the corresponding magnetic flux leakage detection channel in the defect comparison sequence; When the maximum peak-valley midline quantity and the minimum local defect comparison quantity are unique and correspond to the same magnetic flux leakage detection channel, the Spearman rank correlation coefficient between the common mode variation sequence and the defect comparison sequence is further determined. The first Peak-valley midline quantity of each magnetic flux leakage detection channel The order value in the common-mode variation sequence is denoted as Comparison of local defects in the same magnetic flux leakage detection channel The order value in the defect comparison sequence is denoted as ; Let the number of leakage magnetic field detection channels included in the current cross-channel response group be . Then Spearman rank correlation coefficient Determined according to the following formula: in, This represents the arithmetic mean of the peak-valley midline quantity order values ​​within the current cross-channel response group; This represents the arithmetic mean of the order values ​​of all local defect comparison quantities within the current cross-channel response group; This only indicates the number of magnetic flux leakage detection channels currently included in the cross-channel response group; When there are identical values ​​in a sequence, the common order value is the arithmetic mean of all the order positions occupied by the identical values ​​when they are arranged in order of numerical size. Before calculating the Spearman rank correlation coefficient, it is checked whether the order values ​​corresponding to the common mode variation sequence and the defect comparison sequence have non-zero dispersion; when all peaks and valleys in the common mode variation sequence have the same midline value, all The same, or all local defect comparison values ​​in the defect comparison sequence are the same, so that all At the same time, the denominator of the Spearman rank correlation coefficient is zero; in this case, the current cross-channel response group is determined to not meet the negative rank correlation judgment condition, the reverse collapse channel interval is not established, and the original leakage magnetic signal corresponding to the current cross-channel response group remains unchanged. The Spearman rank correlation coefficient is calculated only when the order values ​​of both sequences have non-zero discreteness. when: When the common-mode variation sequence and the defect comparison sequence are generally negatively rank correlated; when the maximum peak-valley midline quantity and the minimum local defect comparison quantity correspond to the same magnetic flux leakage detection channel, and When this happens, the leakage magnetic flux detection channel is determined as the reverse change positioning benchmark. In this embodiment, the reverse change positioning benchmark refers to a single leakage magnetic flux detection channel that, in the absence of parallel extreme values, simultaneously corresponds to the maximum peak-valley midline quantity in the common mode change sequence and the minimum local defect comparison quantity in the defect comparison sequence, and whose cross-channel response group satisfies a negative rank correlation. The reverse change positioning benchmark is used to determine the starting position of the boundary of the reverse collapse channel interval to both sides of the pipeline circumferential direction.

[0026] Starting from the reverse change positioning reference, the nearest local low point in the common mode change sequence is retrieved along the two opposite directions of the pipeline circumference; For the first one located within the current cross-channel response group Each magnetic flux leakage detection channel, when its peak-valley midline quantity When the value is not greater than the midline value of the peak and valley of the adjacent leakage magnetic flux detection channels on both sides of the pipeline circumferential direction, the leakage magnetic flux detection channel is determined as the local low point of the common mode variation sequence. When searching sequentially along the circumferential direction of a pipe, the flux leakage detection channel that first satisfies the above local low point condition is the nearest local low point in that direction; the same search is performed in the opposite direction to obtain the nearest local low point on the other side; the flux leakage detection channels corresponding to the two nearest local low points are taken as the two side boundaries, and the range of continuous flux leakage detection channels between the two boundary flux leakage detection channels is determined as the reverse collapse channel interval. like Figure 2As shown, within the same cross-channel response group, the peak-valley midline quantity and local defect contrast quantity of each leakage magnetic flux detection channel are arranged according to the circumferential position of the leakage magnetic flux detection channel as the lateral spatial order. When the dynamic lift-off area and the actual pipe wall defect area spatially overlap, as the relative degree of the dynamic lift-off influence on adjacent leakage magnetic flux detection channels increases, the leakage magnetic flux signal level of the overall peak-valley response gradually increases, causing the peak-valley midline quantity to rise from both sides to the middle, forming a corresponding high value position in the middle. Due to the increased distance between the magnetic sensing position and the pipe wall, the local peak-valley fluctuations corresponding to the actual defect are weakened, causing the local defect contrast quantity to decrease from both sides to the middle, forming a low value at the circumferential position corresponding to the high value position of the peak-valley midline quantity. An inverse relationship is formed between adjacent leakage magnetic flux detection channels, where the peak-valley midline quantity increases while the local defect contrast quantity decreases. Figure 2 In this embodiment, the position corresponding to the maximum peak-valley midline value and the minimum local defect comparison value is represented as the reverse change positioning reference. The nearest local low point of the peak-valley midline value is retrieved along both sides of the pipeline circumferentially from the reverse change positioning reference. The continuous magnetic flux leakage detection channel range between the left and right local low points is defined as the reverse collapse channel interval. The reverse collapse channel interval is used to limit the circumferential processing range for subsequent peak-valley midline correction and local defect comparison recovery, thereby enabling the overall common-mode offset caused by dynamic lifting and the attenuation of the actual defect local response to be processed separately. In this embodiment, the reverse collapse channel interval refers to the continuous magnetic flux leakage detection channel range between the nearest common-mode change local low points determined by the reverse change positioning reference or the reverse change positioning area described later, within the cross-channel response group that satisfies the conditions for cross-channel common-mode rise and local defect comparison reverse change. The reverse collapse channel interval limits the circumferential range of the pipeline for which this invention performs common-mode offset correction and local defect comparison recovery, and its two boundary magnetic flux leakage detection channels provide the measured boundary data required for subsequent local relative continuous recovery. When it is impossible to obtain the nearest local low point on both sides of the pipeline circumferential direction of the reverse change positioning reference at the same time, it is determined that the current cross-channel response group does not have the condition for continuous recovery of the double boundary, and the subsequent reverse collapse recovery is not performed, and the original leakage magnetic signal corresponding to the current cross-channel response group remains unchanged.

