CN115047064A - Pipeline defect quantification method, processor and pipeline defect quantification device - Google Patents
Pipeline defect quantification method, processor and pipeline defect quantification device Download PDFInfo
- Publication number
- CN115047064A CN115047064A CN202210638895.6A CN202210638895A CN115047064A CN 115047064 A CN115047064 A CN 115047064A CN 202210638895 A CN202210638895 A CN 202210638895A CN 115047064 A CN115047064 A CN 115047064A
- Authority
- CN
- China
- Prior art keywords
- defect
- pipeline
- magnetic flux
- principal component
- flux leakage
- 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.)
- Pending
Links
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
- G01N27/85—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 using magnetographic methods
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
- F17D5/02—Preventing, monitoring, or locating loss
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Chemical & Material Sciences (AREA)
- Mechanical Engineering (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)
Abstract
The embodiment of the invention provides a pipeline defect quantification method, a processor and a pipeline defect quantification device, wherein the pipeline defect quantification method comprises the following steps: acquiring a pipeline magnetic flux leakage detection image; extracting characteristic information of the pipeline magnetic flux leakage detection image to construct a defect characteristic database of the pipeline; performing dimension reduction operation on the defect features in the defect feature database to determine principal component defect features; constructing a pipeline defect quantification model; acquiring an actual value of the principal component defect characteristic; the actual value of the main component defect characteristic is input into the pipeline defect quantization model to determine the predicted value of the defect characteristic dimension of the pipeline, the predicted value of the defect characteristic dimension of the pipeline can be rapidly and accurately obtained, and the practicability is high.
Description
Technical Field
The invention relates to the technical field of pipeline defect quantification, in particular to a pipeline defect quantification method, a processor and a pipeline defect quantification device.
Background
With the development of economy, the worldwide energy demand is rapidly increased every year, most of the energy exists in the form of oil and gas, and the main transportation mode is pipeline transportation. Because the pipeline passes through a wide area, is complex in service and hidden in position, pipeline leakage caused by corrosion, abrasion, accidental damage and the like sometimes occurs. Once the energy is not found and prevented in time, not only energy waste and economic loss are caused, but also the environment is seriously polluted, and even the personal safety is endangered. Therefore, in order to ensure the safe operation of the oil and gas transportation pipeline and provide accurate prediction for pipeline repair, the pipeline needs to be periodically detected, and research is carried out on the defect condition of the pipeline. Therefore, the research on the detection technology in the pipeline is developed, and the accurate quantitative analysis on the defects of the pipeline becomes the key point of the current in-service pipeline safety research.
The research of the pipeline detection technology relates to multiple disciplines and fields, and the research content is complex. Among many pipeline detection technologies, the magnetic flux leakage detection technology is one of the most widely used and most mature nondestructive detection technologies. The pipeline defect identification based on the magnetic flux leakage detection data is mainly carried out manually, a large number of human factors exist in the result, the existence and the position of the pipeline defect can be basically judged, and accurate quantitative information such as the shape, the size and the like of the defect cannot be provided. With the improvement of the precision of the pipeline magnetic flux leakage detection equipment and the extension of the detection distance, data up to dozens of GB can be generated when detecting a pipeline of 100km, so that the manual analysis of a large amount of detection data consumes a lot of time, and an intelligent pipeline defect quantitative identification technology is urgently needed.
Disclosure of Invention
The invention aims to provide a pipeline defect quantification method, a processor and a pipeline defect quantification device, which can accurately and quickly obtain the defect characteristic size of a pipeline.
In order to achieve the above object, an embodiment of the present invention provides a method for quantifying pipeline defects, where the method includes:
acquiring a pipeline magnetic flux leakage detection image;
extracting characteristic information of the pipeline magnetic flux leakage detection image to construct a defect characteristic database of the pipeline;
performing dimension reduction operation on defect feature data in a defect feature database to determine principal component defect features;
constructing a pipeline defect quantification model;
acquiring an actual value of the principal component defect characteristics;
and inputting the actual value of the main component defect characteristic into a pipeline defect quantization model to determine a predicted value of the defect characteristic size of the pipeline.
In an embodiment of the invention, the defect features in the defect feature database include at least axial, radial, and circumferential triaxial signals such as valley-to-valley spacing, peak-to-peak spacing, peak-to-valley spacing, inflection point spacing, special point spacing, peak-to-valley difference, volume, surface area, surface energy, and the like.
