CN114676730A - Bent deformed pipe section identification method and device, electronic equipment and storage medium - Google Patents

Bent deformed pipe section identification method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN114676730A
CN114676730A CN202210329101.8A CN202210329101A CN114676730A CN 114676730 A CN114676730 A CN 114676730A CN 202210329101 A CN202210329101 A CN 202210329101A CN 114676730 A CN114676730 A CN 114676730A
Authority
CN
China
Prior art keywords
pipe section
strain
pipe
identified
data
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
Application number
CN202210329101.8A
Other languages
Chinese (zh)
Inventor
刘啸奔
刘燊
刘思佳
郭梦琪
张东
张宏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China University of Petroleum Beijing
Original Assignee
China University of Petroleum Beijing
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China University of Petroleum Beijing filed Critical China University of Petroleum Beijing
Priority to CN202210329101.8A priority Critical patent/CN114676730A/en
Publication of CN114676730A publication Critical patent/CN114676730A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching
    • G06F2218/14—Classification; Matching by matching peak patterns
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
    • G01B21/32—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring the deformation in a solid
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
    • 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
    • G06N3/045—Combinations of networks
    • 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
    • G06N3/047—Probabilistic or stochastic networks
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Signal Processing (AREA)
  • Length Measuring Devices With Unspecified Measuring Means (AREA)

Abstract

The application provides a method and a device for identifying a bent and deformed pipe section, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring first strain data measured by an Inertial Measurement Unit (IMU); obtaining a bending strain curve according to the first strain data; determining a wave crest of which the bending strain value is greater than a preset value in a bending strain curve to obtain a pipe section to be identified, wherein the pipe section to be identified is a pipe section corresponding to a preset interval on two sides of the wave crest in a pipeline; and inputting the second strain data corresponding to the pipe section to be identified into the pipe section type identification model to obtain the pipe section abnormal type of the pipe section to be identified. According to the method, the pipe section to be identified is determined according to the acquired first strain data, the second strain data corresponding to the pipe section to be identified are input into the pipe section type identification model, the pipe section abnormal type of the pipe section to be identified can be obtained, the pipe section abnormal type is not required to be judged manually, time and labor can be saved, the standard is unified, and therefore the judgment efficiency of the pipe section abnormal type is improved.

Description

Bent deformed pipe section identification method and device, electronic equipment and storage medium
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a method and an apparatus for identifying a curved deformed pipe segment, an electronic device, and a storage medium.
Background
The oil and gas long-distance pipeline has the characteristics of long-distance transmission, shallow pipeline buried depth, wide crossing area, complex and variable crossing geological conditions and the like, so that the pipeline is easy to be damaged by natural disasters such as earthquake, collapse, landslide, debris flow and the like. Geological disasters are one of the main reasons for damage and failure of buried oil and gas pipelines, the change of the surface shape and the displacement of soil bodies caused by the geological disasters can cause the pipe body to generate bending deformation, stress concentration areas are easy to appear at points with the most serious deformation, welding seams or deformation restriction points after the pipelines generate the bending deformation under the action of external force, and the pipe body is damaged and failed in serious conditions, so that immeasurable economic loss, casualties and environmental damage are caused. Therefore, in-pipeline detection techniques are needed to monitor whether a pipeline is abnormal or not, so as to avoid various losses caused by the failure of the pipe body.
The in-pipeline detection technology is a common technology which relies on the pressure of a conveying medium in a pipeline to drive an inner detector to operate so as to measure and obtain defects and anomalies on an oil and gas pipeline. The Inertial Measurement Unit (IMU) internal detection technology is a common pipeline internal detection technology, and is an effective method for detecting pipeline bending strain, wherein the core components of the IMU are a three-dimensional orthogonal gyroscope and an accelerometer, after the IMU completes the Measurement of the whole pipeline to be measured, the IMU collects and records the stored data to perform integral calculation and the like, so that the speed, position and attitude data information of the internal detector at any moment can be obtained, the deformation and strain state borne by each pipe section in the whole pipeline to be measured can be calculated based on the attitude data obtained by the IMU detection, and the risk brought by pipeline displacement and bending strain can be identified and controlled through long-term detection.
According to the existing identification and analysis method for the IMU strain data, all pipe sections with bending strain exceeding 0.125% in the IMU strain data are found in advance through Matlab programming, the pipe sections comprise elbows, recesses, bending deformation sections and circular welding seam abnormal section pipelines, the mileage positions of the given elbows and recesses are marked based on the aligned geometric detection data, the strain data in the influence ranges of the geometric characteristic points such as the recesses and the elbows are deleted, and the interference of the elbows and the recesses is eliminated. For the pipe sections with the residual geometric detection unmarked bending strain value larger than 0.125 percent, technicians with practical experience judge the types of the pipe sections section by section in a drawing mode, and mark the pipe sections by a manual identification method, wherein the pipe sections are respectively a suspected elbow, a suspected dent, an abnormal section of a girth weld and a bending deformation section. According to the method, the positions of the geological disaster pipelines are obtained by carrying out section-by-section identification and classification on IMU data through a geometric detection and manual identification method.
The existing identification and analysis of IMU strain data still stays at the stage of manually identifying high-risk pipe sections of geological disasters section by section, and the method has the problems of long time consumption, low identification accuracy and the like.
Disclosure of Invention
The application provides a bending deformation pipe section identification method and device, electronic equipment and a storage medium, which are used for solving the problems of long time consumption and low identification accuracy rate of the existing identification analysis of IMU strain data.
In a first aspect, the present application provides a curved deformed pipe section identification method, including:
acquiring first strain data measured by an Inertial Measurement Unit (IMU), wherein the first strain data are pipeline strain data obtained when the IMU completes detection in a pipeline.
And obtaining a bending strain curve according to the first strain data.
Determining a wave crest of which the bending strain value is greater than a preset value in the bending strain curve to obtain a pipe section to be identified, wherein the pipe section to be identified is a pipe section corresponding to a preset interval on two sides of the wave crest in the pipeline.