[0027] like Figure 3As shown, before performing local relative continuous recovery of the reverse collapse channel interval, the original multi-channel magnetic flux leakage signal matrix is ​​mapped to a two-dimensional coordinate space according to the axial position of the pipe and the circumferential position of the magnetic flux leakage detection channel, which can form a pre-correction magnetic flux leakage image of the pipe wall. In the position where the real pipe wall defect is not significantly affected by dynamic lift-off, the local magnetic flux leakage anomaly corresponding to the defect maintains a relatively obvious peak-valley change. When the dynamic lift-off affected area and the real pipe wall defect area overlap spatially, in the magnetic flux leakage detection channel that is strongly affected by dynamic lift-off, the overall horizontal shift of the magnetic flux leakage response occurs, and at the same time, the local peak-valley response corresponding to the real defect is attenuated, so that the defect response at the corresponding position in the pre-correction magnetic flux leakage image of the pipe wall is weakened relative to its circumferentially adjacent position, thus forming a weakened defect response zone in the middle.

[0028] S5. Local relative continuous recovery of the reverse collapse channel interval; Let the left boundary of the reverse collapse channel interval be the magnetic flux leakage detection channel. The magnetic flux leakage detection channel is the first one, and the magnetic flux leakage detection channel on the right boundary is the second one. One magnetic flux leakage detection channel; in, and This only indicates the number of the two boundary magnetic flux leakage detection channels; Let the first The peak-valley midline quantity of each magnetic flux leakage detection channel is , No. The peak-valley midline quantity of each magnetic flux leakage detection channel is ;No. The local defect comparison quantity of each magnetic flux leakage detection channel is , No. The local defect comparison quantity of each magnetic flux leakage detection channel is ; Based on the actual installation position of the magnetic sensing position of each leakage magnetic field detection channel in the circumferential direction of the pipeline, determine the corresponding circumferential position coordinates of each leakage magnetic field detection channel; In this embodiment, the arc length coordinates calculated from a predetermined circumferential reference position along the circumference of the pipe are used as the circumferential position coordinates; let the arc length coordinates be... The circumferential position coordinates of each magnetic flux leakage detection channel are: , No. The circumferential position coordinates of each magnetic flux leakage detection channel are: The first section inside the reverse collapse channel The circumferential position coordinates of each magnetic flux leakage detection channel are: ; Linear interpolation is performed on the peak-valley midline values ​​of the two boundary magnetic flux leakage detection channels based on the actual circumferential position coordinates to obtain the first... Reference centerline of each magnetic flux leakage detection channel : Linear interpolation is performed on the local defect comparison values ​​of the two boundary magnetic flux leakage detection channels based on the same actual circumferential position coordinates to obtain the first... Reference defect comparison quantity for each magnetic flux leakage detection channel : in, Only indicates the first Reference centerline value of each magnetic flux leakage detection channel; Only indicates the first The reference defect comparison value of each magnetic flux leakage detection channel; when the reverse collapse channel interval crosses the circumferential reference position, the circumferential position coordinates after passing the circumferential reference position are increased by a pipe circumference, so that... , and The linear interpolation is performed by increasing the values ​​of the reverse collapse channel intervals in the actual arrangement order on the same continuous circumferential coordinate axis. The reference centerline value in this embodiment refers to the local continuous reference value obtained by interpolating the peak-valley centerline value actually measured by the leakage magnetic flux detection channels at both sides of the reverse collapse channel interval and the actual circumferential position of the leakage magnetic flux detection channels inside the interval. The reference defect comparison value in this embodiment refers to the local continuous reference value obtained by interpolating the local defect comparison value actually measured by the leakage magnetic flux detection channels at both sides of the reverse collapse channel interval and the actual circumferential position of the leakage magnetic flux detection channels inside the interval. Both the reference centerline value and the reference defect comparison value belong to the relative recovery benchmark inside the current reverse collapse channel interval and do not represent the theoretical leakage magnetic flux signal centerline value or theoretical defect amplitude under absolute zero lift-off conditions. This embodiment uses two nearest local low points as local continuous recovery boundaries, which does not mean that the actual lift-off distance at the two boundary leakage magnetic flux detection channels is zero; the two boundary leakage magnetic flux detection channels only represent two local termination positions that can be determined from the actual detection data during the weakening process of the current common mode lifting region from the inside to the outside; therefore, the present invention establishes a relatively continuous reference for the current local region through the actual leakage magnetic flux response of the two boundary leakage magnetic flux detection channels, rather than reconstructing the theoretical leakage magnetic field under absolutely no lift-off conditions; For the section inside the reverse collapse channel, the first One magnetic flux leakage detection channel, based on the reference defect comparison quantity. Comparison with actual local defects Determine the coefficient of recovery of defects ; when: hour: when: hour: therefore: in, Only indicates the first Defect comparison recovery coefficient of each magnetic flux leakage detection channel; Since the peak-valley responses entering the cross-channel response group are valid peak-valley responses, satisfying the condition that the peak value and valley value are not the same, therefore: Thus, the above ratio can be determined; When the reference defect comparison value is greater than the actual local defect comparison value, the local magnetic flux leakage response subjected to dynamic lift-off compression is recovered using a defect comparison recovery coefficient greater than one; when the reference defect comparison value is not greater than the actual local defect comparison value, the defect comparison recovery coefficient is kept at one, and the existing actual local magnetic flux leakage response in the current magnetic flux leakage detection channel is not actively weakened. For the The original magnetic flux leakage signal within the axial range corresponding to the current cross-channel response group of each magnetic flux leakage detection channel. Obtain the corrected leakage magnetic signal : in, Only indicates the first One magnetic flux leakage detection channel is located at the axial position of the pipeline. Corrected magnetic leakage signal at the location; The correction first utilizes a reference centerline quantity. Adjust the overall midline level of the current peak-valley response, and then use the defect comparison recovery coefficient. The original leakage magnetic signal relative to the original peak-valley midline quantity To restore local deviations; Instead of deleting the common anomalies in multiple adjacent magnetic flux leakage detection channels as a whole, or uniformly amplifying all magnetic flux leakage signals in the reverse collapse channel interval, the overall centerline shift and local defect contrast attenuation are processed separately in the local area where the relationship between common mode rise and local defect contrast reverse collapse has been identified. The local relative continuous recovery in this implementation refers to using only the actual leakage magnetic response on both sides of the current reverse collapse channel interval as boundary constraints to continuously correct the leakage magnetic response between the two boundaries affected by dynamic lift-off by comparing the centerline level and local defects, so that the leakage magnetic response inside the interval forms a continuous circumferential transition of the pipeline with the measured leakage magnetic response on both sides; the local relative continuous recovery does not mean reconstructing the theoretical leakage magnetic field under absolutely no lift-off conditions. For magnetic flux leakage detection channels outside the reverse collapse channel interval, the original magnetic flux leakage signals remain unchanged; for pipe axial positions within the reverse collapse channel interval but not within the current cross-channel response group axial range, the original magnetic flux leakage signals also remain unchanged; the corrected magnetic flux leakage signals within the reverse collapse channel interval are recombined with the remaining uncorrected original magnetic flux leakage signals according to the original pipe axial position and the original magnetic flux leakage detection channel circumferential position to form an anti-lift-off multi-channel magnetic flux leakage signal matrix; the anti-lift-off multi-channel magnetic flux leakage signal matrix in this embodiment refers to the complete multi-channel magnetic flux leakage data matrix formed by keeping the pipe axial position dimension and the magnetic flux leakage detection channel circumferential position dimension of the original multi-channel magnetic flux leakage signal matrix unchanged, replacing the original magnetic flux leakage signals in the corresponding axial range within the already determined reverse collapse channel interval with the corrected magnetic flux leakage signals, and keeping the original magnetic flux leakage signals at other positions unchanged.