In an embodiment of the present invention, the defect feature sizes include a defect length, a defect circumferential width, and a defect depth.
In an embodiment of the present invention, acquiring the pipeline magnetic flux leakage detection image includes:
acquiring pipeline magnetic flux leakage data;
preprocessing the pipeline magnetic flux leakage data to obtain the preprocessed pipeline magnetic flux leakage data;
and drawing a curve according to the preprocessed pipeline magnetic flux leakage data to obtain a pipeline magnetic flux leakage detection image.
In an embodiment of the present invention, performing a dimension reduction operation on the defect feature data in the defect feature database to determine the principal component defect feature comprises:
normalizing the defect feature data in the defect feature database to determine the defect feature data after normalization processing;
and performing dimensionality reduction operation on the defect characteristic data after the normalization processing based on a principal component analysis algorithm to determine principal component defect characteristics.
In an embodiment of the present invention, constructing a pipeline defect quantification model includes:
acquiring a radial basis function neural network;
acquiring an actual value of the feature size of the defect;
constructing a pipeline defect quantitative training set according to the actual value of the defect feature size and the actual value of the principal component defect feature;
and inputting the pipeline defect quantitative training set into a radial basis function neural network for training so as to construct a pipeline defect quantitative model.
In the embodiment of the invention, inputting the pipeline defect quantification training set into the radial basis function neural network for training to construct the pipeline defect quantification model comprises:
inputting the pipeline defect quantitative training set into a radial basis function neural network for training so as to construct a pipeline defect quantitative test network;
constructing a pipeline defect quantitative test set according to the actual value of the defect feature size and the actual value of the principal component defect feature;
and inputting the pipeline defect quantitative test set into a pipeline defect quantitative test network for testing to construct a pipeline defect quantitative model.
A second aspect of the invention provides a processor configured to perform the above-described method of quantifying pipe defects.
A third aspect of the present invention provides a pipeline defect quantifying apparatus, which includes the processor described above.
In an embodiment of the present invention, the apparatus for quantifying pipeline defects further includes:
and the magnetic flux leakage detection equipment is used for acquiring magnetic flux leakage data of the pipeline.
According to the technical scheme, the characteristic information is extracted from the obtained pipeline magnetic flux leakage detection image to construct a defect characteristic database of the pipeline, the dimension reduction operation is performed on the defect characteristics in the defect characteristic database to determine the principal component defect characteristics, then the pipeline defect quantitative model inputs the actual values of the principal component defect characteristics into the pipeline defect quantitative model to determine the predicted value of the defect characteristic dimension of the pipeline.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention and not to limit the embodiments of the invention. In the drawings:
FIG. 1 is a flow chart illustrating a method for quantifying pipeline defects according to an embodiment of the present invention;
FIG. 2 is a graph of a signal of axial leakage flux of a defect according to an embodiment of the present invention;
FIG. 3 is a graph of a signal of radial leakage flux of a defect in an embodiment of the present invention;
FIG. 4 is a graph of a signal of circumferential leakage flux of a defect in an embodiment of the present invention;
FIG. 5 is a graph showing principal component analysis in the embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
An embodiment of the present invention provides a method for quantifying a pipeline defect, as shown in fig. 1, the method for quantifying a pipeline defect includes the following steps:
step S101: and acquiring a pipeline magnetic flux leakage detection image.
Further, in this embodiment, step S101: the step of obtaining the pipeline magnetic leakage detection image comprises a step S201 and a step S203, wherein:
step S201: and acquiring pipeline magnetic flux leakage data.
Specifically, the method for quantifying the pipeline defect in the embodiment is suitable for a device for quantifying the pipeline defect, and the device for quantifying the pipeline defect comprises a processor and a magnetic leakage detection device which are connected through signals, wherein the magnetic leakage detection device performs magnetic leakage detection on a pipeline and collects magnetic leakage data of the pipeline, and the magnetic leakage detection device transmits the magnetic leakage data of the pipeline to the processor after the detection is completed.
Step S202: and preprocessing the pipeline magnetic flux leakage data to obtain the preprocessed pipeline magnetic flux leakage data.
