CN115096987A - Pipeline defect quantification method based on magnetic flux leakage signal characteristics - Google Patents
Pipeline defect quantification method based on magnetic flux leakage signal characteristics Download PDFInfo
- Publication number
- CN115096987A CN115096987A CN202210729456.6A CN202210729456A CN115096987A CN 115096987 A CN115096987 A CN 115096987A CN 202210729456 A CN202210729456 A CN 202210729456A CN 115096987 A CN115096987 A CN 115096987A
- Authority
- CN
- China
- Prior art keywords
- defect
- depth
- peak
- component
- width
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 359
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000004907 flux Effects 0.000 title claims abstract description 32
- 238000011002 quantification Methods 0.000 title claims abstract description 13
- 238000013528 artificial neural network Methods 0.000 claims abstract description 41
- 238000013139 quantization Methods 0.000 claims abstract description 41
- 238000012549 training Methods 0.000 claims description 30
- 238000012360 testing method Methods 0.000 claims description 20
- 239000008186 active pharmaceutical agent Substances 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 16
- 238000012937 correction Methods 0.000 claims description 9
- 238000003062 neural network model Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 abstract description 3
- 238000004088 simulation Methods 0.000 description 11
- 238000001514 detection method Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000004445 quantitative analysis Methods 0.000 description 5
- 238000012795 verification Methods 0.000 description 4
- 230000007423 decrease Effects 0.000 description 3
- 230000005284 excitation Effects 0.000 description 3
- 230000006698 induction Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000005415 magnetization Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/72—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables
- G01N27/82—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws
- G01N27/83—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws by investigating stray magnetic fields
Landscapes
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)
Abstract
The invention discloses a pipeline defect quantification method based on magnetic flux leakage signal characteristics, which is used for obtaining three-axis magnetic flux leakage signals of defects under different defect sizes and constructing a sample set; selecting the peak-valley distance of the axial component differential signal as a quantization parameter of the defect length to obtain a defect length quantization formula: selecting the median distance between the circumferential component peaks and valleys as a quantization parameter of the defect width to obtain defect width quantization; and acquiring a mapping relation between the length and the width of the defect and a defect depth fitting parameter by establishing a neural network, and calculating the depth of the defect by taking the defect depth fitting parameter and the radial component peak-valley value as quantitative parameters of the depth of the defect. The method can accurately quantize the defect size under the conditions that the geometrical parameters of the residual defects are unknown or the lifting distance fluctuates within a certain range and the like, and has stronger anti-interference capability and simpler quantizing mode.
Description
Technical Field
The invention relates to the technical field of pipeline defect detection, in particular to a pipeline defect quantification method based on magnetic flux leakage signal characteristics.
Background
In pipeline safety engineering, pipeline inspection is a basic method for ensuring pipeline safety. Among various types of pipeline detection technologies, the magnetic flux leakage internal detection technology is the most widely applied and technically mature magnetic pipeline defect detection technology. The size of the pipeline defect cannot be directly measured in the operation process of the internal magnetic flux leakage detection technology, the core problem of the internal magnetic flux leakage detection technology is the size inversion problem of the pipeline defect, and the size of the defect is reduced through the collected magnetic flux leakage signal. Due to the complex relationship between the leakage magnetic signal and the defect size, the analysis of the leakage magnetic signal to realize the defect quantization is a technical problem. The traditional defect quantification method mostly takes axial component characteristics as evaluation, the signal characteristic source is single, the identification degree is not high, and the defect quantification precision is reduced.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a pipeline defect quantification method based on magnetic flux leakage signal characteristics, which can identify the type of the pipeline defect according to the magnetic flux leakage signal of the pipeline, has stronger anti-interference capability and simpler quantification mode, and has great engineering significance and good application prospect.
