CN-121994908-A - Method and device for identifying defects or flanges based on magnetic leakage signals

CN121994908ACN 121994908 ACN121994908 ACN 121994908ACN-121994908-A

Abstract

The embodiment of the specification provides a method and a device for identifying defects or flanges based on magnetic leakage signals, wherein the method comprises the steps of detecting a pipeline to be identified by using a magnetic leakage detection instrument to obtain abnormal data of the magnetic leakage signals, performing invalid elimination and blank filling processing on the abnormal data of the magnetic leakage signals to obtain abnormal data segments, extracting characteristic values of the abnormal data segments to obtain a plurality of characteristic values for expressing waveforms of the magnetic leakage signals, selecting a first characteristic value and a second characteristic value in the characteristic values, inputting the first characteristic value into a trained defect and flange identification model to obtain defects or flanges of the pipeline to be identified, and inputting the second characteristic value into a trained defect type identification model to obtain defect types corresponding to the pipeline to be identified when the defects exist in the pipeline to be identified. According to the embodiment of the specification, the defects or the flanges can be identified, and effective detection of the health condition of the pipeline is further realized.

Inventors

HE MO
QIN LIN
LIN DONG
ZHOU ZHIYONG
LI CHAOLANG
SUN MINGNAN
CHEN HAN
GAO JIAN
TANG YU

Assignees

中国石油天然气股份有限公司

Dates

Publication Date: 20260508
Application Date: 20241107

Claims (12)