Specifically, the leakage flux detection device in this embodiment includes a plurality of parallel leakage flux detection sensors, the plurality of parallel leakage flux detection sensors are movably disposed in the pipeline cavity, and due to a baseline difference between the respective leakage flux detection sensors, the pipeline leakage flux data detected by the respective leakage flux detection sensors under the same condition are inconsistent, so that the initially obtained pipeline leakage flux data needs to be preprocessed by methods such as comparing data mean values, and differences between the respective leakage flux detection sensors are eliminated, so that subsequent steps can be performed normally. The pipeline magnetic leakage data are preprocessed, and operations such as deviation correction and data filling are performed, so that the correctness and the integrity of the pipeline magnetic leakage data are improved, and the accuracy of the pipeline defect quantification method is further improved.
Step S203: and drawing a curve according to the preprocessed pipeline magnetic flux leakage data to obtain a pipeline magnetic flux leakage detection image.
Specifically, after the preprocessing is completed, the processor performs curve drawing on the pipeline magnetic leakage data in a dot-drawing manner (as shown in fig. 2 to 4) to draw a pipeline magnetic leakage detection image, where the types of the curves in the pipeline magnetic leakage detection image in the embodiment include a defect maximum channel curve, a defect peak-valley value curve, a defect axial center signal curve, a defect maximum channel first-order differential curve, and the like.
Step S102: and (4) extracting characteristic information of the pipeline magnetic flux leakage detection image to construct a defect characteristic database of the pipeline.
Specifically, after the pipeline magnetic flux leakage detection image is obtained, the processor performs feature information extraction operation on the pipeline magnetic flux leakage detection image, that is, preliminarily extracts the defect features of the pipeline from a defect maximum channel curve, a defect peak-valley value curve, a defect axial center signal curve, a defect maximum channel first-order differential curve and the like. Further, the defect characteristics of the pipeline in the embodiment at least include axial, radial and circumferential triaxial signals such as valley-to-valley distance, peak-to-peak distance, peak-to-valley distance, inflection point distance, special point distance, peak-to-valley difference, volume, surface area, surface energy and the like. After the defect characteristics are obtained, a defect characteristic database of the pipeline is constructed based on the defect characteristics, wherein the defect characteristic database comprises the defect characteristics and the characteristic value data corresponding to the defect characteristics.
Step S103: and performing dimensionality reduction operation on the defect features in the defect feature database to determine principal component defect features.
Further, step S103: performing a dimension reduction operation on the defect features in the defect feature database to determine principal component defect features further includes steps S301 to S302, where:
step S301: and normalizing the defect feature data in the defect feature database to determine the normalized defect feature data.
Different defect features may have different dimensions and dimension units, so as to avoid affecting the accuracy of the result of the subsequent step and eliminate the dimension influence of different defect features, the processor in the embodiment performs normalization processing on the defect features in the defect feature database through python software, and each defect feature after the normalization processing is in the same order of magnitude, thereby eliminating the adverse effect caused by singular sample data.
Step S302: and performing dimensionality reduction operation on the defect characteristic data after the normalization processing based on a principal component analysis algorithm to determine principal component defect characteristics.
Principal Component Analysis (PCA) is a method of converting a set of variables that may have correlation into a set of linearly uncorrelated variables by orthogonal transformation, and the converted set of variables is called Principal components.
After obtaining the defect features after the normalization processing, the processor judges the correlation among the defect features by adopting a principal component analysis algorithm, and specifically, preliminarily determines the principal component defect features based on the following formula:
F p ＝a 1i ×Z x1 +a 2i ×Z x2 +...+a pi ×Z xp (2)
wherein, F p For the p-th linear combination, a 1i ，a 2i ，…，a pi (i is 1, … …, m) is a feature vector corresponding to the feature value of the covariance matrix Σ of the defect feature X, Z x1 ，Z x2 ，……，Z xp Is the defect characteristic data after normalization processing.
As shown in fig. 5, followed by F p Reasonably selecting the first m principal components from the determined p principal components by taking the contribution rate of the medium defect characteristic data as a selection standard; if the cumulative contribution rate reaches a certain value (e.g. 80% -90%, preferably 85% in this embodiment), it can be ensured that the selected principal components contain most of the information of the original defect features, and the m selected principal components are the principal component defect features (m is 7 in the embodiment shown in fig. 5).
Step S104: and constructing a pipeline defect quantification model.