And inputting the second strain data corresponding to the pipe section to be identified into the pipe section type identification model to obtain the pipe section abnormal type of the pipe section to be identified.
Optionally, the pipe segment type identification model is obtained based on one-dimensional convolutional neural network model training.
Optionally, before inputting the second strain data corresponding to the pipe segment to be identified into the pipe segment type identification model, the method includes:
training an initial pipe section type recognition model based on a sample database and a label corresponding to sample data to obtain a pipe section type recognition model; the sample database comprises the type of the pipe section, the absolute mileage range where the strain data of the pipe section are located, the number range of the girth weld, the detection date, the number of the pipeline and the strain value of the pipe section.
Optionally, training an initial pipe segment type recognition model based on the sample database and the label corresponding to the sample data to obtain a pipe segment type recognition model, including:
and acquiring sample data corresponding to the abnormal types of the 4 pipe sections with the third strain data, wherein the sample data comprises the sample data corresponding to the abnormal types of the 4 pipe sections.
Training an initial pipe section type recognition model based on sample data corresponding to the 4 pipe section types and labels corresponding to the sample data to obtain a pipe section type recognition model; the label corresponding to the sample data comprises 4 abnormal types of pipe sections, such as sag, bend, bending deformation and circumferential weld abnormity.
Optionally, training an initial pipe segment type recognition model based on sample data corresponding to the 4 pipe segment types and a label corresponding to the sample data to obtain a pipe segment type recognition model, including:
inputting sample data into an initial pipe section type identification model, and obtaining a pipe section prediction type after characteristic processing of an input layer, a convolution layer, a pooling layer, a flat layer, a full-connection layer and an output layer.
And updating the network parameters of the initial pipe section type identification model according to the pipe section prediction type and the label to obtain the pipe section type identification model.
Optionally, updating the network parameters of the initial pipe segment type identification model according to the pipe segment prediction type and the label to obtain the pipe segment type identification model, including:
and constructing a loss function corresponding to the sample data according to the prediction type and the label of the pipe section.
And updating the network parameters of the initial pipe section type identification model according to the loss function corresponding to the sample data to obtain the pipe section type identification model.
In a second aspect, the present application provides a curved deformed tube section identification device comprising:
the acquisition module is used for acquiring first strain data obtained by measurement of the inertial measurement unit IMU, and the first strain data is pipeline strain data obtained when the IMU completes detection in a pipeline.
The acquisition module is further used for obtaining a bending strain curve according to the first strain data.
The determining module is used for determining a wave crest of which the bending strain value is larger than a preset value in the bending strain curve to obtain a pipe section to be identified, wherein the pipe section to be identified is a pipe section corresponding to a preset interval on two sides of the wave crest in the pipeline.
And the identification module is used for inputting the second strain data corresponding to the pipe section to be identified into the pipe section type identification model to obtain the pipe section abnormal type of the pipe section to be identified.
Optionally, the pipe section type identification model is obtained based on one-dimensional convolutional neural network model training.
Optionally, the identification device for the bending deformation pipe section further comprises a training module.
The training module is used for training an initial pipe section type recognition model based on a sample database and a label corresponding to sample data before inputting second strain data corresponding to a pipe section to be recognized into the pipe section type recognition model to obtain a pipe section type recognition model; the sample database comprises the type of the pipe section, the absolute mileage range where the strain data of the pipe section are located, the number range of the girth weld, the detection date, the number of the pipeline and the strain value of the pipe section.
Optionally, the training module is specifically configured to:
and acquiring sample data corresponding to the abnormal types of the 4 pipe sections with the third strain data, wherein the sample data comprises the sample data corresponding to the abnormal types of the 4 pipe sections.
Training an initial pipe section type recognition model based on sample data corresponding to the 4 pipe section types and labels corresponding to the sample data to obtain a pipe section type recognition model; the label corresponding to the sample data comprises 4 abnormal types of pipe sections, such as sag, bend, bending deformation and circumferential weld abnormity.
Optionally, the training module is specifically configured to:
inputting sample data into an initial pipe section type identification model, and obtaining a pipe section prediction type after characteristic processing of an input layer, a convolution layer, a pooling layer, a flat layer, a full-connection layer and an output layer.
And updating the network parameters of the initial pipe section type identification model according to the pipe section prediction type and the label to obtain the pipe section type identification model.
Optionally, the training module is specifically configured to:
and constructing a loss function corresponding to the sample data according to the prediction type and the label of the pipe section.
And updating the network parameters of the initial pipe section type identification model according to the loss function corresponding to the sample data to obtain the pipe section type identification model.
In a third aspect, the present application provides an electronic device, comprising: a memory and a processor;
a memory for storing a computer program.
A processor for reading the computer program stored in the memory and executing the method for identifying a curved and deformed pipe section according to the first aspect of the present invention.
In a fourth aspect, the present application provides a readable storage medium having stored thereon a computer program having stored therein computer executable instructions for implementing the curved deformed tube section identification method as described above in the first aspect when the computer executable instructions are executed by a processor.
In a fifth aspect, the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the curved deformed pipe section identification method of the first aspect is implemented.
The method, the device, the electronic equipment and the storage medium for identifying the bending deformation pipe section are characterized in that first strain data obtained by measurement of an Inertial Measurement Unit (IMU) are obtained, the first strain data are pipeline strain data obtained when the IMU completes detection in a pipeline, a bending strain curve is obtained according to the first strain data, a wave crest of which the bending strain value is larger than a preset value in the bending strain curve is determined, the pipe section to be identified is obtained, the pipe section to be identified is a pipe section corresponding to a preset interval on two sides of the wave crest in the pipeline, and second strain data corresponding to the pipe section to be identified are input into a pipe section type identification model, so that the abnormal type of the pipe section to be identified is obtained. According to the method, the abnormal type of the pipe section to be identified can be obtained only by acquiring the IMU strain data and determining the pipe section to be identified and inputting the second strain data corresponding to the pipe section to be identified into the pipe section type identification model, the abnormal type of the pipe section does not need to be judged manually, time and labor can be saved, the judgment standard is unified, and therefore the judgment efficiency of the abnormal type of the pipe section is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application.