In order to achieve the purpose, the invention adopts the following technical scheme that:
a pipeline defect quantification method based on leakage magnetic signal characteristics comprises the following steps:
s1, obtaining triaxial magnetic flux leakage signals of the defects under different defect sizes, and constructing a sample set;
the defect size includes: defect length, defect width, defect depth;
the three-axis magnetic flux leakage signal of the defect refers to: the axial magnetic leakage signal is an axial component, the radial magnetic leakage signal is a radial component, and the circumferential magnetic leakage signal is a circumferential component;
the axial direction is the direction along the length of the pipeline, the radial direction is the direction vertical to the inner wall of the pipeline, and the circumferential direction is the circumferential direction along the pipeline;
the selection mode of the circumferential component is as follows: determining the peak position of the radial component on the axial path, and selecting the value of the radial component on the circumferential path where the peak position is as the value of the circumferential component;
s2, differentiating the axial component in the sample set to obtain an axial component differential signal, and extracting the peak-to-valley interval DS of the axial component differential signal xp-p As a quantization parameter for the defect length L, the peak-to-valley distance DS of the differential signal of the axial component in the sample set xp-p Linear fitting is carried out on the defect length L to obtain the defect lengthThe metric quantization formula:
L＝a 1 *DS xp-p +b 1
wherein, a 1 Is a length scale factor, b 1 Is a length correction factor;
s3, extracting the median space S between the peaks and the valleys of the circumferential components in the sample set y-50％ As a quantization parameter of the defect width W, the peak-to-valley median spacing S of the circumferential components in the sample set y-50％ And performing linear fitting on the defect width W to obtain a defect width measurement formula:
W＝a 2 *S y-50％ +b 2
wherein, a 2 Is a width proportionality coefficient, b 2 Is the width correction factor;
median circumferential component peak-to-valley spacing S y-50％ The extraction method comprises the following steps: extracting the wave crest and the wave trough of the circumferential component, calculating the middle value of the wave crest and the wave trough, namely 50% of the difference between the wave crest and the wave trough as the median of the wave crest and the wave trough, and the distance between the median of the wave crest and the wave trough on the circumferential component is the distance S between the median of the wave crest and the wave trough on the circumferential component y-50％ ；
S4, extracting the radial component peak-valley value B in the sample set zp-p I.e. the difference between the peak and the trough of the radial component, the peak-to-trough value B of the radial component zp-p As a quantization parameter of the defect depth D, the peak-to-valley value B of the radial component in the sample set zp-p Fitting with the defect depth D to obtain a defect depth quantization formula:
D＝a*B zp-p +b
wherein a and b are both defect depth fitting parameters and are both variables, and when the defect length or defect width changes, the corresponding defect depth fitting parameters a and b also change;
s5, constructing a neural network, wherein the input of the neural network is the defect length L and the defect width W, the output of the neural network is defect depth fitting parameters a and b, and the neural network is generated by training with a sample set;
s6, the process of quantifying the unknown defects is as follows:
s61, differentiating the peak-to-valley distance D of the signal according to the axial component of the unknown defectS xp-p And a defect length quantization formula, calculating to obtain the defect length L of the unknown defect;
s62, according to the mean distance S between the peaks and the valleys of the circumferential component of the unknown defect y-50％ And a defect width measurement formula, calculating to obtain the defect width W of the unknown defect;
s63, inputting the calculated length L and width W of the unknown defect into a neural network model, and predicting to obtain defect depth fitting parameters a and b;
s64, determining the radial component peak-to-valley value B of the unknown defect zp-p And substituting the predicted defect depth fitting parameters a and b into a defect depth quantization formula to calculate the defect depth D of the unknown defect.
Preferably, in step S1, three-axis leakage magnetic signals of the defect with different defect sizes are simulated by finite element software modeling.
Preferably, in step S5, the sample set is divided into a training set and a test set, and the neural network is trained using the training set:
for the radial component peak-valley value B with the same defect length and defect width in the training set zp-p Fitting with the defect depth D to obtain defect depth fitting parameters a and b under the defect length and the defect width so as to obtain training data of the neural network, and training the neural network by using the training data of the neural network.
Preferably, in step S5, the sample set is divided into a training set and a test set, and the neural network is tested by using the test set:
inputting the defect length L and the defect width W of the known defects in the test set into the trained neural network, and predicting to obtain corresponding defect depth fitting parameters a and b; fitting corresponding defect depth to parameters a and B and radial component peak-valley value B of the defect zp-p Substituting the defect depth into a defect depth quantization formula, and calculating to obtain a predicted value of the defect depth D; and comparing the predicted value of the defect depth D with the real value of the defect depth D to test the prediction accuracy of the neural network.
The invention has the advantages that:
(1) according to the method, through analysis of sample data, the defect length is characterized to be most stable by utilizing the peak-valley distance of the axial component differential signal, the peak-valley distance of the axial component differential signal is not influenced by factors such as defect width, defect depth, sensor lift-off value and the like, namely when the interference amount changes, the peak-valley distance of the axial component differential signal cannot be changed along with the change of the defect length, and the peak-valley distance of the axial component differential signal only changes along with the change of the defect length, so that the peak-valley distance of the axial component differential signal is selected as a quantization parameter of the defect length according to the method, and a defect length quantization formula is constructed according to a sample set.
(2) According to the method, through analysis of sample data, only the signal characteristic quantity of the circumferential component peak-valley median distance can be called as a stable quantity and does not change along with the change of the defect depth and the lifting-away distance, so that the circumferential component peak-valley median distance is selected as a quantization parameter of the defect width according to the method, and a defect width quantization formula is constructed according to a sample set.
(3) According to the method, through analysis of sample data, although two signal characteristic quantities, namely a radial component peak-valley value and an axial component peak-valley value, are suitable for being used as signal characteristic quantities for quantifying the defect depth, the two signal characteristic quantities are also influenced by the change of the defect size, therefore, the defect length is determined by the distance between the axial component differential signal peaks and valleys, the defect width size is determined by the median distance between the circumferential component peaks and valleys, the mapping relation between the length and the width of the defect and defect depth fitting parameters a and b is obtained through establishing a BP neural network model, then the defect depth is calculated by taking the defect depth fitting parameter and the radial component peak-valley value as the quantification parameter of the defect depth, and the accuracy is higher than that of a method for quantifying the defect depth by directly using the radial component peak-valley value.