[0029] S6. Pipe wall magnetic flux leakage imaging and determination of defect areas in in-service pipelines; Based on the axial position of the pipeline and the actual circumferential position of the magnetic flux leakage detection channel, the anti-lift-off multi-channel magnetic flux leakage signal matrix is ​​mapped to the two-dimensional coordinate space of pipeline axial-pipeline circumferential direction to generate a magnetic flux leakage image of the pipe wall. In order to ensure that the determination of the final defect area still relies on the physical signal of magnetic flux leakage detection, rather than introducing an image classification model that is not related to the main line of the present invention, the peak and valley response is re-extracted and recovered from the anti-lift-off multi-channel magnetic flux leakage signal matrix in the same way as S2 for determining local extrema; For the The recovered peak-valley response of a single magnetic flux leakage detection channel is used to determine the recovered local maxima as the recovered peak value. The local minimum value after recovery is defined as the valley value after recovery. And calculate the local defect contrast after restoration. : in, Only indicates the first The peak value after recovery of each magnetic flux leakage detection channel. Only indicates the first The valley value after recovery of each magnetic flux leakage detection channel. This only represents the comparison of local defects after the recovery of the corresponding peak-valley response; Call the upper limit of the background peaks and valleys obtained in S2 using the same magnetic flux leakage detection device, a defect-free calibration pipe section of the corresponding specification, and the same magnetization and acquisition conditions. ; when: When the current peak-valley response after recovery is determined, it is defined as the effective peak-valley response after recovery; when Not greater than When the current recovered peak-valley response is determined as background fluctuation, it is not used to form defect areas in in-service pipelines; the recovered effective peak-valley response in this embodiment refers to the peak-valley response that is re-extracted from the anti-lift-off multi-channel magnetic flux leakage signal matrix and the absolute difference between the recovered peak value and the recovered valley value is greater than the upper limit of the background peak-valley of the corresponding magnetic flux leakage detection channel; the recovered effective peak-valley response is used to exclude local signal changes that are still within the defect-free background fluctuation range after anti-lift-off correction; For continuously adjacent magnetic flux leakage detection channels in the circumferential direction of the pipeline, when the corresponding recovered effective peak and valley responses have a common axial position in the axial direction of the pipeline, these recovered effective peak and valley responses are correlated across channels, and their mapping areas in the magnetic flux leakage image of the pipe wall are spatially connected. Only the continuous mapping region consisting of the recovered effective peak-valley responses of two or more circumferentially adjacent magnetic flux leakage detection channels, and where the recovered effective peak-valley responses overlap in the axial direction of the pipeline, is defined as the in-service pipeline defect region. In this embodiment, the in-service pipeline defect region refers to the continuous magnetic flux leakage anomaly mapping region formed in the two-dimensional space of pipeline axial-circumferential direction, consisting of the recovered effective peak-valley responses of two or more circumferentially adjacent magnetic flux leakage detection channels, and where the axial ranges of the recovered effective peak-valley responses overlap.

[0030] When finally determining the defect area of ​​an in-service pipeline, two conditions must be met simultaneously: First, the local defect contrast quantity of the restored peak-valley response that makes up the region is greater than the upper limit of the background peak-valley of the corresponding magnetic flux leakage detection channel; Second, multiple restored effective peak-valley responses have a spatial overlap relationship in the pipeline axial direction between adjacent magnetic flux leakage detection channels. Even if small fluctuations in the defect-free background form local maxima and local minima in a mathematical sense, they will not be determined as defect areas of the in-service pipeline as long as the local defect contrast quantity after restoration is not greater than the upper limit of the corresponding background peak-valley. For a identified defect area in an in-service pipeline, the minimum and maximum axial positions of the pipeline corresponding to the effective peak-valley responses after full recovery of the region are obtained, and the axial position range of the defect is determined by the minimum and maximum axial positions of the pipeline. Furthermore, all magnetic flux leakage detection channels covered by the defect area of ​​the in-service pipeline are obtained, and the circumferential coverage range of the defect is determined according to the actual circumferential positions corresponding to the magnetic flux leakage detection channels. Finally, the magnetic flux leakage image of the pipe wall, the defect area of ​​the in-service pipeline, the axial position range of the defect, and the circumferential coverage range of the defect are output.