Fig. 1 is a schematic flow chart of a curved deformed pipe section identification method according to an embodiment of the present disclosure;
FIG. 2 is a schematic view of a pipe segment to be identified according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a one-dimensional convolutional neural network according to an embodiment of the present disclosure;
FIG. 4 is a schematic flow chart illustrating another curved deformed pipe section identification method according to an embodiment of the present application;
FIG. 5 is a schematic structural diagram of an identification device for a curved deformed pipe section according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
With the above figures, there are shown specific embodiments of the present application, which will be described in more detail below. These drawings and written description are not intended to limit the scope of the inventive concepts in any manner, but rather to illustrate the inventive concepts to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. The following description refers to the accompanying drawings in which the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
The terms referred to in this application are explained first:
machine learning: the machine learning method is a method for constructing a machine learning model by training a large amount of data and then classifying and predicting new data by using the model, and can divide a machine learning algorithm into three types of supervised learning, unsupervised learning and reinforcement learning based on different learning forms. Machine learning relates to the interdiscipline of multiple fields, and mainly researches how a computer simulates or realizes the learning behavior of a human brain to acquire new knowledge or skills and reorganizes the existing knowledge structure to continuously improve the performance of a model so as to solve optimal parameters.
And (3) supervised learning: and training to obtain an optimal model to realize data classification and prediction by using a sample with known certain characteristics as sample data.
Characteristic engineering: in machine learning or statistics, also known as variable selection, attribute selection, or variable subset selection, is the process of selecting relevant features and constructing feature subsets in model construction.
The technical scheme provided by the embodiment of the application can be applied to bending strain detection of buried oil and gas pipelines, and particularly defect and abnormity on the pipelines are detected by applying IMU internal detection technology. The international leading internal detection technology represented by IMU, magnetic flux leakage and the like is formed in China, and with the continuous popularization of the detection technology in pipelines, the obtained IMU strain data can be continuously increased, so that the intelligent identification research of the geological disaster high-risk pipeline section based on machine learning is carried out, and the intelligent identification research has great significance for identifying the large strain pipeline section under the influence of the geological disaster and carrying out the integrity evaluation on the geological disaster section pipeline.
In the existing method, semi-automatic identification of high-risk pipe sections of geological disasters is realized based on a Matlab algorithm program, data alignment based on a girth weld number is realized on IMU strain data and geometric detection data of the same pipe, all pipe sections with bending strain exceeding 0.125% in the IMU strain data are found in advance through Matlab programming, the pipe sections comprise elbows, pits, bending deformation sections and girth weld abnormal section pipes, the mileage positions of the given elbows and pits are marked based on the aligned geometric detection data, strain data in the influence ranges of geometric characteristic points such as the pits and the elbows are deleted, and the interference of the elbows and the pits is eliminated. For the pipe sections with the residual geometric detection unmarked bending strain value larger than 0.125 percent, technicians with practical experience mark the pipe sections in a certain range at two sides of the large strain point as abnormal sections to be checked in a drawing mode, judge the type of the pipe sections section by section according to the characteristics of the pipe sections in the range, and mark the pipe sections with labels, namely a suspected elbow, a suspected recess, an abnormal section of a girth weld and a bending deformation section. In this method, part of the pipe sections can be marked as bends or recesses by means of the aligned geometric inspection data, but a large number of pipe sections still need to be judged manually, which requires a lot of time and effort, and requires a worker with sufficient practical experience, and a layman has difficulty in understanding the characteristic differences between different pipe section types. In addition, because the cardinality of the data is large, slight characteristic differences still exist among the same pipe section types, and the unified standard is difficult to maintain in the manual identification process, so that the condition of misjudgment is easy to occur, and the accurate judgment is difficult to be given for the pipe section characteristics with two sharp edges.
In order to solve the problems, the application provides a method for identifying a bending deformation pipe section, which includes the steps of obtaining first strain data obtained by measurement of an inertial measurement unit IMU, wherein the first strain data are pipe strain data obtained when the IMU completes detection in a pipe, obtaining a bending strain curve according to the first strain data, determining a wave peak of which the bending strain value in the bending strain curve is larger than a preset value, determining pipe sections corresponding to preset intervals on two sides of the wave peak in the pipe to be identified as pipe sections to be identified, and inputting second strain data corresponding to the pipe sections to be identified into a pipe section type identification model to obtain abnormal pipe section types of the pipe sections to be identified. In the method, the pipe section to be identified is determined according to the acquired first strain data obtained by IMU measurement, the second strain data corresponding to the pipe section to be identified is input into the pipe section type identification model, the pipe section abnormal type of the pipe section to be identified can be obtained, the pipe section abnormal type is not required to be judged manually, time and labor can be saved, the judgment standard is unified, and therefore the judgment efficiency of the pipe section abnormal type is improved.
The following describes the technical solutions of the present application and how to solve the above technical problems with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a curved deformed pipe section identification method according to an embodiment of the present disclosure, where the curved deformed pipe section identification method may be executed by software and/or a hardware device, for example, the hardware device may be an electronic device, such as a terminal or a server. For example, referring to fig. 1, the identification method of the bent and deformed pipe section may include:
s101, first strain data obtained by measurement of an inertial measurement unit IMU are obtained.
And the first strain data are pipeline strain data obtained when the IMU completes detection in the pipeline.