(4) The method simulates pipeline magnetic flux leakage signals under different defect sizes through finite element modeling, researches and analyzes the mutual influence relationship among a plurality of signal characteristic quantities, the defect length, the defect width and the defect depth, can obtain the representation effectiveness of the defect length, the defect width and the defect depth and each signal characteristic quantity, specifically analyzes the superiority and inferiority of the signal characteristic quantity capable of representing the geometrical parameters of the defect, and obtains an optimal representation result.
(5) The method provided by the invention has the advantages that the sample data acquired under the conditions of different working scenes, different equipment, different materials, different magnetization conditions and the like are different, but the integral quantitative prediction method is suitable for all occasions.
Drawings
Fig. 1 is a flowchart of a method for quantifying a pipe defect based on characteristics of a leakage magnetic signal according to the present invention.
Fig. 2 is a leakage flux plot of an axial component differential signal.
Fig. 3 is a magnetic flux leakage graph of the circumferential component.
Fig. 4 is a graph of leakage flux for the radial component.
Fig. 5 is a graph of leakage flux for differential signals of axial components of different defect lengths.
FIG. 6 is a graph of axial component differential signal peak-to-valley spacing versus defect length.
FIG. 7 is a graph of peak-to-valley spacing of axial component differential signals versus defect width for different defect lengths.
FIG. 8 is a graph of peak-to-valley spacing of axial component differential signals versus defect depth for different defect lengths.
FIG. 9 is a graph of peak-to-valley spacing of differential signals of axial components versus lift-off values for different defect lengths.
Fig. 10 is a graph of leakage flux for circumferential components of different defect widths.
FIG. 11 is a graph of circumferential component median peak-to-valley spacing versus defect width.
FIG. 12 is a graph of median peak-to-valley spacing versus defect depth for circumferential components of different defect widths.
FIG. 13 is a graph of the median spacing between peaks and valleys versus lift-off for circumferential components of different defect widths.
Fig. 14 is a plot of leakage flux for radial components of different defect depths.
FIG. 15 is a graph of radial component peak-to-valley value versus defect depth for the same defect length and width.
FIG. 16 is a graph of radial component peak-to-valley values versus defect depth for different defect lengths and widths.
Fig. 17 is a schematic diagram of the training result of the BP neural network.
Fig. 18 is a comparison graph of predicted values of defect depths and actual values of defect depths for different defect lengths and widths.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
As shown in fig. 1, a method for quantifying a pipeline defect based on a leakage magnetic signal characteristic includes the following steps:
and S1, acquiring the triaxial magnetic flux leakage signals of the defects under different defect sizes, and constructing a sample set.
The defect size includes: defect length, defect width, defect depth.
The three-axis magnetic flux leakage signal of the defect refers to: the axial magnetic leakage signal is an axial component, the radial magnetic leakage signal is a radial component, and the circumferential magnetic leakage signal is a circumferential component;
the axial direction is the direction along the length of the pipeline, the radial direction is the direction vertical to the inner wall of the pipeline, namely the normal direction of the inner wall of the pipeline, and the circumferential direction is the circumferential direction along the circumference of the pipeline, namely the circumferential direction surrounding the pipeline;
the circumferential component is constructed by selecting a value of a radial component on a circumferential path of the pipeline, and the specific selection mode is as follows: determining the peak position of the radial component on the axial path, and selecting the value of the radial component on the circumferential path where the peak position is as the value of the circumferential component;
s2, differentiating the axial component in the sample set to obtain an axial component differential signal, and extracting the axial component differential signalPeak-to-valley spacing DS xp-p The peak-to-valley spacing DS of the differential signal of the axial component in the sample set as a quantization parameter for the defect length L xp-p And performing linear fitting on the defect length L to obtain a defect length quantization formula:
L＝a 1 *DS xp-p +b 1
wherein, a 1 Is a length scale factor, b 1 Is a length correction factor;
axial component differential signal peak-to-valley spacing DS xp-p As shown in fig. 2, fig. 2 is a magnetic flux leakage graph of the axial component differential signal, in which the ordinate represents the magnetic induction differential value and the abscissa represents the axial distance.
S3, extracting the median space S between the peaks and the valleys of the circumferential components in the sample set y-50％ As a quantization parameter of the defect width W, the peak-to-valley median spacing S of the circumferential components in the sample set y-50％ And performing linear fitting on the defect width W to obtain a defect width quantization formula:
W＝a 2 *S y-50％ +b 2
wherein, a 2 Is the width proportionality coefficient, b 2 Is a width correction factor;
median spacing S between circumferential component peaks and valleys y-50％ As shown in fig. 3, the extraction method is as follows: extracting the wave crest and the wave trough of the circumferential component, and calculating the intermediate value of the wave crest and the wave trough, namely the difference S between the wave crest and the wave trough max 50% of the circumferential component is used as the median value of the peak valley, and the distance between the median values of the peak valley on the circumferential component is the median distance S between the peak valley of the circumferential component y-50％ . Fig. 3 is a magnetic flux leakage graph of a circumferential component, in which the vertical axis represents magnetic induction and the horizontal axis represents a circumferential distance.