1. A method for identifying a defect or flange based on a leakage signal, comprising: Detecting the pipeline to be identified by using a magnetic flux leakage detection instrument to obtain abnormal data of a magnetic flux leakage signal; performing invalid elimination and blank filling treatment on the abnormal data of the magnetic leakage signal to obtain an abnormal data segment; Extracting the characteristic values of the abnormal data segments to obtain a plurality of characteristic values for expressing the waveform of the magnetic leakage signal; Selecting a first characteristic value and a second characteristic value in the plurality of characteristic values, wherein the first characteristic value has correlation with the flange and the defects, and the second characteristic value has correlation with different types of defects; Inputting the first characteristic value into a trained defect and flange identification model to obtain the defect or flange of the pipeline to be identified; when the pipeline to be identified has defects, the second characteristic value is input into a trained defect type identification model, and the defect type corresponding to the pipeline to be identified is obtained.
2. The method of claim 1, wherein obtaining abnormal data of the magnetic flux leakage signal after detecting the pipe to be identified by the magnetic flux leakage detecting instrument further comprises: detecting the pipeline to be identified by using a plurality of probes on the magnetic flux leakage detection instrument; Taking a probe of which abnormal data of a magnetic leakage signal is detected as an affected probe, and taking four channels of the affected probe as affected channels; And extracting abnormal data of a group of magnetic flux leakage signals detected by each affected channel in the axial direction of the pipeline to be identified.
3. The method of claim 2, wherein obtaining the abnormal data segment after performing the invalidation rejection and the blank filling processing on the abnormal data of the magnetic flux leakage signal further comprises: Carrying out curve processing on the abnormal data of each group of magnetic leakage signals to obtain each initial data segment, wherein the initial data segments are in curve forms; removing invalid data in a single form in each initial data segment to obtain abnormal data after each group of removal; And calculating the average value of the abnormal data after each group of elimination, filling the blank of the initial data segment by using the average value to obtain each abnormal data segment, wherein the abnormal data segment is in the form of a continuous curve, the horizontal axis of the abnormal data segment is a mileage value, and the vertical axis of the abnormal data segment is an abnormal data value.
4. The method of claim 3, wherein the plurality of characteristic values for describing the waveform of the leakage signal comprises a peak-to-valley value, an axial spacing value, an amplitude value, a waveform area, a waveform energy, and an affected channel number.
5. The method of claim 4, wherein extracting the eigenvalues of the anomaly data segments to obtain a plurality of eigenvalues representing leakage signal waveforms further comprises: selecting an abnormal data segment with the largest abnormal data difference as a target abnormal data segment; obtaining the maximum value and the minimum value of the abnormal data in the target abnormal data segment, and calculating the absolute value of the difference value between the maximum value and the minimum value to be used as a peak-valley value; Acquiring a mileage value of first abnormal data and a mileage value of last abnormal data in the target abnormal data segment, and calculating an absolute value of a difference value of the two mileage values as an axial distance value; acquiring the absolute value of the abnormal data with the maximum absolute value in the target abnormal data segment as an amplitude value; acquiring the area surrounded by the waveform with the maximum abnormal data fluctuation in the target abnormal data segment as a waveform area; Obtaining the discrete degree of the abnormal data in the target abnormal data segment to obtain waveform energy; the number of affected channels is obtained as the number of affected channels.
6. The method of claim 5, wherein the obtaining the degree of dispersion of the anomaly data in the target anomaly data segment yields waveform energy calculated by the formula: Wherein t n is the mileage value of the nth abnormal data, N is the number of abnormal data in the target abnormal data segment, x (t n ) is the value of the nth abnormal data, min [ x (t n ) ] is the minimum value of the abnormal data in the target abnormal data segment, and E is waveform energy.
7. The method of claim 1, wherein the training of the defect and flange identification model comprises: detecting the known pipeline by using a magnetic flux leakage detection instrument to obtain known abnormal data of the known magnetic flux leakage signal; invalid elimination and blank filling processing are carried out on the known abnormal data of the known magnetic leakage signal, and a known abnormal data segment is obtained; Extracting the characteristic values of the known abnormal data segments to obtain a plurality of known characteristic values for expressing the waveforms of the known magnetic leakage signals; selecting a first known characteristic value in a plurality of known characteristic values; and taking the first known characteristic value of the known pipeline as an input value of a defect and flange identification model, taking the defect or the flange of the known pipeline as an output value, and training the defect and flange identification model.
8. The method of claim 1, wherein the training process of the defect type identification model comprises: detecting the known pipeline by using a magnetic flux leakage detection instrument to obtain known abnormal data of the known magnetic flux leakage signal; invalid elimination and blank filling processing are carried out on the known abnormal data of the known magnetic leakage signal, and a known abnormal data segment is obtained; Extracting the characteristic values of the known abnormal data segments to obtain a plurality of known characteristic values for expressing the waveforms of the known magnetic leakage signals; selecting a second known feature value of the plurality of known feature values; and taking the second known characteristic value of the known pipeline as an input value of a defect type identification model, taking the defect type of the known pipeline as an output value, and training the defect type identification model.
9. An apparatus for identifying defects or flanges based on magnetic leakage signals, the apparatus comprising: the detection module is used for detecting the pipeline to be identified by using the magnetic leakage detection instrument to obtain abnormal data of the magnetic leakage signal; The processing module is used for carrying out invalid elimination and blank filling processing on the abnormal data of the magnetic leakage signal to obtain an abnormal data segment; The extraction module is used for extracting the characteristic values of the abnormal data segments to obtain a plurality of characteristic values for expressing the waveform of the magnetic leakage signal; The selection module is used for selecting a first characteristic value and a second characteristic value in the plurality of characteristic values, wherein the first characteristic value has correlation with the flange and the defects, and the second characteristic value has correlation with the defects of different types; The first recognition module is used for inputting the first characteristic value into a trained defect and flange recognition model to obtain the defect or flange of the pipeline to be recognized; And the second recognition module is used for inputting the second characteristic value into the trained defect type recognition model to obtain the defect type corresponding to the pipeline to be recognized when the pipeline to be recognized has defects.
10. A computer device comprising a memory, a processor, and a computer program stored on the memory, characterized in that the computer program, when being executed by the processor, performs the instructions of the method according to any of claims 1-8.
11. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor of a computer device, executes instructions of the method according to any one of claims 1-8.
12. A computer program product, characterized in that the computer program product, when being executed by a processor of a computer device, executes instructions of the method according to any of claims 1-8.