In the step, an inertial measurement unit IMU is a device for measuring the three-axis attitude angle (or angular rate) and acceleration of an object, the core components of the inertial measurement unit IMU are a three-dimensional orthogonal gyroscope and an accelerometer, after the IMU completes the measurement of the whole pipeline to be measured, the IMU collects and records the stored data to perform calculation such as integration and the like, so that the speed, position and attitude data information of the IMU at any time can be obtained, and the deformation and strain state borne by the pipeline section can be estimated based on the attitude data detected by the IMU. The first strain data can be understood as IMU strain data obtained by measuring the pipeline to be measured by an IMU internal detection technology, and includes information such as velocity, position and attitude data of the IMU at any time in the pipeline.
Specifically, an inertial measurement unit IMU arranged in the pipeline is used for measuring in the pipeline, and speed, position and attitude data of the IMU at any time in the pipeline, namely first strain data, are obtained.
And S102, obtaining a bending strain curve according to the first strain data.
In this step, the bending strain curve may be expressed as a relationship between a bending strain value and an absolute distance, and the bending strain value may be calculated from the IMU strain data.
Specifically, a corresponding bending strain value can be obtained according to the first strain data, and then a bending strain curve can be drawn according to the bending strain value and the absolute mileage corresponding to the bending strain value. The absolute distance can be understood as the distance between the strain point and the measurement starting position of the whole pipeline to be measured.
S103, determining a wave crest of which the bending strain value is larger than a preset value in the bending strain curve to obtain the pipe section to be identified.
In this step, the pipe sections to be identified are corresponding pipe sections in the pipeline in the preset interval on both sides of the wave crest.
The preset value may be set according to actual conditions or experience, for example, may be set to 0.125%, and the specific value of the preset value is not limited herein. When determining the peak in the bending strain curve, where the bending strain value is greater than the preset value, the peak may be determined by using a findpeak function in python, or may be implemented by a programming method. The findpeak function can be expressed as:
peaks＝scipy.signal.find_peaks(x,threshold,distance) (1)
in formula (1): x is the peaked signal, threshold is the set recognition threshold, and distance is the minimum horizontal separation between two peaks.
Specifically, according to a set preset value, a wave crest with a bending strain value larger than the preset value in a bending strain curve can be found, then two strain points with bending strain values as the preset values on the bending strain curves on the left side and the right side of the wave crest are respectively found, an absolute mileage interval corresponding to the two strain points is a preset interval on the two sides of the wave crest, and then pipe sections in the absolute mileage interval in the pipeline can be found, namely the pipe sections to be identified.
Fig. 2 is a schematic diagram of a pipe segment to be identified, as shown in fig. 2, where the preset value is 0.125%, and according to the preset value, 0.125%, a wave crest A, B and a wave crest C with a bending strain value greater than 0.125% in a bending strain curve can be found, taking the wave crest a as an example, two strain points with a bending strain value of 0.125% on the bending strain curves on the left and right sides of the wave crest a are respectively E and F, an absolute distance corresponding to the strain point E is 3330m, an absolute distance corresponding to the strain point F is 3340m, an absolute distance corresponding to the two strain points is (3330, 3340), and then preset intervals (3330, 3340) on the two sides of the wave crest a can be obtained, and at this time, the pipe segment to be identified corresponding to the preset intervals (3330, 3340) is the pipe segment 1 to be identified.
And S104, inputting second strain data corresponding to the pipe section to be identified into the pipe section type identification model to obtain the pipe section abnormal type of the pipe section to be identified.
In this step, the second strain data may be understood as IMU strain data measured on the pipe section to be identified by an intra-IMU detection technique.
For example, the pipe segment type recognition model may be trained based on a one-dimensional neural network model, such as a one-dimensional convolutional neural network model. The initial pipe section type recognition model is trained through the one-dimensional neural network model, characteristic values do not need to be extracted through characteristic engineering, the calculation amount can be reduced, and the characteristic extraction process is simplified.
The pipe section type identification model can be obtained by training other network models, such as a multidimensional convolution neural network model, a cyclic neural network model or a random forest model.
For example, before inputting the second strain data corresponding to the pipe section to be identified into the pipe section type identification model, the initial pipe section type identification model is trained, and the specific steps may be: and training an initial pipe section type recognition model based on the sample database and the label corresponding to the sample data to obtain a pipe section type recognition model.
The sample database comprises the type of the pipe section, the absolute mileage range where the strain data of the pipe section are located, the number range of the girth weld, the detection date, the number of the pipeline and the strain value of the pipe section. The sample database may be a Mysql database.
Specifically, when the initial pipe section type identification model is trained, sample data in a sample database and a label corresponding to the sample data are input into the initial pipe section type identification model, and the pipe section type identification model is obtained through training according to the sample data in the sample database and the label corresponding to the sample data.
Through years of pipeline data monitoring practice, a large amount of pipeline IMU strain data is obtained, and although the strain data of the pipeline IMU is provided with a large amount of IMU strain data, corresponding data processing means are still lacked. In the scheme, a machine learning method taking big data as drive is formed by establishing sample databases of different pipe section types, and the database is continuously expanded along with popularization and application of detection technology in the pipeline, so that the method can be reacted on the machine learning method, and the accuracy of pipe section type identification is improved.
Illustratively, the sample database includes sample data corresponding to sag, bend distortion and girth weld anomaly 4 pipe segment anomaly types. And when training the initial pipe section type identification model according to the sample data in the sample database and the label corresponding to the sample data, acquiring sample data corresponding to 4 pipe section abnormal types of the sag, the bend, the bending deformation and the girth weld abnormality in the third strain data, and training the initial pipe section type identification model based on the sample data corresponding to the 4 pipe section types and the label corresponding to the sample data to obtain a pipe section type identification model.
The label corresponding to the sample data comprises 4 abnormal types of pipe sections, such as sag, bend, bending deformation and circumferential weld abnormity. The third strain data may be understood as historical IMU strain data obtained based on detection techniques within the IMU, and may be obtained from a local database or downloaded from a server.