S4, extracting the peak-valley value B of the radial component in the sample set zp-p I.e. the difference between the peak and the trough of the radial component, the peak-to-trough value B of the radial component zp-p As a quantization parameter of the defect depth D, the peak-to-valley value B of the radial component in the sample set zp-p Fitting with the defect depth D to obtain a defect depth quantization formula:
D＝a*B zp-p +b
and when the defect length or the defect width is changed, the corresponding defect depth fitting parameters a and b are also changed.
Radial component peak-to-valley B zp-p As shown in fig. 4, and S in fig. 4 zp-p Representing the radial component peak-to-valley spacing. Fig. 4 is a graph of leakage flux of a radial component, in which the vertical axis represents magnetic induction and the horizontal axis represents axial distance.
And S5, constructing a neural network, inputting the neural network into the defect length L and the defect width W, outputting the parameters a and b of defect depth fitting, and training by using a sample set to generate the neural network.
Dividing a sample set into a training set and a testing set, and training a neural network by using the training set:
for the radial component peak-valley value B with the same defect length and defect width in the training set zp-p Fitting with the defect depth D to obtain defect depth fitting parameters a and b under the defect length and the defect width so as to obtain training data of the neural network, and training the neural network by using the training data of the neural network.
Testing the neural network by using a test set: inputting the defect length L and the defect width W of the known defects in the test set into the trained neural network, and predicting to obtain corresponding defect depth fitting parameters a and b; fitting corresponding defect depth to parameters a and B and radial component peak-valley value B of the defect zp-p Substituting the defect depth into a defect depth quantization formula, and calculating to obtain a predicted value of the defect depth D; and comparing the predicted value of the defect depth D with the true value of the defect depth D to test the prediction accuracy of the neural network.
S6, the process of quantifying the unknown defects is as follows:
s61, differentiating the signal peak-to-valley spacing DS according to the axial component of the unknown defect xp-p And a defect length quantization formula, calculating to obtain the defect length L of the unknown defect;
s62, according to the median distance S between the peaks and valleys of the circumferential component of the unknown defect y-50％ And a defect width measurement formula, calculating to obtain the defect width W of the unknown defect;
s63, inputting the calculated length L and width W of the unknown defect into a neural network model, and predicting to obtain defect depth fitting parameters a and b;
s64, determining the radial component peak-to-valley value B of the unknown defect zp-p And substituting the predicted defect depth fitting parameters a and b into a defect depth quantization formula to calculate the defect depth D of the unknown defect.
In the embodiment, a sample set is constructed by simulating the defect flux leakage signals under different defect sizes through finite element software modeling. The sample data is: the size of the defect and the corresponding triaxial leakage flux signal. The dimensions of the defect include the length, width, and depth of the defect. The three-axis leakage signal of the defect refers to: the axial magnetic leakage signal is an axial component, the radial magnetic leakage signal is a radial component, and the circumferential magnetic leakage signal is a circumferential component; in this embodiment, the circumferential component is constructed by selecting a value of a radial component on a circumferential path of the pipeline, and the specific selection method is as follows: the peak position of the radial component on the axial path is determined, and the value of the radial component on the circumferential path where the peak position is located is selected as the value of the circumferential component.
In this embodiment, after the length, width, and depth of the defect are quantitatively analyzed according to the sample set, a specific quantitative manner of the length, width, and depth of the defect is obtained, and the quantitative analysis is specifically as follows:
(1) quantitative analysis of defect length:
according to the sample set, the axial component peak-valley distance, the axial component differential signal peak-valley distance and the radial component peak-valley distance are all monotonically increased along with the increase of the defect length, namely a linear relation is formed; the radial component peak-to-valley is only an approximate monotone decrease; the axial component peak-to-valley is non-linear and decreases with increasing defect length.
Taking the defect width as 10mm and the depth as 7mm as an example, as shown in fig. 5, axial component differential signals of different defect lengths are obtained by finite element simulation, and the axial component differential signals are axial component derivatives along the axial direction; in this embodiment, finite element simulations were used to obtain defect lengths of 3, 6, 9, 12, 15, and 18mm, respectivelyThe axial component of the signal. According to the simulation result, the peak-valley spacing DS of the differential signals of the axial components of the defects with different lengths xp-p Proportional to the defect length, as shown in fig. 6.