Further, in the present embodiment, step S104: the construction of the pipeline defect quantification model further comprises steps S401 to S404, wherein:
step S401: and acquiring the radial basis function neural network.
On the Basis of a Radial Basis Function Neural Network (RBFNN), a momentum term and a self-adaptive learning mechanism are adopted and applied to quantitative analysis of the defects of the long-distance pipeline. The radial basis function has various forms, generally taking a Gauss function, and then the input/output relationship of the radial basis function neural network is expressed as follows:
wherein: x represents input, Y ═ Y 1 ,y 2 ,...,y m ) T Representing the output, H represents the number of basis functions used, | is the Euclidean norm, c j Representing the center of the basis function, σ j Is the radius of the basis function, ω ij Then the connection weight between the jth basis function and the ith output point.
The establishment of the radial basis function neural network comprises two steps:
first, selection of the center of the basis function: generally, the radius sigma of the basis function can be calculated after the clustering analysis of the samples is completed according to experience or a clustering algorithm or by adopting a K-order mean algorithm j I.e. the average between the center of the basis function and the training samples belonging to the cluster.
Secondly, training is to determine the connection weight from the hidden layer to the output layer, and the embodiment adjusts the parameters of the radial basis function neural network by using a gradient descent method with an adaptive mechanism. The error function is defined as follows:
wherein: j (k) is an error function, t (k) represents the desired output, and y (W, k) represents the actual output of the network.
Obtaining a self-adaptive adjustment algorithm of a weight matrix connecting the hidden layer to the output layer as shown in the following formula:
wherein μ (k) is a first learning rate; alpha (k) is a momentum term and has the function of stabilizing the change of the radial basis function neural network so as to facilitate the convergence of the radial basis function neural network; w (k +1) is the connection weight from the hidden layer to the output layer in the k +1 th iteration; w (k) is the connection weight from the hidden layer to the output layer in the kth iteration; w (k-1) is the connection weight from the hidden layer to the output layer in the k-1 iteration;
is the partial derivative of the error function J (W) with respect to W.
The adjustment algorithm of the hidden layer center value matrix is shown as the following formula:
wherein, C (k +1) is the central value of the hidden layer in the (k +1) th iteration; c (k) is the center value of the hidden layer at the kth iteration; c (k-1) is the central value of the hidden layer in the k-1 iteration;
is the partial derivative of the error function J (C) with respect to C.
The adjustment algorithm of the radius matrix of the hidden layer is shown as the following formula:
wherein σ (k +1) is a radius matrix of the hidden layer in the (k +1) th iteration; σ (k) is a radius matrix of the hidden layer at the kth iteration; sigma (k-1) is a radius matrix of the hidden layer at the k-1 iteration;
is the partial derivative of the error function J (σ) with respect to σ.
Specifically, the processor of the embodiment is pre-stored with a radial basis function neural network, and the radial basis function neural network is called out when needed; the processor is also pre-stored with measured pipeline body measurement data, the pipeline body measurement data comprises a pipeline inner diameter, a pipeline wall thickness, a sensor lift-off value and the like, and the pipeline body measurement data is input into the radial basis function neural network, so that the pertinence of the radial basis function neural network to the pipeline with the characteristic size of the solved defect is further improved, and the accuracy of solving is improved.
Step S402: and acquiring an actual value of the feature size of the defect.
The actual value of the defect feature size can be measured in advance, stored in a processor in advance and retrieved when needed.
Step S403: and constructing a pipeline defect quantification training set according to the actual value of the defect feature size and the actual value of the principal component defect feature.
And (3) according to the actual value of the defect characteristic size and the actual value of the principal component defect characteristic as 7: and 3, dividing according to the proportion, wherein a set consisting of the actual values of the defect feature sizes with a large number and the actual values of the principal component defect features after division is a pipeline defect quantification training set.
Step S404: and inputting the pipeline defect quantitative training set into a radial basis function neural network for training so as to construct a pipeline defect quantitative model.
Further, step S404: inputting the pipeline defect quantification training set into a radial basis function neural network for training, so as to construct a pipeline defect quantification model, and further comprising steps S501-S503, wherein:
step S501: inputting the pipeline defect quantitative training set into a radial basis function neural network for training so as to construct a pipeline defect quantitative test network;
after a pipeline defect quantitative training set is obtained, the pipeline defect quantitative training set is input into a radial basis function neural network for training, so that the error value predicted by the radial basis function neural network is lower than a preset error value, and an inverse process is solved according to an iterative method by combining the trained radial basis function neural network so as to construct a pipeline defect quantitative test network.