Illustratively, table 1 is a sample database, and as shown in table 1, the sample database includes a pipe segment type, an absolute mileage range in which pipe segment strain data is located, a girth weld number range, a detection date, a pipe number, and a pipe segment strain value. The pipe section types comprise 4 pipe section abnormal types of sag, bend, bending deformation and girth weld abnormity, and the 4 pipe section abnormal types of sag, bend, bending deformation and girth weld abnormity are respectively represented by digital labels 1, 2, 3 and 4. Taking the pipe section type with the number of 1 as an example, the pipe section type is a recess, the corresponding label is 1, the absolute mileage range is 600.9-620.5m, the circumferential weld number range is 400-420, the detection date is 2016 year and 5 months, the pipeline number is A section, and the pipe section strain value is [ -0.86%, …, 0.23% ].
TABLE 1 sample database
Figure BDA0003574558050000101
Specifically, firstly, sample data corresponding to the abnormal types of the 4 pipe sections with the concave, the elbow, the bending deformation and the circumferential weld abnormal are obtained from historical IMU strain data. And extracting sample data corresponding to the depressions and the bends in the third strain data by a geometric detection method, and extracting sample data corresponding to bending deformation and circumferential weld abnormity in the third strain data by a manual identification method. And training an initial pipe section type recognition model according to the sample data corresponding to the 4 pipe section types and the label corresponding to the sample data to obtain a pipe section type recognition model.
In the scheme, a data base is established for the application and continuous improvement of the machine learning method through the established sample databases of different pipe section types, the detailed information of the required pipe section can be quickly found based on the existing database, and meanwhile, a judgment basis can be provided for the subsequent development of the pipe section. In addition, the pipe section type identification model obtained by training sample data corresponding to 4 pipe section abnormal types of the sink, the elbow, the bending deformation and the girth weld and the label only needs to provide geometric detection data and a manual identification result when a sample database is established, after the pipe section type identification model is obtained by training, the pipe section type can be predicted only based on IMU strain data, the judgment of the pipe section abnormal type is not needed through a manual identification method, the time and the labor can be saved, the standard can be uniformly judged, and the judgment efficiency of the pipe section abnormal type is improved.
For example, the pipe section type identification model may be obtained based on training of a one-dimensional convolutional neural network model, and the model may include 6 parts, namely an input layer, a convolutional layer, a pooling layer, a flat layer, a fully-connected layer, and an output layer.
Specifically, a basic framework of a one-dimensional convolutional neural network model can be built by using a tensorflow module in python, and fig. 3 is a schematic structural diagram of the one-dimensional convolutional neural network, and the model may include 6 portions, namely an input layer (not shown), a convolutional layer, a pooling layer, a flat layer, a fully-connected layer and an output layer (not shown).
The input layer of the one-dimensional convolutional neural network model may receive a one-dimensional or two-dimensional array. Inputting sample data at an input layer of the initial pipe segment type identification model, wherein the sample data is IMU strain data and comprises components in the horizontal direction and the vertical direction, namely a 2 x n pipe segment strain characteristic matrix, which can be expressed as:
Figure BDA0003574558050000111
in formula (2): t is a strain value matrix;
Figure BDA0003574558050000112
a horizontal strain component that is the nth strain point;
Figure BDA0003574558050000113
is the vertical strain component of the nth strain point.
The input layer is a sample data input port of the neural network model, the preset input dimension of the input layer needs to be consistent with the dimension of the sample data, and the one-dimensional convolution neural network needs to convert the multi-dimensional sample data, namely the 2 x n pipe section strain characteristic matrix corresponding to the second strain data, into the one-dimensional data.
The convolution layer is a feature extraction layer, reduces the dimension of original sample data through convolution operation, and simultaneously extracts the main features of the sample to prevent the model from being over-fitted due to excessive parameters.
The pooling layer is used for further reducing the size of the characteristic data, can greatly reduce the number of parameters in the network and remove redundant information, and mainly comprises two modes of mean value sampling and maximum value sampling.
The data dimension is changed due to the filter, and the flat layer is used for converting the data to be processed into a one-dimensional vector corresponding to the neural unit of the full connection layer before being input into the neural network of the full connection layer.
The full connection layer is composed of a plurality of neurons, all the neurons are connected completely, and the local features transmitted from the pooling layer are assembled again through the weight matrix to form complete global features.
The output layer consists of a regression classifier, and the output of a plurality of neurons can be mapped into a fixed interval by using a softmax function, so that the identification of various abnormal types of the pipe sections of the pipe section type identification model is realized.
And continuously updating the network parameters of the initial pipe section type identification model according to the pipe section prediction type and the label to obtain the pipe section type identification model.
In the scheme, the initial pipe section type recognition model is trained through the one-dimensional convolution neural network model, the characteristic value is not required to be extracted through the characteristic engineering, the operation amount can be reduced, and the characteristic extraction process can be simplified.
Illustratively, when updating the network parameters of the initial pipe section type identification model according to the pipe section prediction type and the label, firstly, a loss function corresponding to sample data is constructed according to the pipe section prediction type and the label, and the network parameters of the initial pipe section type identification model are updated according to the loss function corresponding to the sample data to obtain the pipe section type identification model.
Specifically, when a loss function corresponding to the sample data is constructed according to the pipe segment prediction type and the label, the loss function may be a cross entropy loss function, and the formula may be represented as:
Figure BDA0003574558050000121
in formula (3): n is the number of samples, x is the sample number, y represents the true value, and a represents the predicted value.
For example, all sample data in the sample database is divided into a training data set and a testing data set according to a certain proportion, for example, the training data set is set according to 4:1, a batch size batch and an iteration number epoch are set in the training process, and a trained pipe section type identification model and parameters thereof are stored.
In the scheme, the network parameters of the initial pipe section type identification model are continuously updated by constructing the loss function corresponding to the label corresponding to the pipe section prediction type and the sample data, so that the accuracy of the training of the pipe section type identification model can be improved.