[0031] like Figure 4 As shown, after the local relative continuous recovery of the reverse collapse channel section is completed, the anti-lift multi-channel magnetic flux leakage signal matrix is ​​mapped to the two-dimensional coordinate space of pipe axis-pipe circumferential according to the pipe axial position and the circumferential position of the magnetic flux leakage detection channel to form a corrected pipe wall magnetic flux leakage image. Figure 4The grayscale values ​​are used to indicate the relative spatial distribution of the magnetic flux leakage response corresponding to different axial positions and different magnetic flux leakage detection channels, without limiting the absolute value of the specific magnetic flux leakage signal. Compared with the magnetic flux leakage image of the pipe wall before correction, the locally weakened local defect response in the dynamically extracted area is restored locally and continuously, forming a continuous spatial expression with the defect response in the adjacent magnetic flux leakage detection channel. For the restored effective peak-valley response where the local defect contrast is greater than the upper limit of the corresponding background peak-valley and there is an axial overlap between the adjacent magnetic flux leakage detection channels in the circumferential direction of the pipe, the continuous mapping area of ​​the restored effective peak-valley response in the magnetic flux leakage image of the pipe wall is determined as the in-service pipeline defect area. The axial position range of the defect is determined according to the axial position of the pipe corresponding to the in-service pipeline defect area, and the circumferential coverage range of the defect is determined according to the actual circumferential position of the covered magnetic flux leakage detection channel.

[0032] Example 2: Based on Example 1, this example provides a set of recalculated sample data to illustrate the determination process of cross-channel common mode lifting-local defect contrast reverse collapse and the local relative continuous recovery process. Within a certain axial region of a pipeline, the first to the ninth magnetic flux leakage detection channels all extract effective peak-valley responses that meet the corresponding background peak-valley upper limit requirements, and the axial ranges of the effective peak-valley responses overlap with each other. Therefore, the first to the ninth magnetic flux leakage detection channels constitute the same cross-channel response group. The peak and valley values ​​for each magnetic flux leakage detection channel are as follows: The peak value for the first channel is 130, and the valley value is 90; the peak value for the second channel is 125.5, and the valley value is 76.5; the peak value for the third channel is 127.5, and the valley value is 82.5; the peak value for the fourth channel is 130.5, and the valley value is 91.5; the peak value for the fifth channel is 133.5, and the valley value is 100.5; the peak value for the sixth channel is 131, and the valley value is 93; the peak value for the seventh channel is 128, and the valley value is 84; the peak value for the eighth channel is 126, and the valley value is 78; the peak value for the ninth channel is 129.5, and the valley value is 88.5. The peak-valley midline values ​​for each magnetic flux leakage detection channel, calculated based on the peak and valley values, are as follows: The corresponding local defect comparison quantities are as follows: Therefore, the common mode change sequence is: The defect comparison sequence is as follows: The maximum peak-to-valley midline value in the common-mode variation sequence is 117, corresponding to the 5th magnetic flux leakage detection channel; the minimum local defect comparison value in the defect comparison sequence is 33, also corresponding to the 5th magnetic flux leakage detection channel. Since the two extreme values ​​mentioned above are unique, we continue to calculate the Spearman rank correlation coefficient between the two sequences; Arranged from smallest to largest, the order values ​​corresponding to the midline values ​​of each peak and valley are as follows: 6 for the 1st leakage magnetic field detection channel; 1 for the 2nd leakage magnetic field detection channel; 3 for the 3rd leakage magnetic field detection channel; 7 for the 4th leakage magnetic field detection channel; 9 for the 5th leakage magnetic field detection channel; 8 for the 6th leakage magnetic field detection channel; 4 for the 7th leakage magnetic field detection channel; 2 for the 8th leakage magnetic field detection channel; and 5 for the 9th leakage magnetic field detection channel. The order values ​​corresponding to the comparison values ​​of each local defect are as follows: 4 for the first magnetic flux leakage detection channel; 9 for the second magnetic flux leakage detection channel; 7 for the third magnetic flux leakage detection channel; 3 for the fourth magnetic flux leakage detection channel; 1 for the fifth magnetic flux leakage detection channel; 2 for the sixth magnetic flux leakage detection channel; 6 for the seventh magnetic flux leakage detection channel; 8 for the eighth magnetic flux leakage detection channel; and 5 for the ninth magnetic flux leakage detection channel. In this embodiment, the two sequences do not have any parallel values, so the Spearman rank correlation simplification in the case of no parallel values ​​can also be used for verification. The sum of the squares of the differences between the two order values ​​for each magnetic flux leakage detection channel is 240. The current cross-channel response group contains 9 magnetic flux leakage detection channels, therefore: Therefore, the common mode variation sequence and the defect comparison sequence show a completely negative rank correlation; Since the maximum peak-valley midline quantity in the common-mode variation sequence and the minimum local defect contrast quantity in the defect contrast sequence both correspond to the 5th leakage magnetic flux detection channel, at the same time... Therefore, the fifth magnetic flux leakage detection channel was determined as the reverse change positioning reference; The common mode variation sequence is sequentially retrieved along a circumferential direction of the pipe from the fifth leakage magnetic flux detection channel; the peak-valley midline value of the second leakage magnetic flux detection channel is 101, and the peak-valley midline values ​​of the first and third leakage magnetic flux detection channels adjacent to it on both sides of the pipe circumference are 110 and 105, respectively. Therefore, the second leakage magnetic flux detection channel satisfies the local low point condition. When the magnetic flux leakage detection channel is searched along the opposite circumferential direction of the pipeline from the 5th leakage magnetic flux detection channel, the peak-valley midline value of the 8th leakage magnetic flux detection channel is 102. The peak-valley midline values ​​of the adjacent 7th and 9th leakage magnetic flux detection channels are 106 and 109, respectively. Therefore, the 8th leakage magnetic flux detection channel satisfies the local low point condition. Therefore, the second magnetic flux leakage detection channel is defined as one boundary of the reverse collapse channel interval, and the eighth magnetic flux leakage detection channel is defined as another boundary. The reverse collapse channel interval is from the second magnetic flux leakage detection channel to the eighth magnetic flux leakage detection channel. Let the actual circumferential position coordinates of the 2nd, 5th, and 8th magnetic flux leakage detection channels be 20, 50, and 80, respectively, then: The peak-valley midline value of the second magnetic flux leakage detection channel is: The peak-valley midline value of the 8th magnetic flux leakage detection channel is: The local defect comparison value of the second magnetic flux leakage detection channel is: The local defect comparison value of the 8th magnetic flux leakage detection channel is: Therefore, the reference centerline value for the 5th magnetic flux leakage detection channel is: The reference defect comparison value for the 5th magnetic flux leakage detection channel is: The actual local defect comparison value of the 5th magnetic flux leakage detection channel is 33, therefore: Assume the fifth leakage magnetic flux detection channel is located at a certain axial position in the pipe within the current cross-channel response group axial range. If the original leakage magnetic field signal is 130, then: The corresponding corrected leakage magnetic field signal is: get: Through the above calculations, the overall centerline in the original leakage magnetic signal was adjusted from 117, which was affected by the common-mode rise, to 101.5, which was determined by the two actual boundary channels. At the same time, the local changes of the original leakage magnetic signal relative to the original peak-valley centerline were compared and restored.