Specifically, a defect characteristic is randomly set, in this embodiment, a defect circumferential width is taken as an example, main component defect characteristics corresponding to the defect circumferential width are a defect leakage magnetic field circumferential waveform width Wp-p and a circumferential waveform peak-valley value V, actual values of the defect circumferential width, the defect leakage magnetic field circumferential waveform width Wp-p and the circumferential waveform peak-valley value V are also measured in advance and stored in a processor in advance, and x (0) is an actual value of the defect circumferential width; inputting the actual value of the circumferential width of the defect, the actual value of the circumferential waveform width of the defect leakage magnetic field and the actual value of the peak-valley value of the circumferential waveform as a training set into a radial basis function neural network for training so as to calculate a calculated value of the circumferential waveform width of the defect leakage magnetic field and a calculated value of the peak-valley value of the circumferential waveform; after a calculated value of the circumferential waveform width of the leakage magnetic field of the defect and a calculated value of the peak-valley value of the circumferential waveform are obtained, solving in an inverse process, and in the k-th cycle of training, inputting x (k) into a radial basis function neural network which is subjected to certain training to obtain a predicted output value as follows:
y(k)＝F(x(k)) (8)
wherein x (k) is an input in the kth cycle, which is an iteration value of the kth circumferential width of the defect in this embodiment; y (k) is an output in the k-th cycle, which is a calculated value of the circumferential waveform width and a calculated value of the circumferential waveform peak-valley value of the leakage magnetic field of the defect at the k-th cycle in the embodiment.
After obtaining the calculated values of the principal component defect features, calculating a sum of squared errors E (k) between the actual values of the principal component defect features and the calculated values of the principal component defect features, and then iterating based on the following equation:
where η is the second learning rate of the iterative process,
is the gradient value of E (k) for x (k).
If the calculated sum of squared errors E (k) is greater than or equal to the preset sum of squared errors, continuing the iteration; if the calculated error square sum e (k) is smaller than the preset error square sum, stopping iteration, and exiting the loop, where the obtained network is the pipeline defect quantitative test network, and x (k) of the last iteration before stopping iteration is a predicted value of the feature size of the defect (in this embodiment, a predicted value of the circumferential width of the defect) generated by the network. Further, table 1 shows calculated values and actual value comparison results of the circumferential waveform width Wp-p and the circumferential waveform peak-to-valley value V of the leakage magnetic field of the defect in the present embodiment; table 2 shows the comparison result of the predicted value and the actual value of the circumferential width of the defect in the solution of the inverse problem in this embodiment.
TABLE 1 calculation value and actual value comparison result of circumferential waveform peak-to-valley value V and circumferential waveform width Wp-p of defect leakage magnetic field
TABLE 2 predicted value and actual value comparison result of defect circumferential width in solving inverse problem
Step S502: and constructing a pipeline defect quantitative test set according to the actual value of the defect feature size and the actual value of the principal component defect feature.
And (3) according to the actual value of the defect characteristic size and the actual value of the principal component defect characteristic as 7: and 3, dividing according to the proportion, wherein a set consisting of the actual values of the defect feature sizes with small quantity and the actual values of the principal component defect features after division is a pipeline defect quantification training set.
Step S503: inputting the pipeline defect quantitative test set into a pipeline defect quantitative test network for testing so as to construct a pipeline defect quantitative model.
In the testing process, whether the predicted value of the characteristic dimension of the defect is accurate or not can be judged according to the accuracy of the predicted value of the characteristic dimension of the defect generated by the pipeline defect quantitative testing network, and if the accuracy of the predicted value of the characteristic dimension of the defect generated by the pipeline defect quantitative testing network is always within the preset accuracy range, the pipeline defect quantitative testing network is determined to be a pipeline defect quantitative model; if the accuracy of the predicted value of the feature size of the defect generated by the pipeline defect quantitative test network is not always within the preset accuracy range, the method should return to step S403 to train the radial basis function neural network again and obtain the pipeline defect quantitative test network until the accuracy of the predicted value of the feature size of the defect generated by the pipeline defect quantitative test network is always within the preset accuracy range.
Step S105: and acquiring an actual value of the principal component defect characteristic.
The actual values of the constituent defect features are measured in advance and stored in a processor.
Step S106: and inputting the actual value of the principal component defect characteristic into a pipeline defect quantization model to determine a predicted value of the defect characteristic dimension of the pipeline.
In the practical application process of the pipeline defect quantification model, the actual value of the principal component defect characteristic corresponding to the defect characteristic to be solved is input into the pipeline defect quantification model, so that the defect characteristic size of the pipeline can be predicted, and the predicted value of the defect characteristic size is obtained.
In one embodiment of the invention, the defect feature sizes include a defect length, a defect circumferential width, and a defect depth.
Another embodiment of the present invention provides a processor configured to perform the pipeline defect quantifying method of the above-described embodiment.
Another embodiment of the present invention provides a pipeline defect quantifying apparatus, which includes the processor of the above embodiment.
In another embodiment of the present invention, the apparatus for quantifying the defect of the pipeline further comprises a leakage flux detecting device connected to the processor for collecting leakage flux data of the pipeline.
The embodiment of the invention provides a pipeline defect quantification method, a processor and a pipeline defect quantification device, wherein the method comprises the steps of extracting characteristic information from an obtained pipeline magnetic flux leakage detection image to construct a defect characteristic database of a pipeline, then performing dimension reduction operation on defect characteristics in the defect characteristic database to determine principal component defect characteristics, and then determining a predicted value of a defect characteristic dimension of the pipeline by using a pipeline defect quantification model and inputting an actual value of the principal component defect characteristics into the pipeline defect quantification model.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art to which the present application pertains. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method for quantifying pipeline defects is characterized by comprising the following steps:
acquiring a pipeline magnetic flux leakage detection image;
extracting characteristic information of the pipeline magnetic flux leakage detection image to construct a defect characteristic database of the pipeline;
performing dimension reduction operation on the defect feature data in the defect feature database to determine principal component defect features;
constructing a pipeline defect quantification model;
acquiring an actual value of the principal component defect characteristics;
and inputting the actual value of the principal component defect characteristic into the pipeline defect quantification model to determine a predicted value of the defect characteristic dimension of the pipeline.
2. The method of claim 1, wherein the defect features in the database of defect features comprise at least axial, radial, and circumferential triaxial signals including valley-to-valley spacing, peak-to-peak spacing, peak-to-valley spacing, inflection point spacing, special point spacing, peak-to-valley difference, volume, surface area, surface energy, and the like.
3. The method of claim 1, wherein the defect feature sizes comprise a defect length, a defect circumferential width, and a defect depth.
4. The method of claim 1, wherein the obtaining of the flux leakage inspection image of the pipeline comprises:
acquiring pipeline magnetic flux leakage data;
preprocessing the pipeline magnetic flux leakage data to obtain preprocessed pipeline magnetic flux leakage data;
and drawing a curve according to the preprocessed pipeline magnetic flux leakage data to obtain the pipeline magnetic flux leakage detection image.
5. The method of claim 1, wherein the performing a dimensionality reduction operation on the defect feature data in the defect feature database to determine the principal component defect feature comprises:
normalizing the defect feature data in the defect feature database to determine normalized defect feature data;
and performing dimensionality reduction operation on the defect characteristic data after the normalization processing based on the principal component analysis algorithm to determine principal component defect characteristics.
6. The method of claim 1, wherein the constructing the pipeline defect quantification model comprises:
acquiring a radial basis function neural network;
acquiring an actual value of the feature size of the defect;
constructing a pipeline defect quantitative training set according to the actual value of the defect feature size and the actual value of the principal component defect feature;
and inputting the pipeline defect quantitative training set into the radial basis function neural network for training so as to construct the pipeline defect quantitative model.
7. The method of claim 6, wherein the inputting the pipeline defect quantification training set into the radial basis function neural network for training to construct the pipeline defect quantification model comprises:
inputting the pipeline defect quantitative training set into the radial basis function neural network for training so as to construct a pipeline defect quantitative test network;
constructing a pipeline defect quantitative test set according to the actual value of the defect feature size and the actual value of the principal component defect feature;
inputting the pipeline defect quantitative test set into the pipeline defect quantitative test network for testing so as to construct the pipeline defect quantitative model.
8. A processor configured to perform the method of any one of claims 1 to 7.
9. A pipeline defect quantification apparatus, comprising the processor according to claim 8.
10. The apparatus of claim 9, wherein the apparatus further comprises:
and the magnetic flux leakage detection equipment is used for acquiring magnetic flux leakage data of the pipeline.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210638895.6A CN115047064A (en) | 2022-06-07 | 2022-06-07 | Pipeline defect quantification method, processor and pipeline defect quantification device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210638895.6A CN115047064A (en) | 2022-06-07 | 2022-06-07 | Pipeline defect quantification method, processor and pipeline defect quantification device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115047064A true CN115047064A (en) | 2022-09-13 |
Family
ID=83161644
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210638895.6A Pending CN115047064A (en) | 2022-06-07 | 2022-06-07 | Pipeline defect quantification method, processor and pipeline defect quantification device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115047064A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116384253A (en) * | 2023-04-20 | 2023-07-04 | 中国石油大学(北京) | Method and device for establishing and predicting depth prediction model of pipeline magnetic flux leakage detection defect |
| CN116754632A (en) * | 2023-08-16 | 2023-09-15 | 清华大学 | Oil gas pipeline crack quantification method and device based on orthogonal twin and storage medium |
-
2022
- 2022-06-07 CN CN202210638895.6A patent/CN115047064A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116384253A (en) * | 2023-04-20 | 2023-07-04 | 中国石油大学(北京) | Method and device for establishing and predicting depth prediction model of pipeline magnetic flux leakage detection defect |
| CN116384253B (en) * | 2023-04-20 | 2024-04-05 | 中国石油大学(北京) | Method and device for establishing and predicting depth prediction model of pipeline magnetic flux leakage detection defect |
| CN116754632A (en) * | 2023-08-16 | 2023-09-15 | 清华大学 | Oil gas pipeline crack quantification method and device based on orthogonal twin and storage medium |
| CN116754632B (en) * | 2023-08-16 | 2023-11-21 | 清华大学 | Oil gas pipeline crack quantification method and device based on orthogonal twin and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Thöns | On the value of monitoring information for the structural integrity and risk management | |
| CN115047064A (en) | Pipeline defect quantification method, processor and pipeline defect quantification device | |
| CN108052770B (en) | Long-span bridge girder performance early warning method considering time-varying effect | |
| CN112508105B (en) | Fault detection and retrieval method for oil extraction machine | |
| CN100483126C (en) | Defect distinguish based on three-dimensional finite element NN and quantified appraisal method | |
| CN108875178B (en) | Sensor arrangement method for reducing structural modal identification uncertainty | |
| CN112393931A (en) | Detection method, detection device, electronic equipment and computer readable medium | |
| CN114299359B (en) | Method, equipment and storage medium for detecting power transmission line faults | |
| CN115824993B (en) | Method and device for determining water body chemical oxygen demand, computer equipment and medium | |
| CN115758084B (en) | Deep neural network crack quantification method and device and storage medium | |
| Chencho et al. | Development and application of random forest technique for element level structural damage quantification | |
| CN115935241B (en) | Multi-parameter mutually-fused pipe cleaner real-time positioning method and device | |
| CN106569069A (en) | Power transformer fault diagnosis method | |
| Huang et al. | Damage identification of a steel frame based on integration of time series and neural network under varying temperatures | |
| CN112926636A (en) | Method and device for detecting abnormal temperature of traction converter cabinet body | |
| Sudirman et al. | Pulsed Eddy Current signal processing using wavelet scattering and Gaussian process regression for fast and accurate ferromagnetic material thickness measurement | |
| US20230091681A1 (en) | Method for determining the geometry of a defect based on non-destructive measurement methods using direct inversion | |
| CN112949735A (en) | Liquid hazardous chemical substance volatile concentration abnormity discovery method based on outlier data mining | |
| CN111639722A (en) | Transformer fault diagnosis method based on principal component analysis and twin support vector machine | |
| Liangou et al. | Statistical pattern recognition for optimal sensor placement in damage detection applications | |
| CN118211882A (en) | Product quality management system and method based on big data | |
| CN117216844B (en) | Bridge structure damage detection method, system and storage medium | |
| Slimani et al. | Improved ANN for damage identification in laminated composite plate | |
| CN112329590B (en) | Pipeline assembly detection system and detection method | |
| CN117763436A (en) | Training method and device of fault diagnosis model based on small sample |
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 |