Description

Method and device for identifying defects or flanges based on magnetic leakage signals Technical Field Embodiments of the present disclosure relate to the field of non-destructive testing, and in particular, to a method and apparatus for identifying defects or flanges based on magnetic leakage signals. Background In general, oil gas is conveyed through a pipeline, defects in the pipeline, such as pit-shaped metal loss, axial grooves and the like, need to be avoided during conveying, and a magnetic leakage signal refers to a phenomenon that when a pipeline material has defects of cutting magnetic lines, magnetic permeability at the defect is small and magnetic resistance is large, so that magnetic flux in a magnetic circuit is distorted, and a magnetic leakage field is formed on the surface of the material. The leakage magnetic field can be captured by a magnetic induction sensor and transmitted to a computer for processing, and the defect characteristics of the material, such as width, depth and the like, can be further known through analyzing the magnetic flux density component of the leakage magnetic field. In reality, however, the pipeline may generate magnetic flux leakage signals except for defects, and the flange at the joint of the pipeline and the pipeline may also generate magnetic flux leakage signals, while the flange belongs to a necessary device for pipeline connection, and when pipeline detection is performed, whether the magnetic flux leakage signals are generated due to defects or the flange needs to be distinguished, so that effective detection of the health condition of the pipeline is realized. Therefore, there is a need for a method for identifying defects or flanges based on magnetic leakage signals, which can identify the defects or flanges, thereby realizing effective detection of the health condition of the pipeline. Disclosure of Invention An object of the embodiments of the present disclosure is to provide a method and an apparatus for identifying a defect or a flange based on a magnetic leakage signal, so as to identify the defect or the flange, and further realize effective detection of a pipeline health condition. To achieve the above object, in one aspect, an embodiment of the present disclosure provides a method for identifying a defect or a flange based on a magnetic leakage signal, including: Detecting the pipeline to be identified by using a magnetic flux leakage detection instrument to obtain abnormal data of a magnetic flux leakage signal; performing invalid elimination and blank filling treatment on the abnormal data of the magnetic leakage signal to obtain an abnormal data segment; Extracting the characteristic values of the abnormal data segments to obtain a plurality of characteristic values for expressing the waveform of the magnetic leakage signal; Selecting a first characteristic value and a second characteristic value in the plurality of characteristic values, wherein the first characteristic value has correlation with the flange and the defects, and the second characteristic value has correlation with different types of defects; Inputting the first characteristic value into a trained defect and flange identification model to obtain the defect or flange of the pipeline to be identified; when the pipeline to be identified has defects, the second characteristic value is input into a trained defect type identification model, and the defect type corresponding to the pipeline to be identified is obtained. Preferably, after the detecting the pipeline to be identified by using the magnetic flux leakage detecting instrument, obtaining the abnormal data of the magnetic flux leakage signal further includes: detecting the pipeline to be identified by using a plurality of probes on the magnetic flux leakage detection instrument; Taking a probe of which abnormal data of a magnetic leakage signal is detected as an affected probe, and taking four channels of the affected probe as affected channels; And extracting abnormal data of a group of magnetic flux leakage signals detected by each affected channel in the axial direction of the pipeline to be identified. Preferably, after the invalidation rejection and the blank filling processing are performed on the abnormal data of the magnetic leakage signal, obtaining the abnormal data segment further includes: Carrying out curve processing on the abnormal data of each group of magnetic leakage signals to obtain each initial data segment, wherein the initial data segments are in curve forms; removing invalid data in a single form in each initial data segment to obtain abnormal data after each group of removal; And calculating the average value of the abnormal data after each group of elimination, filling the blank of the initial data segment by using the average value to obtain each abnormal data segment, wherein the abnormal data segment is in the form of a continuous curve, the horizontal axis of the abnormal data segment is a mileag