[0033] The 3rd, 4th, 6th and 7th magnetic flux leakage detection channels are processed in the same way; the 2nd and 8th magnetic flux leakage detection channels serve as the two boundaries of the reverse collapse channel interval, and their actual peak-valley midline quantity and local defect comparison quantity directly constitute the interpolation boundary, without needing to be calculated backward from the internal magnetic flux leakage detection channels; the 1st and 9th magnetic flux leakage detection channels are located outside the reverse collapse channel interval, and the corresponding original magnetic flux leakage signals remain unchanged; For a region where a real circumferentially extended defect overlaps with a non-uniform dynamic lift-off, it is possible to reduce the common-mode rise in the middle channel and restore the local peak-valley changes weakened by dynamic lift-off without completely deleting the abnormal response of the second to eighth magnetic flux leakage detection channels, thereby maintaining the magnetic flux leakage expression of the actual circumferentially extended defect across multiple consecutive magnetic flux leakage detection channels.

[0034] Example 3 illustrates the handling methods when there are parallel values ​​for the maximum peak-valley midline quantity or the minimum local defect comparison quantity, when the reverse collapse channel section crosses the circumferential reference position of the pipeline, and when multiple recovery areas overlap in two dimensions. When the maximum peak-valley midline quantity in the common-mode variation sequence corresponds to multiple leakage magnetic flux detection channels, all leakage magnetic flux detection channels corresponding to the maximum peak-valley midline quantity are combined into a first leakage magnetic flux detection channel set; when the minimum local defect comparison quantity in the defect comparison sequence corresponds to multiple leakage magnetic flux detection channels, all leakage magnetic flux detection channels corresponding to the minimum local defect comparison quantity are combined into a second leakage magnetic flux detection channel set; the Spearman rank correlation coefficient between the common-mode variation sequence and the defect comparison sequence is calculated in the manner described in Example 1. When the Spearman rank correlation coefficient is less than zero, and there is a common magnetic flux leakage detection channel in the first set of magnetic flux leakage detection channels and the second set of magnetic flux leakage detection channels, the magnetic flux leakage detection channels that are continuously adjacent in the circumferential direction of the pipe in the common magnetic flux leakage detection channel are formed into a reverse change positioning region. In this implementation, the reverse change positioning region refers to a continuous channel segment consisting of magnetic flux leakage detection channels that coexist in the set of magnetic flux leakage detection channels corresponding to the maximum peak-valley midline quantity and the set of magnetic flux leakage detection channels corresponding to the minimum local defect comparison quantity when there are parallel values; the reverse change positioning region is used to replace a single reverse change positioning reference to determine the starting position of the retrieval of the two side boundaries. For example, when the fourth and fifth magnetic flux leakage detection channels have the same maximum peak-valley midline quantity, and the fourth and fifth magnetic flux leakage detection channels also have the same minimum local defect comparison quantity, the common channels of the first set of magnetic flux leakage detection channels and the second set of magnetic flux leakage detection channels are the fourth and fifth magnetic flux leakage detection channels. Since the fourth and fifth magnetic flux leakage detection channels are continuously adjacent in the circumferential direction of the pipe, the continuous channel segment formed by the fourth and fifth magnetic flux leakage detection channels is defined as the reverse change positioning area. The nearest local low point is retrieved from one end of the reverse change positioning area along the circumferential direction of the pipe away from the area, and the nearest local low point is retrieved from the other end along the opposite direction. The reverse collapse channel interval is determined by the leakage magnetic field detection channel corresponding to the two nearest local low points. When the first set of magnetic flux leakage detection channels and the second set of magnetic flux leakage detection channels do not have a common magnetic flux leakage detection channel, even if the Spearman rank correlation coefficient between the two sequences is less than zero, a reverse change localization region is not established, reverse collapse recovery is not performed, and the corresponding original magnetic flux leakage signal remains unchanged. When the reverse collapse channel section crosses the reference starting position of the pipe circumferential position coordinates, the pipe circumferential direction is still a continuous closed direction according to the actual physical relationship. At this time, the circumferential position coordinates of the leakage magnetic field detection channel after passing the circumferential reference position are increased by a pipe circumference, so that all circumferential position coordinates of the current reverse collapse channel section are located on the same continuous coordinate axis. Then, the reference centerline quantity and reference defect comparison quantity are interpolated according to the continuously unfolded circumferential position coordinates. For example, if a reverse collapse channel section sequentially covers the magnetic flux leakage detection channel near the end of the pipe circumference and the magnetic flux leakage detection channel near the zero point along the actual pipe circumference direction, then the original circumferential position coordinates of the magnetic flux leakage detection channel after the zero point will be increased by the pipe circumference, so that it continues to be located after the end of the pipe circumference in numerical terms, thereby maintaining the continuity of the actual circumferential position. When there are multiple independent cross-channel response groups in the same section of in-service pipeline, reverse collapse identification is performed on each cross-channel response group separately; when the pipeline axial range of two reverse collapse recovery areas does not overlap, recovery is performed independently; when the pipeline axial range of two reverse collapse recovery areas overlaps, but the corresponding reverse collapse channel intervals do not overlap in the pipeline circumferential direction, recovery is also performed separately. When two or more reverse collapse recovery regions overlap in both the axial and circumferential directions of the pipeline, resulting in the same magnetic flux leakage detection channel at the same axial position of the pipeline... Original leakage magnetic signal at the location When a signal belongs to multiple candidate correction regions, only one final correction is performed on the original leakage magnetic signal. Calculate the corresponding candidate corrected magnetic flux leakage signal based on the boundary data of each candidate inverse collapse recovery region, and determine the difference between each candidate corrected magnetic flux leakage signal and the original magnetic flux leakage signal. The absolute difference between the original and the candidate corrected magnetic flux leakage signal is taken as the final corrected magnetic flux leakage signal at that location. The above processing ensures that the overlapping position adopts the effective correction result with the least change to the original detection result, avoiding over-recovery caused by the same leakage magnetic signal being repeatedly corrected by multiple recovery regions; When the absolute difference between multiple candidate corrected leakage magnetic flux signals and the original leakage magnetic flux signal is the same, the candidate correction result corresponding to the reverse collapse channel interval with a smaller circumferential coverage of the pipeline is selected; when the circumferential coverage of the pipeline is still the same, the candidate correction result corresponding to the cross-channel response group with a smaller axial range of the pipeline is selected. A unique final correction result can be obtained for any leakage magnetic signal that belongs to multiple candidate recovery regions at the same time. When searching from the reverse change positioning reference or the reverse change positioning area to any circumferential direction of the pipeline, if the corresponding nearest local low point cannot be obtained within the current cross-channel response group, then the double-boundary reverse collapse channel interval is not established, and the original leakage magnetic signal of the current cross-channel response group remains unchanged.

[0035] Example 4 further illustrates the calibration and verification process of the reverse collapse relationship and the handling method when the reverse collapse state does not occur; Before formally conducting in-service pipeline inspection, the reverse collapse response direction of the current magnetic flux leakage detection device is verified by using a standard defect calibration pipe section corresponding to the material, outer diameter and wall thickness of the pipeline to be inspected. The standard defect calibration pipe section is equipped with a metal loss area with a known location; First, the magnetic flux leakage detection channel is placed at the normal working distance specified by the detection device. Under fixed magnetization conditions and fixed signal acquisition conditions, it passes through the known metal loss area and acquires the corresponding magnetic flux leakage signal. Then, the peak value, valley value, peak-valley midline value and local defect comparison value under the normal working distance are determined according to the method described in Example 1. Keeping the magnetization conditions, leakage magnetic field detection channel, detection direction, and signal acquisition conditions unchanged, the distance between the magnetic sensing position and the pipe wall is increased, and the same known metal loss area is passed through again. The leakage magnetic field signal after the increase in the lift-off distance is collected, and the corresponding peak value, valley value, peak-valley midline value, and local defect comparison value are determined. When the peak-to-valley midline quantity decreases along the current signal polarity direction after increasing the lift-off distance, the polarity of the leakage magnetic signal in the current detection task is uniformly reversed; after completing the polarity unification, the two states of normal working distance and increased lift-off distance are re-compared; When the peak-valley midline value increases and the local defect contrast value decreases after increasing the lift-off distance, it is confirmed that the current magnetic flux leakage signal of the magnetic flux leakage detection device meets the change direction adopted by the reverse collapse identification of the present invention, and the magnetic flux leakage signal is allowed to enter the reverse collapse identification process described in Embodiment 1; when the increase in the lift-off distance does not simultaneously form the change relationship of the increase in the peak-valley midline value and the decrease in the local defect contrast value, the magnetic flux leakage signal is not used to perform the reverse collapse channel interval identification and local relative continuous recovery described in the present invention. The above calibration verification only confirms whether the current leakage magnetic signal has the lift-off response direction utilized in this invention. It does not require establishing a fixed function between the actual lift-off distance and the peak-valley midline quantity, nor does it require establishing a fixed proportional relationship between the actual lift-off distance and the local defect comparison quantity. Therefore, in the actual detection process of this invention, the relative spatial change within the same cross-channel response group is used, rather than directly inverting the absolute lift-off distance based on the peak-valley midline quantity or directly inverting the actual defect depth based on the local defect comparison quantity. Furthermore, for a circumferentially extended defect in an actual in-service pipeline, when there is no obvious non-uniform dynamic lift-off in the region, multiple circumferentially consecutive adjacent leakage magnetic flux detection channels may still generate effective peak-valley responses and can establish cross-channel response groups. However, if the Spearman rank correlation coefficient between its common mode variation sequence and the defect comparison sequence is not less than zero, then the negative rank correlation condition of this invention is not met, and the reverse collapse channel interval is not established. Alternatively, even if the Spearman rank correlation coefficient between the common mode variation sequence and the defect comparison sequence is less than zero, the magnetic flux leakage detection channel corresponding to the maximum peak-valley midline quantity is different from the magnetic flux leakage detection channel corresponding to the minimum local defect comparison quantity, and there is no common magnetic flux leakage detection channel under the condition of parallel extreme values, and no reverse collapse channel interval is established. For the cross-channel response group that does not meet the reverse collapse condition, keep the corresponding original leakage magnetic signal unchanged and directly enter the anti-lift-off multi-channel leakage magnetic signal matrix. Therefore, this invention does not correct all abnormal responses that are continuously distributed across multiple leakage magnetic flux detection channels, but only performs local relative continuous recovery on cross-channel response groups that simultaneously have inverse changes in peak-valley midline quantity and local defect contrast quantity, and that have a spatial correspondence between the maximum common mode position and the minimum local defect contrast position.