In the three signal characteristic quantities of the axial component peak-valley distance, the axial component differential signal peak-valley distance and the radial component peak-valley distance, the defect length is characterized to be most stable by using the axial component differential signal peak-valley distance, the axial component differential signal peak-valley distance is not influenced by factors such as defect width, defect depth, sensor lift-off value and the like, namely when the interference quantity changes, the axial component differential signal peak-valley distance cannot be changed along with the change of the defect length, the axial component differential signal peak-valley distance only changes along with the change of the defect length, and the axial component differential signal peak-valley distance fluctuation precision error is verified in finite element simulation to be within 10%. To verify the conclusion, when the lifting distances are respectively set to be 1, 2, 3, 4 and 5mm, the defect widths are respectively 6, 8, 10, 12 and 14mm, and the defect depths are respectively 4, 6, 8, 10 and 12mm, magnetic flux leakage detection is performed to obtain the peak-valley distances of the axial component differential signals when the defect size and the lifting distance are respectively changed, wherein one defect length is a group and is divided into 5 groups, the defect lengths are respectively 6, 9, 12, 15 and 18mm, the peak-valley distances of the axial component differential signals and the corresponding defect widths, defect depths and lifting distances are obtained, the obtained data are generated into a broken line graph, as shown in fig. 7, 8 and 9, and the fluctuation percentage of each broken line is calculated to verify that the fluctuation precision errors of the peak-valley distances of the axial component differential signals are all within 10%. Secondly, the peak-to-valley distance of the axial component and the peak-to-valley distance of the radial component are relatively stable, and although the two signal characteristic quantities are not influenced by the defect width and the defect depth, the two signal characteristic quantities are influenced by the fluctuation of the lift-off value of the sensor. As the lift-off value increases, the axial component peak-to-valley spacing decreases and then becomes a single peak, while the radial component peak-to-valley spacing increases.
In summary, the axial component differential signal peak-to-valley spacing DS is selected xp-p As a quantization parameter for the defect length L, the peak-to-valley distance DS of the differential signal of the axial component in the sample set xp-p Linear fitting is carried out on the defect length L to obtain the defect length quantityThe formula is shown as follows:
L＝a 1 *DS xp-p +b 1
wherein, a 1 Is a length scale factor, b 1 As a length correction factor, a 1 、b 1 The device is a fixed value under general conditions, and can be adjusted according to specific conditions such as different working environments, pipes, excitation intensity, detector attributes and the like.
(2) Quantitative analysis of defect width:
when the magnetic leakage detector is excited axially, the sensors for detecting the magnetic leakage signals are arranged in the circumferential direction, the circumferential distance of the magnetic leakage signals detected by the sensors is approximately equal to the width of the defect, and the number of the sensors subjected to the defect excitation signals is close to the direct proportional relation with the width of the defect.
And extracting the circumferential component, namely determining the maximum position of the radial component on the axial path, namely the peak position, and then extracting the radial component on the circumferential path where the peak position is positioned as the circumferential component.
The circumferential components with different defect widths are obtained through finite element simulation, as shown in fig. 10, according to the simulation result, the peak-to-valley median spacing S of the circumferential components with different widths is known y-50％ Proportional to the defect width, as shown in fig. 11.
As can be known from finite element simulation analysis, the peak-to-valley value of the axial component and the peak-to-valley value of the radial component are monotonically increased along with the increase of the defect width, but the two signal characteristic quantities are influenced by other factors such as the change of the defect depth, the fluctuation of the lift-off value and the like. Only the signal characteristic quantity of the circumferential component peak-valley median distance can be called as a stable quantity, does not change along with the change of the defect depth and the lifting distance, and is verified in finite element simulation, and the fluctuation precision error is within 10 percent. It should be emphasized that the characteristic parameters such as the peak-to-valley distance of the axial component, the peak-to-valley distance of the radial component, the peak-to-valley distance of the differential signal of the axial component, etc. are kept unchanged when the defect width is changed, corresponding to the determination of the quantization parameter of the defect length.
Circumferential component peak-to-valley median spacing S in verifying quantified defect width y-50％ Setting the lifting distance to be 1, 2, 3 and 4mm respectively, setting the value range of the defect depth to be 4, 6, 8, 10 and 12mm respectively, carrying out magnetic flux leakage detection to obtain the circumferential component when the defect depth and the lifting distance change respectively, wherein one defect width is a group and is divided into 5 groups, the defect widths are 6, 8, 10, 12 and 14mm respectively, and obtaining the peak-to-valley median spacing S of the circumferential component y-50％ And corresponding defect depth and lifting distance, generating a line graph from the obtained data, and verifying the median space S between the peaks and valleys of the circumferential component by calculating the fluctuation percentage of each line graph as shown in FIGS. 12 and 13 y-50％ The fluctuation precision errors of the pressure sensor are all within 10 percent.
In summary, the median circumferential component peak-to-valley spacing S is selected y-50％ As a quantization parameter of the defect width W, the peak-to-valley median spacing S of the circumferential components in the sample set y-50％ And performing linear fitting on the defect width W to obtain a defect width quantization formula:
W＝a 2 *S y-50％ +b 2
wherein, a 2 Is the width proportionality coefficient, b 2 Is a width correction factor; a is 2 、b 2 The device is a fixed value under general conditions, and can be adjusted according to specific conditions such as different working environments, pipes, excitation intensity, detector attributes and the like.
(3) Quantitative analysis of defect depth
According to simulation result analysis, the radial component peak-valley value and the axial component peak-valley value are in good direct proportional linear relation with the defect depth, increase along with the increase of the defect depth, and are suitable for being used as signal characteristic quantity for quantifying the defect depth. For example, the radial components of different defect depths are obtained by finite element simulation, as shown in fig. 14, and the peak-to-valley values B of the radial components of different depths are known from the simulation results zp-p Proportional to the defect width, as shown in fig. 15.