And when the comprehensive performance of the model is evaluated, performing parameter optimization on the network model to determine the optimal hyper-parameter. In the classification supervised learning model, the commonly used classification model evaluation indexes include accuracy, precision, recall rate, FI value and the like, and the formula can be expressed as follows:
Figure BDA0003574558050000122
Figure BDA0003574558050000123
Figure BDA0003574558050000124
Figure BDA0003574558050000125
in the above formula, Accuracy represents Accuracy, Precision represents Accuracy, Recall represents Recall, FI Score represents FI value, TP represents correct classification as positive sample, TN represents correct classification as negative sample, FP represents incorrect classification as positive sample, and FN represents incorrect classification as negative sample.
For example, when the performance of the pipe segment type identification model is evaluated, when the accuracy of the pipe segment type identification model is greater than or equal to a preset value, the pipe segment type identification model can accurately identify the abnormal type of the pipe segment.
According to the identification method for the bending deformation pipe section, first strain data obtained by measurement of an inertial measurement unit IMU are obtained, the first strain data are pipe strain data obtained when the IMU completes detection in a pipe, a bending strain curve is obtained according to the first strain data, a wave crest in the bending strain curve, of which the bending strain value is larger than a preset value, is determined, the pipe sections corresponding to preset intervals on two sides of the wave crest in the pipe are to-be-identified pipe sections, second strain data corresponding to the to-be-identified pipe sections are input into a pipe section type identification model, and pipe section abnormal types of the to-be-identified pipe sections are obtained. According to the method and the device, the pipe section to be identified is determined according to the acquired IMU strain data, the second strain data corresponding to the pipe section to be identified is input into the pipe section type identification model, the pipe section abnormal type of the pipe section to be identified can be obtained, the pipe section abnormal type is not required to be judged manually, time and labor can be saved, the judgment standard is unified, and therefore the judgment efficiency of the pipe section abnormal type is improved.
Fig. 4 is a schematic flow chart of another identification method for a curved deformed pipe section according to an embodiment of the present application, please refer to fig. 4, where the identification method for a curved deformed pipe section may include a training process and an application process:
when the initial pipe section type identification model is trained, preprocessing third strain data, namely extracting sample data corresponding to a recess and an elbow in the third strain data by a geometric detection method, and extracting sample data corresponding to bending deformation and circumferential weld abnormity in the third strain data by a manual identification method; then, establishing a sample database based on the extracted sample data corresponding to the 4 pipe segment types, and training an initial pipe segment type identification model based on the sample database and the labels corresponding to the sample data to obtain a pipe segment type identification model; and finally, performing performance evaluation on the pipe section type identification model, wherein when the accuracy of the pipe section type identification model is greater than or equal to a preset value, the pipe section type identification model can accurately identify the abnormal type of the pipe section.
When the pipe section type identification model is used for identifying the abnormal type of the pipe section, first strain data obtained by measurement of an Inertial Measurement Unit (IMU) is obtained, the first strain data are pipeline strain data obtained when the IMU completes detection in a pipeline, a bending strain curve is obtained according to the first strain data, a wave crest of which the bending strain value is larger than a preset value in the bending strain curve is determined, the pipe section to be identified is obtained, second strain data corresponding to the pipe section to be identified are input into the pipe section type identification model, the abnormal type of the pipe section to be identified is obtained, and therefore intelligent identification and classification of the high-risk section of the geological disaster are achieved.
According to the identification method for the bending deformation pipe section, sample data corresponding to 4 abnormal pipe section types of the pipe section, such as the recess, the elbow, the bending deformation and the circular weld seam, and a pipe section type identification model obtained through label training are obtained, first strain data obtained through measurement by an Inertial Measurement Unit (IMU) are obtained, the first strain data are pipeline strain data obtained when the IMU completes detection in a pipeline, a bending strain curve is obtained according to the first strain data, a wave crest with a bending strain value larger than a preset value in the bending strain curve is determined, the pipe section corresponding to a preset interval on two sides of the wave crest in the pipeline is a pipe section to be identified, second strain data corresponding to the pipe section to be identified are input into the pipe section type identification model, and the pipe section abnormal type of the pipe section to be identified is obtained. According to the method and the device, geometric detection data and manual identification results need to be provided when a sample database is established, after a pipe section type identification model is obtained through training, the abnormal type of the pipe section can be predicted only based on IMU strain data, the abnormal type of the pipe section does not need to be judged through a manual identification method, time and labor can be saved, the judgment standard is unified, and therefore the judgment efficiency of the abnormal type of the pipe section is improved.
Fig. 5 is a schematic structural diagram of a curved deformed pipe section identification device 50 according to an embodiment of the present application, and for example, referring to fig. 5, the curved deformed pipe section identification device 50 includes:
the acquiring module 501 is configured to acquire first strain data obtained by measurement by an inertial measurement unit IMU, where the first strain data is pipeline strain data obtained when the IMU completes detection in a pipeline.
The obtaining module 501 is further configured to obtain a bending strain curve according to the first strain data.
The determining module 502 is configured to determine a peak of the bending strain curve where the bending strain value is greater than a preset value, to obtain a pipe segment to be identified, where the pipe segment to be identified is a pipe segment corresponding to a preset interval on two sides of the peak in the pipeline.
The identification module 503 is configured to input the second strain data corresponding to the pipe segment to be identified into the pipe segment type identification model, so as to obtain the pipe segment abnormal type of the pipe segment to be identified.
Optionally, the pipe segment type identification model is obtained based on one-dimensional convolutional neural network model training.
Optionally, the apparatus for identifying a curved deformed pipe section further comprises a training module 504.
The training module 504 is configured to train an initial pipe section type identification model based on a sample database and a label corresponding to sample data before inputting second strain data corresponding to a pipe section to be identified into the pipe section type identification model, so as to obtain a pipe section type identification model; the sample database comprises the type of the pipe section, the absolute mileage range where the strain data of the pipe section are located, the number range of the girth weld, the detection date, the number of the pipeline and the strain value of the pipe section.
Optionally, the training module 504 is specifically configured to:
and acquiring sample data corresponding to the abnormal types of the 4 pipe sections with the third strain data, wherein the sample data comprises the sample data corresponding to the abnormal types of the 4 pipe sections.
Training an initial pipe section type recognition model based on sample data corresponding to the 4 pipe section types and labels corresponding to the sample data to obtain a pipe section type recognition model; the label corresponding to the sample data comprises 4 abnormal types of pipe sections, such as sag, bend, bending deformation and circumferential weld abnormity.
Optionally, the training module 504 is specifically configured to:
inputting sample data into an initial pipe section type identification model, and obtaining a pipe section prediction type after characteristic processing of an input layer, a convolution layer, a pooling layer, a flat layer, a full-connection layer and an output layer.
And updating the network parameters of the initial pipe section type identification model according to the pipe section prediction type and the label to obtain the pipe section type identification model.
Optionally, the training module 504 is specifically configured to:
and constructing a loss function corresponding to the sample data according to the prediction type and the label of the pipe section.
And updating the network parameters of the initial pipe section type identification model according to the loss function corresponding to the sample data to obtain the pipe section type identification model.
The curved deformed pipe section identification apparatus 50 shown in the embodiment of the present application may implement the technical solution of the curved deformed pipe section identification method in the above embodiment, and the implementation principle and the beneficial effects thereof are similar to those of the curved deformed pipe section identification method, and reference may be made to the implementation principle and the beneficial effects of the curved deformed pipe section identification method, which are not described herein again.
Fig. 6 is a schematic structural diagram of an electronic device 60 according to an embodiment of the present application, and for example, please refer to fig. 6, the electronic device 60 may include a processor 601 and a memory 602; wherein the content of the first and second substances,
a memory 602 for storing a computer program.
The processor 601 is configured to read the computer program stored in the memory 602, and execute the curved deformed pipe segment identification method in the above embodiment according to the computer program in the memory 602.
Alternatively, the memory 602 may be separate or integrated with the processor 601. When the memory 602 is a separate device from the processor 601, the electronic device 60 may further include: a bus for connecting the memory 602 and the processor 601.
Optionally, this embodiment further includes: a communication interface, which may be connected to the processor 601 through a bus. The processor 601 may control the communication interface to implement the above-described functions of acquisition and transmission of the electronic device 60.
For example, in the embodiment of the present application, the electronic device 60 may be a terminal, or may also be a server, and may be specifically configured according to actual needs.
The electronic device 60 shown in the embodiment of the present application can execute the technical solution of the identification method for a bent and deformed pipe section in the above embodiment, and the implementation principle and the beneficial effects thereof are similar to those of the identification method for a bent and deformed pipe section, and reference may be made to the implementation principle and the beneficial effects of the identification method for a bent and deformed pipe section, which are not described herein again.
The embodiment of the present application further provides a computer-readable storage medium, where a computer execution instruction is stored in the computer-readable storage medium, and when a processor executes the computer execution instruction, the technical solution of the identification method for a curved deformed pipe section in the foregoing embodiment is implemented, and the implementation principle and the beneficial effect of the identification method for a curved deformed pipe section are similar to those of the implementation principle and the beneficial effect of the identification method for a curved deformed pipe section, which are not described herein again.
The embodiment of the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the technical solution of the identification method for a curved deformed pipe section in the foregoing embodiment is implemented, and the implementation principle and the beneficial effect of the identification method for a curved deformed pipe section are similar to those of the identification method for a curved deformed pipe section, which can be referred to as the implementation principle and the beneficial effect of the identification method for a curved deformed pipe section, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and in actual implementation, there may be other divisions, for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts shown as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment. In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
The integrated module implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute some steps of the methods according to the embodiments of the present application.
It should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of hardware and software modules.
The Memory may include a Random Access Memory (RAM), a Non-Volatile Memory (NVM), for example, at least one disk Memory, and may also be a usb disk, a removable hard disk, a read-only Memory, a magnetic disk or an optical disk.
The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, the buses in the figures of the present application are not limited to only one bus or one type of bus.
The computer readable storage medium may be any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk or optical disk. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.

Claims (10)

1. A method of identifying a curved deformed pipe section, comprising:
acquiring first strain data measured by an Inertial Measurement Unit (IMU), wherein the first strain data are pipeline strain data obtained when the IMU completes detection in a pipeline;
obtaining a bending strain curve according to the first strain data;
determining a wave crest of which the bending strain value is greater than a preset value in the bending strain curve to obtain a pipe section to be identified, wherein the pipe section to be identified is a pipe section corresponding to a preset interval on two sides of the wave crest in the pipeline;
and inputting the second strain data corresponding to the pipe section to be identified into a pipe section type identification model to obtain the pipe section abnormal type of the pipe section to be identified.
2. The method of claim 1, wherein the pipe segment type recognition model is trained based on a one-dimensional convolutional neural network model.
3. The method according to claim 1 or 2, wherein before entering the second strain data corresponding to the pipe segment to be identified into a pipe segment type identification model, the method comprises:
training an initial pipe section type identification model based on a sample database and a label corresponding to sample data to obtain the pipe section type identification model; the sample database comprises the type of the pipe section, the absolute mileage range of the strain data of the pipe section, the number range of the girth joint, the detection date, the number of the pipeline and the strain value of the pipe section.
4. The method of claim 3, wherein the training of the initial pipe segment type recognition model based on the sample database and the labels corresponding to the sample data to obtain the pipe segment type recognition model comprises:
acquiring sample data corresponding to the abnormal types of 4 pipe sections with abnormal girth welds, bending deformation and sag in third strain data, wherein the sample data comprises the sample data corresponding to the abnormal types of the 4 pipe sections;
training the initial pipe section type recognition model based on the sample data corresponding to the 4 pipe section types and the label corresponding to the sample data to obtain the pipe section type recognition model; and the label corresponding to the sample data comprises 4 abnormal types of pipe sections, such as a recess, an elbow, bending deformation and circumferential weld abnormity.