[0036] For cross-channel continuous magnetic flux leakage response formed purely by actual circumferential expansion defects but without the aforementioned reverse collapse relationship, the present invention preserves its original magnetic flux leakage signal, thereby avoiding the mistaken deletion of the real circumferential expansion defect as common-mode interference simply because multiple magnetic flux leakage detection channels show abnormality together. After completing all the reverse collapse recovery that meets the conditions, a complete anti-lift-off multi-channel leakage magnetic signal matrix is ​​formed, and the effective peak-valley response after recovery is extracted again in accordance with the method described in Example 1 S6. For peak-valley responses where the local defect contrast after recovery is not greater than the upper limit of the corresponding background peak-valley, they are excluded as background fluctuations. For magnetic flux leakage responses where the local defect contrast after recovery is greater than the upper limit of the corresponding background peak-valley, and which can form an axial overlap relationship with the effective peak-valley responses after recovery in the continuously adjacent magnetic flux leakage detection channels in the circumferential direction of the pipeline, their corresponding continuous mapping area is determined as the defect area of ​​the in-service pipeline.

[0037] The above processing can reduce the attenuation of magnetic flux leakage response and imaging breakage in the middle of the real circumferential expansion defect caused by non-uniform dynamic lifting, while avoiding the simple weakening of the real cross-channel continuous defect as a whole as common mode interference, thus improving the continuity of multi-channel magnetic flux leakage imaging of defects in in-service pipelines and the reliability of defect area identification.

[0038] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A multi-channel magnetic flux leakage signal anti-interference imaging and identification method for defects in in-service pipelines, characterized in that, Includes the following steps: The leakage magnetic signals of multiple leakage magnetic detection channels arranged along the circumference of the pipeline are collected, and a multi-channel leakage magnetic signal matrix is ​​established according to the axial position of the pipeline and the circumferential position of the leakage magnetic detection channels. Peak and valley responses are extracted from the multi-channel magnetic flux leakage signal matrix. Peak and valley responses that overlap in axial range and are continuously adjacent in circumferential direction are associated to form cross-channel response groups, and the peak and valley values ​​of each magnetic flux leakage detection channel are determined. Based on the peak and valley values, the peak-valley midline quantity and local defect comparison quantity of each magnetic flux leakage detection channel are determined, and a common mode variation sequence and a defect comparison sequence are formed respectively. The reverse collapse channel interval is determined based on the inverse change corresponding to the increase in the peak-valley midline amount in the common mode variation sequence and the decrease in the local defect comparison amount in the defect comparison sequence. Based on the peak-valley midline quantity and local defect comparison quantity of the magnetic flux leakage detection channels on both sides of the reverse collapse channel interval, the reference midline quantity and reference defect comparison quantity of each magnetic flux leakage detection channel in the interval are determined, and the magnetic flux leakage signal in the interval is corrected to obtain the anti-lift-off multi-channel magnetic flux leakage signal matrix. Based on the anti-lift-off multi-channel magnetic flux leakage signal matrix, a pipe wall magnetic flux leakage image is generated, and the defect area of ​​the in-service pipeline is determined.

2. The multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 1, characterized in that, The establishment of the multi-channel leakage magnetic signal matrix includes: Acquire the magnetic flux leakage signal sampling data of each magnetic flux leakage detection channel and the axial position of the detection device corresponding to each magnetic flux leakage signal sampling data; Based on the axial position of the detection device, the leakage magnetic signal sampling data of different leakage magnetic detection channels are mapped to a unified pipeline axial position coordinate. The multi-channel magnetic flux leakage signal matrix is ​​obtained by mapping the magnetic flux leakage signal sampling data according to the circumferential arrangement of the magnetic flux leakage detection channels and the axial position coordinates of the pipeline.

3. The multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 1, characterized in that, The extraction of peak and valley responses and the formation of cross-channel response groups include: Along the axial direction of the pipeline, for each magnetic flux leakage detection channel, according to the arrangement order of the axial positions of the pipeline, the magnetic flux leakage signal sampling data corresponding to the axial position of the pipeline is subtracted from the magnetic flux leakage signal sampling data corresponding to the axial position of the pipeline to determine the magnetic flux leakage signal difference between adjacent axial positions of the pipeline. When the difference in the magnetic flux leakage signal on the front side of a pipe's axial position is greater than zero and the difference in the magnetic flux leakage signal on the rear side is less than zero, the axial position of the pipe is determined as a local maximum position; when the difference in the magnetic flux leakage signal on the front side of a pipe's axial position is less than zero and the difference in the magnetic flux leakage signal on the rear side is greater than zero, the axial position of the pipe is determined as a local minimum position. A peak-valley response is formed by taking an axially adjacent local maximum and a local minimum and the leakage magnetic signal between them, and the leakage magnetic signal value corresponding to the local maximum is determined as the peak value, and the leakage magnetic signal value corresponding to the local minimum is determined as the valley value. Obtain the upper limit of the background peak and valley corresponding to each magnetic flux leakage detection channel. Determine the peak and valley response that is greater than the corresponding upper limit of the background peak and valley as the valid peak and valley response, and determine the peak and valley response that does not meet the above conditions as background fluctuation. The effective peak-valley responses of different magnetic flux leakage detection channels that overlap in axial range and are continuously adjacent in circumferential direction are correlated to form the cross-channel response group.

4. The multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 1, characterized in that, The peak-valley midline value is the arithmetic mean of the peak and valley values ​​corresponding to the same magnetic flux leakage detection channel, and the local defect comparison value is the absolute difference between the peak and valley values ​​corresponding to the same magnetic flux leakage detection channel.

5. The multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 4, characterized in that, The formation of the common mode variation sequence and the defect comparison sequence includes: Arrange the midline values ​​of each peak and valley according to the circumferential arrangement order of the leakage magnetic flux detection channels within the cross-channel response group to obtain the common mode variation sequence; The local defect comparison quantities are arranged in the same circumferential arrangement order of the magnetic flux leakage detection channels to obtain the defect comparison sequence.

6. The multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 5, characterized in that, Before determining the reverse collapse channel interval, the following is included: Determine the magnetic flux leakage detection channel corresponding to the maximum peak-valley midline amount in the common mode variation sequence and the magnetic flux leakage detection channel corresponding to the minimum local defect comparison amount in the defect comparison sequence. Determine the Spearman rank correlation coefficient between the common mode variation sequence and the defect comparison sequence; When the maximum peak-valley midline quantity and the minimum local defect comparison quantity correspond to the same magnetic flux leakage detection channel, and the Spearman rank correlation coefficient is less than zero, the magnetic flux leakage detection channel is determined as the reverse change positioning reference.

7. The multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 5, characterized in that, When the maximum peak-to-valley midline value in the common-mode variation sequence or the minimum local defect comparison value in the defect comparison sequence corresponds to multiple magnetic flux leakage detection channels, including: The first set of magnetic flux leakage detection channels corresponding to the maximum peak-valley midline amount and the second set of magnetic flux leakage detection channels corresponding to the minimum local defect comparison amount are respectively formed; Determine the Spearman rank correlation coefficient between the common mode variation sequence and the defect comparison sequence; When the Spearman rank correlation coefficient is less than zero and the first magnetic flux leakage detection channel set and the second magnetic flux leakage detection channel set have a common magnetic flux leakage detection channel, the continuous channel segment where the common magnetic flux leakage detection channel is located is determined as the reverse change positioning region.

8. A multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 6 or 7, characterized in that, The determination of the reverse collapse channel interval includes: The nearest local low point in the common mode change sequence is retrieved from both sides of the pipe circumference using the reverse change positioning reference, or the nearest local low point in the common mode change sequence is retrieved from both ends of the reverse change positioning area outward along the pipe circumference. The magnetic flux leakage detection channel whose peak-valley midline value is not greater than the corresponding peak-valley midline value of the adjacent magnetic flux leakage detection channels on both sides of the circumference is determined as the local low point. The range of continuous channels between the leakage magnetic flux detection channels corresponding to the nearest local low points on both sides is defined as the reverse collapse channel interval.

9. A multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 8, characterized in that, The leakage magnetic signal within the correction interval includes: Using the peak-valley midline value of the leakage magnetic flux detection channel at both sides of the reverse collapse channel interval as the boundary value, linear interpolation is performed according to the circumferential arrangement position of the leakage magnetic flux detection channel to obtain the reference midline value of each leakage magnetic flux detection channel inside the interval. Using the local defect comparison amount of the magnetic flux leakage detection channels on both sides as the boundary value, linear interpolation is performed according to the circumferential arrangement position of the magnetic flux leakage detection channels to obtain the reference defect comparison amount of each magnetic flux leakage detection channel within the interval. Determine the ratio of the reference defect comparison value to the corresponding local defect comparison value for each magnetic flux leakage detection channel. When the ratio is greater than one, the ratio is determined as the defect comparison recovery coefficient. When the ratio is not greater than one, one is determined as the defect comparison recovery coefficient. The difference between the original magnetic flux leakage signal and the corresponding peak-valley midline quantity of each magnetic flux leakage detection channel within the interval is recovered by amplitude recovery according to the defect comparison recovery coefficient, and the difference after amplitude recovery is combined with the corresponding reference midline quantity to obtain the corrected magnetic flux leakage signal.

10. The multi-channel magnetic flux leakage signal anti-interference imaging identification method for defects in in-service pipelines according to claim 9, characterized in that, The process of obtaining the anti-lift-off multi-channel leakage magnetic signal matrix and determining the defect region of the in-service pipeline includes: Keeping the leakage magnetic signal outside the reverse collapse channel interval unchanged, the corrected leakage magnetic signal inside the reverse collapse channel interval and the leakage magnetic signal outside the reverse collapse channel interval are recombined according to the circumferential arrangement position of the leakage magnetic detection channel to obtain the anti-lift-off multi-channel leakage magnetic signal matrix. The anti-lift-off multi-channel magnetic flux leakage signal matrix is ​​mapped to the pipe wall magnetic flux leakage image according to the axial position of the pipe and the circumferential arrangement of the magnetic flux leakage detection channels. The recovered peak-valley response is extracted from the anti-lift-off multi-channel magnetic flux leakage signal matrix. The connected mapping region of the recovered peak-valley response with overlapping axial ranges and corresponding magnetic flux leakage detection channels that are continuously adjacent in the circumferential direction in the pipe wall magnetic flux leakage image is determined as the defect region of the in-service pipeline. Based on the axial position of the pipeline corresponding to the defect area in the in-service pipeline and the covered magnetic flux leakage detection channel, the axial position and circumferential coverage of the defect are determined.

Images