Although the two signal feature quantities, i.e., the radial component peak-to-valley value and the axial component peak-to-valley value, are suitable as signal feature quantities for quantifying the depth of a defect, they are still unstable signal feature quantities due to the influence of the change in the size of the defect and the fluctuation in the lift-off value. However, according to the above quantitative analysis of the defect length and width, the defect length can be determined by the peak-to-valley distance of the axial component differential signal, the defect width size can be determined by the median distance between the peaks and valleys of the circumferential component, and the defect depth can be quantified by taking the peak-to-valley value of the radial component as a quantification parameter.
The radial component peak-valley value and the corresponding defect depth are subjected to linear fitting, the linear fitting effect of the radial component peak-valley value and the defect depth under defect samples of different sizes is shown in fig. 16, the fitting straight line can represent the defect depth in a corresponding range, and the following relational expression is satisfied:
D＝a*B zp-p +b
wherein, B zp-p And a and b are defect depth fitting parameters, namely the difference between the peak and the trough of the radial component, and are variables, when the length and the width of the defect change, the corresponding a and b also change, and the a and b respectively represent a proportionality coefficient and a correction coefficient of the defect depth.
Therefore, the mapping relation between the length and the width of the defect and the defect depth fitting parameters a and b can be obtained by establishing a BP neural network model.
In this embodiment, 200 sets of defects with different lengths and widths are designed, the defect depths are respectively 4mm, 6mm, 8mm, 10mm and 12mm, the three are combined to total 1000 defects, and the radial component peak-to-valley value B of the 200 sets of defects with different lengths and widths is calculated zp-p Fitting with the defect depth D to obtain 200 groups of corresponding defect depth fitting parameters a and b, namely obtaining 200 groups of neural network sample data, and constructing a neural network sample set.
And training sample data by adopting a GA (genetic algorithm) depth optimization BP (back propagation) neural network method by taking the length and the width of the known defect as the input of the neural network and the defect depth fitting parameters a and b as the output. Fitting parameters a and B of the predicted defect depth with a radial component peak-valley value B zp-p And substituting the formula to obtain a predicted value of the defect depth D. Will be short ofAnd comparing the predicted value of the depth of the defect with the actual value of the depth of the defect which is initially set, thereby judging the accuracy rate of the prediction of the depth of the defect in the method.
In this embodiment, the neural network sample set is divided into a training set, a test set, and a verification set, where the number of data in the training set is 140, the number of data in the test set is 30, and the number of data in the verification set is 30. The number of nodes of an input layer of the BP neural network is 2, the number of nodes of a hidden layer is automatically optimized, and the number of nodes of an output layer is 2. The comparison result and the correlation coefficient between the predicted defect depth value and the actual defect depth value in the training set, the test set, the verification set and the total neural network sample set are shown in fig. 17, where R is the correlation coefficient between the predicted defect depth value and the actual defect depth value in the set, and is used for characterizing the prediction accuracy of the defect depth in the set, and the closer R is to 1, the closer the prediction accuracy of the defect depth is to 100%, where R is 0.99801 in the training set, R is 0.99448 in the test set, R is 0.97216 in the verification set, and R is 0.99312 in the total neural network sample set. The ratio of the predicted defect depth value to the actual defect depth value of part of the sample data is shown in fig. 18. As can be seen from fig. 18, the indirect quantification of the defect depth is achieved by predicting the defect depth fitting parameters a and b, the effect is very significant, and the prediction accuracy is higher than 97%, which is higher than that of a method for directly quantifying the defect depth by using the radial component peak-valley value.
In summary, the radial component peak-to-valley B is selected zp-p And obtaining a defect depth quantization formula as a quantization parameter of the defect depth D:
D＝a*B zp-p +b
and the defect depth fitting parameters a and b are obtained by predicting the length and width of the defect after quantization through a neural network model.
The method simulates pipeline magnetic flux leakage signals under different defect sizes through finite element modeling, researches and analyzes the mutual influence relation between several signal characteristic quantities and the defect length, width and depth, can obtain the representation effectiveness of the defect length, width and depth and each signal characteristic quantity, and specifically analyzes the superiority and inferiority of the signal characteristic quantity which can represent the geometrical parameters of the defect, and obtains an optimal representation result.
The method provided by the invention has the advantages that the sample data acquired under the conditions of different working scenes, different equipment, different materials, different magnetization conditions and the like are different, but the integral quantitative prediction method is suitable for all occasions.