5. The method of claim 4, wherein the training the initial pipe segment type recognition model based on the sample data corresponding to the 4 pipe segment types and the labels corresponding to the sample data to obtain the pipe segment type recognition model comprises:
inputting the sample data into the initial pipe section type identification model, and obtaining a pipe section prediction type after the characteristic processing of an input layer, a convolution layer, a pooling layer, a flat layer, a full-connection layer and an output layer;
and updating the network parameters of the initial pipe section type identification model according to the pipe section prediction type and the label to obtain the pipe section type identification model.
6. The method of claim 5, wherein updating network parameters of the initial pipe segment type identification model based on the predicted type of the pipe segment and the label to obtain the pipe segment type identification model comprises:
constructing a loss function corresponding to the sample data according to the pipe section prediction type and the label;
and updating the network parameters of the initial pipe section type identification model according to the loss function corresponding to the sample data to obtain the pipe section type identification model.
7. A curved deformed pipe section identification apparatus, comprising:
the system comprises an acquisition module, a detection module and a control module, wherein the acquisition module is used for acquiring first strain data obtained by measurement of an Inertial Measurement Unit (IMU), and the first strain data is pipeline strain data obtained when the IMU completes detection in a pipeline;
the acquisition module is further used for obtaining a bending strain curve according to the first strain data;
the determining module is used for determining a wave crest in the bending strain curve, wherein the bending strain value of the wave crest is larger than a preset value, so as to obtain a pipe section to be identified, and the pipe section to be identified is a pipe section corresponding to a preset interval on two sides of the wave crest in the pipeline;
and the identification module is used for inputting the second strain data corresponding to the pipe section to be identified into a pipe section type identification model to obtain the pipe section abnormal type of the pipe section to be identified.
8. An electronic device, comprising: a memory and a processor;
the memory for storing a computer program;
the processor is configured to read the computer program stored in the memory and execute the curved deformed pipe section identification method according to any one of claims 1 to 6 according to the computer program in the memory.
9. A readable storage medium having stored thereon a computer program having stored therein computer executable instructions for implementing the method of identifying a bendingly deformed pipe section according to any one of claims 1 to 6 when executed by a processor.
10. A computer program product comprising a computer program for implementing the method of identifying a curved deformed pipe section according to any one of claims 1 to 6 when executed by a processor.
CN202210329101.8A 2022-03-31 2022-03-31 Bent deformed pipe section identification method and device, electronic equipment and storage medium Pending CN114676730A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210329101.8A CN114676730A (en) 2022-03-31 2022-03-31 Bent deformed pipe section identification method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210329101.8A CN114676730A (en) 2022-03-31 2022-03-31 Bent deformed pipe section identification method and device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN114676730A true CN114676730A (en) 2022-06-28

Family

ID=82076405

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210329101.8A Pending CN114676730A (en) 2022-03-31 2022-03-31 Bent deformed pipe section identification method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN114676730A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115060222A (en) * 2022-08-10 2022-09-16 昂坤视觉(北京)科技有限公司 Wafer surface type classification method and system
CN115464018A (en) * 2022-11-02 2022-12-13 江苏新恒基特种装备股份有限公司 Three-dimensional space continuous multi-bend pipe accurate forming measurement control system and method

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115060222A (en) * 2022-08-10 2022-09-16 昂坤视觉(北京)科技有限公司 Wafer surface type classification method and system
CN115464018A (en) * 2022-11-02 2022-12-13 江苏新恒基特种装备股份有限公司 Three-dimensional space continuous multi-bend pipe accurate forming measurement control system and method
CN115464018B (en) * 2022-11-02 2023-01-31 江苏新恒基特种装备股份有限公司 Three-dimensional space continuous multi-bend pipe accurate forming measurement control system and method

Similar Documents

Publication Publication Date Title
CN114676730A (en) Bent deformed pipe section identification method and device, electronic equipment and storage medium
CN112508105B (en) Fault detection and retrieval method for oil extraction machine
CN105510038A (en) Wind turbine generator fault monitoring method and device
CN108363886A (en) Deformation prediction method and system based on deep learning
CN111737909A (en) Structural health monitoring data anomaly identification method based on space-time graph convolutional network
CN117540174B (en) Building structure multi-source heterogeneous data intelligent analysis system and method based on neural network
CN116123988B (en) Pipeline geometric deformation detection method and system based on data processing
CN112836719B (en) Indicator diagram similarity detection method integrating two classifications and triplets
CN112766301B (en) Oil extraction machine indicator diagram similarity judging method
CN115597494B (en) Precision detection method and system for prefabricated part preformed hole based on point cloud
CN114387332B (en) Pipeline thickness measuring method and device
CN116186946A (en) Hydraulic system fault diagnosis method and system based on diagnosis model
CN116342538A (en) Method and device for detecting running and leaking, intelligent equipment and storage medium
CN115171045A (en) YOLO-based power grid operation field violation identification method and terminal
Ye et al. A deep learning-based method for automatic abnormal data detection: Case study for bridge structural health monitoring
CN115719411A (en) Three-dimensional geological modeling method, system, computer and readable storage medium
CN113780111B (en) Pipeline connector defect accurate identification method based on optimized YOLOv3 algorithm
CN117421616B (en) Mine shaft detection system and method
Xiu et al. DS-Net: A dedicated approach for collapsed building detection from post-event airborne point clouds
CN115063337A (en) Intelligent maintenance decision-making method and device for buried pipeline
CN116663126A (en) Bridge temperature effect prediction method based on channel attention BiLSTM model
CN113935377B (en) Pipeline leakage aperture identification method combining characteristic migration with time-frequency diagram
CN113610957B (en) BIM-based automatic management method for three-dimensional defect information of drainage pipeline
CN114330442A (en) Pipeline strain characteristic classification calculation method and system based on K-nearest neighbor method
CN116011183B (en) In-service oil and gas pipeline detection method, device, equipment and storage medium

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