The present invention is not limited to the above embodiments, and any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (4)
1. A pipeline defect quantification method based on leakage magnetic signal characteristics is characterized by comprising the following steps:
s1, obtaining triaxial magnetic flux leakage signals of the defects under different defect sizes, and constructing a sample set;
the defect sizes include: defect length, defect width, defect depth;
the three-axis magnetic flux leakage signal of the defect refers to: the axial magnetic leakage signal is an axial component, the radial magnetic leakage signal is a radial component, and the circumferential magnetic leakage signal is a circumferential component;
the axial direction is the direction along the length of the pipeline, the radial direction is the direction vertical to the inner wall of the pipeline, and the circumferential direction is the circumferential direction along the pipeline;
the selection mode of the circumferential component is as follows: determining the crest position of the radial component on the axial path, and selecting the value of the radial component on the circumferential path where the crest position is as the value of the circumferential component;
s2, differentiating the axial component in the sample set to obtain an axial component differential signal, and extracting the peak-to-valley interval DS of the axial component differential signal xp-p The peak-to-valley spacing DS of the differential signal of the axial component in the sample set as a quantization parameter for the defect length L xp-p Linear fitting is carried out on the defect length L to obtain the defect length quantityThe formula is shown as follows:
L＝a 1 *DS xp-p +b 1
wherein, a 1 Is a length scale factor, b 1 Is a length correction factor;
s3, extracting the median space S between the peaks and valleys of the circumferential components in the sample set y-50％ As a quantization parameter of the defect width W, the peak-to-valley median spacing S of the circumferential components in the sample set y-50％ And performing linear fitting on the defect width W to obtain a defect width quantization formula:
W＝a 2 *S y-50％ +b 2
wherein, a 2 Is a width proportionality coefficient, b 2 Is the width correction factor;
median circumferential component peak-to-valley spacing S y-50％ The extraction method comprises the following steps: extracting the wave crest and the wave trough of the circumferential component, calculating the middle value of the wave crest and the wave trough, namely 50% of the difference between the wave crest and the wave trough as the median of the wave crest and the wave trough, and the distance between the median of the wave crest and the wave trough on the circumferential component is the distance S between the median of the wave crest and the wave trough on the circumferential component y-50％ ；
S4, extracting the peak-valley value B of the radial component in the sample set zp-p I.e. the difference between the peak and the trough of the radial component, the peak-to-trough value B of the radial component zp-p As a quantization parameter of the defect depth D, the radial component peak-to-valley value B is concentrated on the sample zp-p Fitting with the defect depth D to obtain a defect depth quantization formula:
D＝a*B zp-p +b
wherein a and b are both defect depth fitting parameters and are both variables, and when the defect length or defect width changes, the corresponding defect depth fitting parameters a and b also change;
s5, constructing a neural network, inputting the neural network into the defect length L and the defect width W, outputting the parameters a and b of defect depth fitting, training by using a sample set and generating the neural network;
s6, the process of quantifying the unknown defects is as follows:
s61, differentiating the signal peak-to-valley spacing DS according to the axial component of the unknown defect xp-p And defect lengthA quantization formula is used for calculating the defect length L of the unknown defect;
s62, according to the mean distance S between the peaks and the valleys of the circumferential component of the unknown defect y-50％ And a defect width measurement formula, calculating to obtain the defect width W of the unknown defect;
s63, inputting the calculated length L and width W of the unknown defect into a neural network model, and predicting to obtain defect depth fitting parameters a and b;
s64, determining the radial component peak-to-valley value B of the unknown defect zp-p And substituting the predicted defect depth fitting parameters a and b into a defect depth quantization formula to calculate the defect depth D of the unknown defect.
2. The method for quantifying the pipeline defect based on the leakage magnetic signal characteristics of claim 1, wherein in the step S1, the triaxial leakage magnetic signals of the defect under different defect sizes are simulated through finite element software modeling.
3. The method according to claim 1, wherein in step S5, the sample set is divided into a training set and a testing set, and the training set is used to train the neural network:
for the radial component peak-valley value B with the same defect length and defect width in the training set zp-p Fitting with the defect depth D to obtain defect depth fitting parameters a and b under the defect length and the defect width so as to obtain training data of the neural network, and training the neural network by using the training data of the neural network.
4. The method for quantifying the pipeline defect based on the leakage magnetic signal characteristics according to claim 1 or 3, wherein in step S5, the sample set is divided into a training set and a test set, and the test set is used to test the neural network:
inputting the defect length L and the defect width W of the known defects in the test set into the trained neural network, and predicting to obtain corresponding defect depth fitting parameters a,b; fitting corresponding defect depth to parameters a and B and radial component peak-valley value B of the defect zp-p Substituting the defect depth into a defect depth quantization formula, and calculating to obtain a predicted value of the defect depth D; and comparing the predicted value of the defect depth D with the real value of the defect depth D to test the prediction accuracy of the neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210729456.6A CN115096987B (en) | 2022-06-24 | 2022-06-24 | Pipeline defect quantification method based on magnetic leakage signal characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210729456.6A CN115096987B (en) | 2022-06-24 | 2022-06-24 | Pipeline defect quantification method based on magnetic leakage signal characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115096987A true CN115096987A (en) | 2022-09-23 |
| CN115096987B CN115096987B (en) | 2024-04-26 |
Family
ID=83292192
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210729456.6A Active CN115096987B (en) | 2022-06-24 | 2022-06-24 | Pipeline defect quantification method based on magnetic leakage signal characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115096987B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104514987A (en) * | 2014-12-19 | 2015-04-15 | 清华大学 | Three-dimensional pipeline flux leakage imaging defect quantizing method |
| JP2017009549A (en) * | 2015-06-26 | 2017-01-12 | コニカミノルタ株式会社 | Non destructive testing device |
| CN106770627A (en) * | 2016-12-16 | 2017-05-31 | 北京华航无线电测量研究所 | A kind of axial magnetic leakage signal length quantization method |
| CN106870957A (en) * | 2017-03-21 | 2017-06-20 | 东北大学 | A kind of feature extracting method of pipeline defect and magnetic leakage signal |
| WO2020133639A1 (en) * | 2018-12-29 | 2020-07-02 | 东北大学 | Intelligent analysis system for magnetic flux leakage detection data in pipeline |
| CN113049676A (en) * | 2021-03-09 | 2021-06-29 | 中国矿业大学 | Quantitative analysis method for pipeline defects |
-
2022
- 2022-06-24 CN CN202210729456.6A patent/CN115096987B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104514987A (en) * | 2014-12-19 | 2015-04-15 | 清华大学 | Three-dimensional pipeline flux leakage imaging defect quantizing method |
| US20160178580A1 (en) * | 2014-12-19 | 2016-06-23 | Tsinghua University | Method and apparatus for quantifying pipeline defect based on magnetic flux leakage testing |
| JP2017009549A (en) * | 2015-06-26 | 2017-01-12 | コニカミノルタ株式会社 | Non destructive testing device |
| CN106770627A (en) * | 2016-12-16 | 2017-05-31 | 北京华航无线电测量研究所 | A kind of axial magnetic leakage signal length quantization method |
| CN106870957A (en) * | 2017-03-21 | 2017-06-20 | 东北大学 | A kind of feature extracting method of pipeline defect and magnetic leakage signal |
| WO2020133639A1 (en) * | 2018-12-29 | 2020-07-02 | 东北大学 | Intelligent analysis system for magnetic flux leakage detection data in pipeline |
| CN113049676A (en) * | 2021-03-09 | 2021-06-29 | 中国矿业大学 | Quantitative analysis method for pipeline defects |
Non-Patent Citations (3)
| Title |
|---|
| ZHANG GUOGUANG: "Defect quantitative recognition technology of circumferential magnetic flux leakage inspection in pipeline", 《2010 INTERNATIONAL CONFERENCE ON COMPUTER APPLICATION AND SYSTEM MODELING (ICCASM 2010)》, 4 November 2010 (2010-11-04) * |
| 杨涛, 王太勇, 李清, 冷永刚: "油气管道缺陷漏磁检测试验", 天津大学学报, no. 08, 25 August 2004 (2004-08-25) * |
| 田野;高涛;许光达;丁融;: "长输管道漏磁内检测缺陷识别量化技术研究", 油气田地面工程, no. 10, 20 October 2018 (2018-10-20) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115096987B (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ulapane et al. | Non-destructive evaluation of ferromagnetic material thickness using pulsed eddy current sensor detector coil voltage decay rate | |
| Kandroodi et al. | Defect detection and width estimation in natural gas pipelines using MFL signals | |
| Kandroodi et al. | Detection of natural gas pipeline defects using magnetic flux leakage measurements | |
| CN107356677B (en) | Ultrasonic nondestructive testing method based on travel time tomography and reverse time migration imaging | |
| US12018939B2 (en) | System and method for analyzing an oilfield casing using an apparent thickness approach | |
| CN112731377A (en) | Dielectric constant inversion method, roadbed detection method and detection system | |
| CN114487976B (en) | Method and system for evaluating traceability uncertainty of MCM electronic transformer calibrator | |
| CN113049676A (en) | Quantitative analysis method for pipeline defects | |
| Chen et al. | Three-dimensional defect reconstruction from magnetic flux leakage signals in pipeline inspection based on a dynamic taboo search procedure | |
| Zhang et al. | Analysis of MFL model for sucker rod defects and its MFL signal processing | |
| Chady et al. | Neural network models of eddy current multi-frequency system for nondestructive testing | |
| CN111579647B (en) | Concrete member corrosion degree detection method and system based on analytic hierarchy process | |
| CN115096987B (en) | Pipeline defect quantification method based on magnetic leakage signal characteristics | |
| CN106761666B (en) | Method and device for four-probe scattering gamma logging and nonlinear data inversion | |
| Peng et al. | A lift-off revision method for magnetic flux leakage measurement signal | |
| CN112131781A (en) | Steel structure damage detection method based on full-connection neural network and transfer rate function | |
| CN107178710B (en) | It is a kind of based on it is inside and outside detection signal characteristic abstraction defect of pipeline inside and outside discrimination method | |
| CN113075289B (en) | Metal cylinder defect parameter detection method and system | |
| CN115371536A (en) | Metal pipeline crack measuring method and system | |
| US10487643B2 (en) | Two-dimensional imaging with multi-stage processing | |
| Ji et al. | LS-SVMs-based reconstruction of 3-D defect profile from magnetic flux leakage signals | |
| CN111197471B (en) | Transient electromagnetic detection calculation model and detection method for underground screen pipe | |
| Jiang | Study of underground oil-gas pipeline corrosion pits estimation based on MFL inspection method | |
| CN116754632B (en) | Oil gas pipeline crack quantification method and device based on orthogonal twin and storage medium | |
| Zhou et al. | Structural optimisation of a standard alternating current field measurement excitation probe for concave and convex surfaces |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |