CN115034250A - Intelligent evaluation visualization system for pipeline integrity - Google Patents
Intelligent evaluation visualization system for pipeline integrity Download PDFInfo
- Publication number
- CN115034250A CN115034250A CN202110252348.XA CN202110252348A CN115034250A CN 115034250 A CN115034250 A CN 115034250A CN 202110252348 A CN202110252348 A CN 202110252348A CN 115034250 A CN115034250 A CN 115034250A
- Authority
- CN
- China
- Prior art keywords
- pipeline
- defect
- detection
- evaluation
- intelligent
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 190
- 238000012800 visualization Methods 0.000 title claims abstract description 48
- 230000007547 defect Effects 0.000 claims abstract description 351
- 238000001514 detection method Methods 0.000 claims abstract description 258
- 238000012423 maintenance Methods 0.000 claims abstract description 186
- 230000002159 abnormal effect Effects 0.000 claims abstract description 103
- 238000007726 management method Methods 0.000 claims abstract description 83
- 238000013523 data management Methods 0.000 claims abstract description 31
- 230000000007 visual effect Effects 0.000 claims abstract description 10
- 230000002452 interceptive effect Effects 0.000 claims abstract description 4
- 238000010586 diagram Methods 0.000 claims description 48
- 238000000034 method Methods 0.000 claims description 43
- 238000012549 training Methods 0.000 claims description 40
- 230000000875 corresponding effect Effects 0.000 claims description 33
- 238000009412 basement excavation Methods 0.000 claims description 30
- 238000004458 analytical method Methods 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 24
- 238000012360 testing method Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 16
- 230000007797 corrosion Effects 0.000 claims description 15
- 238000005260 corrosion Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 13
- 238000012217 deletion Methods 0.000 claims description 9
- 230000037430 deletion Effects 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 8
- 230000003993 interaction Effects 0.000 claims description 7
- 238000012986 modification Methods 0.000 claims description 7
- 230000004048 modification Effects 0.000 claims description 7
- 238000003062 neural network model Methods 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 7
- 239000000463 material Substances 0.000 claims description 6
- 238000009825 accumulation Methods 0.000 claims description 4
- 238000003491 array Methods 0.000 claims description 4
- 230000008676 import Effects 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 230000002596 correlated effect Effects 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 239000003550 marker Substances 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 claims description 3
- 230000006698 induction Effects 0.000 description 30
- 239000013598 vector Substances 0.000 description 24
- 230000008439 repair process Effects 0.000 description 15
- 238000005520 cutting process Methods 0.000 description 11
- 238000013528 artificial neural network Methods 0.000 description 8
- 239000002184 metal Substances 0.000 description 7
- 238000012544 monitoring process Methods 0.000 description 7
- 230000005856 abnormality Effects 0.000 description 6
- 239000000126 substance Substances 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000002950 deficient Effects 0.000 description 3
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011867 re-evaluation Methods 0.000 description 2
- JXASPPWQHFOWPL-UHFFFAOYSA-N Tamarixin Natural products C1=C(O)C(OC)=CC=C1C1=C(OC2C(C(O)C(O)C(CO)O2)O)C(=O)C2=C(O)C=C(O)C=C2O1 JXASPPWQHFOWPL-UHFFFAOYSA-N 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000008846 dynamic interplay Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000010561 standard procedure Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Images
Classifications
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Software Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Geometry (AREA)
- Remote Sensing (AREA)
- Game Theory and Decision Science (AREA)
- Computer Graphics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Pipeline Systems (AREA)
Abstract
The invention discloses an intelligent evaluation visualization system for pipeline integrity, which comprises a three-dimensional visualization module and an intelligent evaluation module for pipelines; the three-dimensional visualization module comprises a three-dimensional model visualization display submodule, a support environment submodule, an evaluation module data interface submodule, a pipeline information directory tree submodule and a pipeline engineering data management submodule; the intelligent pipeline evaluation module comprises a basic data management submodule, a detection management submodule, a defect management submodule and an intelligent integrity evaluation and maintenance decision submodule; the integrity intelligent evaluation and maintenance decision submodule comprises an intelligent maintenance decision for pipeline abnormal signals which cannot quantify the size and type of the defect and an automatic evaluation and maintenance decision for quantifying the defect, and the three-dimensional model visual display submodule performs three-dimensional dynamic interactive visual display. The invention can realize the intelligent evaluation of the integrity of pipelines, particularly pipelines in stations and gathering and transportation pipe networks in refineries, by interacting with the three-dimensional visual platform of the pipelines.
Description
Technical Field
The invention relates to the technical field of intelligent evaluation visualization of pipeline integrity, in particular to an intelligent evaluation visualization system for pipeline integrity.
Background
The integrity evaluation is one of core services of the pipeline integrity management work, the safety state of the pipeline can be judged through the integrity evaluation, and guidance suggestions are provided for the maintenance of the pipeline. To unable to develop interior detection, unable pipeline such as refinery, station and oil gas field gathering and transportation pipe network that excavation detected, at present, the problem that the pipeline faced mainly has: 1) a large number of raw material lines, mutual supply lines, power lines and the like are distributed in refinery plants and station areas, most of the power lines are distributed underground, underground pipelines are various in types, complex in distribution and high in unified management difficulty; 2) the oil and gas field gathering and transportation pipe network, because the pipeline quantity is numerous, the old pipeline data is lacked seriously, the trend of the pipeline, the distribution situation of the underground pipeline often appear the situation that the drawing is inconsistent with the actual situation on the spot, supervisor know, cause great difficulty for the work such as the construction, examining and repairing and reconstruction of the apparatus and maintenance and renewal of the apparatus; 3) underground conditions are complex, corrosion and thinning of pipelines commonly exist, the old pipe networks are overloaded and operate in an overyear limit, accidents such as pipe explosion, leakage, wire stringing and the like are caused by corrosion, pressure and the like, the leakage phenomenon sometimes occurs, and a large potential safety hazard exists; 4) the pipeline cannot carry out internal detection and cannot master the safety state of the pipeline in time. 5) The existing mode of evaluating the detection data through manual experience cannot ensure the accuracy of the evaluation result.
Disclosure of Invention
Aiming at the problems in the prior art, the embodiment of the invention provides an intelligent pipeline integrity evaluation visualization system, and particularly relates to an intelligent pipeline integrity evaluation visualization system which realizes real-time dynamic interaction with a three-dimensional pipeline model.
Specifically, the embodiment of the invention provides the following technical scheme:
the embodiment of the invention provides a visualization system for intelligent evaluation of pipeline integrity, which comprises: the system comprises a three-dimensional visualization module and a pipeline intelligent evaluation module;
the three-dimensional visualization module comprises a three-dimensional model visualization display submodule, a supporting environment submodule, an evaluation module data interface submodule, a pipeline information directory tree submodule and a pipeline engineering data management submodule;
the intelligent pipeline evaluation module comprises a basic data management submodule, a detection management submodule, a defect management submodule and an intelligent integrity evaluation and maintenance decision submodule;
the evaluation module data interface submodule is used for providing a data interface for a basic data management submodule, a detection management submodule, a defect management submodule and an integrity intelligent evaluation and maintenance decision submodule;
the integrity intelligent evaluation and maintenance decision submodule comprises an intelligent maintenance decision for the pipeline abnormal signals of which the sizes and types of the defects can not be quantized and an automatic evaluation and maintenance decision for the quantized defects, and the intelligent maintenance decision is carried out by adopting an intelligent maintenance decision model based on a signal diagram or signal data which is trained in advance through deep learning and by acquiring a detection signal diagram and a detection signal intensity value of the pipeline section from a detection management submodule for the pipeline abnormal signals of which the sizes and types of the defects can not be quantized;
aiming at automatic evaluation and maintenance decision of quantitative defects, batch and single defects of size and type information are obtained through detection, different evaluation methods are adopted for different defect types, and after the evaluation methods are selected for a certain type of defects, a built-in evaluation algorithm is called to perform batch evaluation, single evaluation and maintenance decision of the same evaluation method on defect points of a uniform type or a certain defect; and the three-dimensional model visual display submodule performs three-dimensional dynamic interactive visual display on the intelligent evaluation and maintenance decision results.
Furthermore, the three-dimensional visualization module is used for carrying out field survey on the refinery equipment and devices, the ground markers and the ground and underground pipeline objects connected with the equipment and the devices on the basis of a man-machine interaction principle, determining the geographic information or the relative position of the modeling object, establishing a model of the object in the three-dimensional model, and adding, checking, deleting and modifying the established pipeline, device and equipment related engineering information through the three-dimensional model or a directory tree, wherein the information of the pipeline is used as a data base for evaluating the integrity of the buried pipeline.
Further, the basic data management submodule is used for acquiring basic information of a pipeline where the pipeline is located from the three-dimensional visualization module and is used for increasing, checking, deleting and modifying information of pipe diameters, wall thicknesses, pipes, conveying media and connecting equipment.
Furthermore, the detection management submodule comprises three functions of detected pipeline management, detection engineering management and detection data management and is used for logically managing detection operation, detection signals and detection results;
wherein, the detected pipeline management: the method is used for managing the detection pipeline and the related detection engineering thereof, the detection pipeline manages the detection range of the detected pipeline, and on the basis of the built-in pipeline number naming rule, the pipeline number of the newly-added detection pipeline is automatically generated based on the pipeline where the detected pipeline is located, and the information of the newly-added detection pipeline comprises the following steps: detecting length, detecting and positioning, maintaining, modifying and checking remark information, and meanwhile, after a detection pipeline is newly built, highlighting and positioning the detected pipeline on the three-dimensional model through a pipeline number; in addition, the detection engineering and the related information of the detected pipeline can be added, checked, deleted and modified under the added detected pipeline; after the pipeline is detected for the first time, on the three-dimensional model of the pipeline, through detecting marker information, manually positioning, clicking and determining the detection range of the detection pipeline on the pipeline to be detected, automatically generating a detection pipeline number by the system according to the pipeline where the detection pipeline is located, and simultaneously realizing highlight positioning with the detected pipeline in the three-dimensional model according to the detection pipeline number; maintaining and modifying the information of the detected pipeline, including detection length, detection positioning point, input and modification of design coefficient information and deletion of the detected pipeline; after the detected pipeline is deleted, the detection project, the position of the defect point and the defect data which are related to the detected pipeline are deleted at the same time;
detection engineering management: the method is used for establishing a detection project after one section of pipeline carries out detection operation, the detection project is newly established under the newly added detected pipeline, and maintenance management is carried out on the detection project and relevant information thereof, including detection project name, detection method, detection equipment, detection time and detection unit, such as addition, search, deletion and modification; detecting the position of a defect point associated with the project after the project is deleted, and deleting the defect data at the same time; the module also checks all detection project lists corresponding to the detected pipelines according to the serial numbers of the detected pipelines;
and (3) detection data management: the system is used for managing detection data of a certain detection project, and the functions comprise detection signal diagram management, detection signal data management and detection result data management; the detection signal diagram management is used for managing the detection signal diagram of the pipeline; the detection signal data is managed in a form, and the data is imported into the system in the form of the form for the strength value of the detection signal along the detection mileage according to the sampling result of the detector; the detection result data is also managed in a form of a form import system; the method also comprises the steps of detecting signals, downloading a result template, importing data, and adding, searching, deleting and modifying the data.
Furthermore, the defect management submodule is used for managing the obtained defect data after the pipeline is detected for one time through expert analysis signals, excavation verification or intelligent maintenance decision; the defect data can interact with the three-dimensional model, so that the defect information of the detected pipeline can be increased, checked, deleted and changed, and meanwhile, the defect information is managed respectively according to different defect types, wherein the corrosion rate of corrosion defects is directly calculated according to the defect depth and the commissioning time through a background built-in algorithm;
wherein, newly-increased defects adopt different processing modes:
for single defects obtained by excavation verification and expert analysis, on the premise of known defect positioning, scale measurement is carried out on a three-dimensional pipeline model through markers on a detected pipeline, the positions of the defects are directly added, and then defect data are recorded;
for the defects obtained by intelligent maintenance decision, outputting the mileage position and the maintenance grade of the abnormal area of the pipeline through an intelligent maintenance decision model, automatically marking an abnormal interval on the pipeline section corresponding to the three-dimensional platform, automatically synchronizing the maintenance grade to the abnormal interval of the corresponding pipeline, and simultaneously storing the defects and the maintenance decision result to a defect management sub-module so as to enable the defects to enter an intelligent evaluation interface to display the maintenance decision result, wherein the defects obtained by intelligent decision can be analyzed by experts and verified by excavation, and are simultaneously stored through a field of a defect adding mode; wherein, the defect addition mode includes: expert analysis, excavation verification or intelligent decision making;
for batch defects, batch defects obtained through intelligent decision are subjected to batch intelligent addition and positioning in the same mode, other defects are imported through data, and the method is specifically operated as follows: downloading a defect information template, editing defect information, importing the defect information into the system, manually inputting defect positioning aiming at each piece of defect data after the defect batch data is successfully imported, displaying newly-added defects in a three-dimensional model in a correlated manner, or selecting the defect to be added and positioned, adding the defect on the three-dimensional pipeline through an adding button of the defect positioning, and automatically updating the defect information to a three-dimensional defect point.
Further, the deletion of defects includes three cases:
A. deleting the position of the defect point and reserving the defect data, and automatically updating the defect data after adding the mileage position of the abnormal area of the pipeline with the same defect number again;
B. deleting the defect data to reserve the position of the three-dimensional defect point, and automatically associating the position defect point information to the three-dimensional defect point after newly adding the position defect point information again;
C. deleting the defect data and the mileage position of the abnormal area of the pipeline on the three-dimensional model pipeline simultaneously;
the defect management submodule realizes highlight positioning of the defect point in three dimensions through the defect point number; on the defect management page, through a page pull-down list, the year, the device, the pipeline number and the detection project are selected, and all defect points and defect information under the detection project can be checked.
Furthermore, the integrity intelligent evaluation and maintenance decision submodule comprises an intelligent maintenance decision for pipeline abnormal signals which cannot quantify the size and type of the defects and an automatic evaluation and maintenance decision for quantifying the defects;
aiming at the pipeline abnormal signals which can not quantify the sizes and types of the defects, a detection signal diagram and a detection signal intensity value of the pipeline section are obtained from a detection management submodule, a pre-trained intelligent maintenance decision model based on the detection signal diagram and an intelligent maintenance decision model based on the detection signal intensity value are adopted for the detection signal diagram and the detection signal intensity value, and a signal diagram or a signal data value with a standard length is determined according to the signal diagram and the signal intensity data of the pipeline
If signal diagram or signal value length
Scaling the signal diagram and the values to a standard size; if the signal length
Starting from the left end of the signal array, establish a length of
Is sliding ofWindow of each
Length clipping the signal once until the right side of the sliding window reaches the boundary of the maximum value of the signal, obtaining
A signal graph or signal array; respectively inputting the signal diagrams or the signal arrays into an intelligent maintenance decision model for prediction, and respectively outputting the mileage interval where the abnormal signal position is located and the maintenance grade of the abnormal signal according to the prediction result;
if the signals comprise ultrasonic guided waves and magnetic stress signals, the results of the ultrasonic guided waves and the magnetic stress signals are simultaneously output, wherein the mileage interval of the abnormal signals is the union of the two results, the maintenance decision level is the result of a higher level, and the abnormal signal mileage interval of the pipeline and the corresponding maintenance decision result are output; or the results of the two are output after data fusion is carried out according to the weight;
and the background automatically marks the abnormal interval on the pipe section in a red form in the three-dimensional platform according to the output abnormal mileage interval, automatically synchronizes the maintenance decision result to the abnormal interval of the pipe section, simultaneously stores the evaluation result, and can display the evaluation result when entering the intelligent evaluation interface next time.
Further, aiming at automatic evaluation and maintenance decision of quantitative defects, batch and single defects of size and type information are obtained through detection, different defect type defect evaluation methods are different, and after an evaluation method is selected for a certain type of defects, the system can call a built-in evaluation algorithm to perform batch evaluation, single evaluation and maintenance decision of the same evaluation method on defect points of a uniform type or a certain defect;
after the system is successfully evaluated, the evaluation result is stored, and the evaluation page is entered for displaying the evaluation result next time; the single defect automatic evaluation is to select a defect point to be evaluated in a single evaluation module according to the defect point with known defect type and defect size, select an evaluation method, automatically evaluate the defect point, store a defect evaluation result for the evaluated defect point, display the evaluation result when entering again, modify the evaluation method and re-evaluate; the automatic evaluation of the batch defects is to select an evaluation method for the defect points needing to be evaluated after a selection device, a pipeline number, a detection project and a defect type are carried out on a batch evaluation page, to input information required by the evaluation, to carry out the batch evaluation of a certain type of defects through a system built-in evaluation algorithm, and to modify the evaluation method for the evaluated defect points for re-evaluation.
Further, when the excavation data is not available in the previous period, the intelligent maintenance decision model is trained and tested through expert analysis data in the defect management submodule;
and after the accumulation degree of the excavation data meets the preset condition, acquiring excavation defect data from the defect management submodule, and optimizing the intelligent maintenance decision model through data migration.
Further, the training process of the intelligent maintenance decision model based on the detection signal diagram comprises the following steps:
acquiring a sample magnetic stress picture from expert analysis data in a defect management submodule; wherein, various defect pixel positions and various maintenance grades need to be covered in the sample magnetic stress picture;
acquiring the mileage position and the maintenance grade of the abnormal area of the pipeline of the sample magnetic stress picture from expert analysis data in a defect management submodule;
taking a sample magnetic stress picture with a predetermined defect pixel position and a predetermined maintenance grade as the input of a model, taking the mileage position and the maintenance grade of the abnormal area of the pipeline in the sample magnetic stress picture as the output of the model, and training and testing a neural network model based on an intelligent learning algorithm to obtain an intelligent maintenance decision model;
the training process of the intelligent maintenance decision model based on the detection signal strength value comprises the following steps:
acquiring a sample magnetic stress detection signal intensity value from expert analysis data in a defect management submodule; wherein, the sample magnetic stress detection signal intensity value needs to cover various defect pixel positions and various maintenance levels;
acquiring the position and maintenance grade of the abnormal area of the pipeline corresponding to the strength value of the sample magnetic stress detection signal from expert analysis data in a defect management submodule; wherein, the data is divided into a training set and a testing set;
the method comprises the steps of taking a sample magnetic stress detection signal intensity value with a predetermined defect pixel position and a predetermined maintenance grade as an input of a model, taking a pipeline abnormal region mileage position and a maintenance grade corresponding to the sample magnetic stress detection signal intensity value as an output of the model, and training and testing a neural network model based on an intelligent learning algorithm to obtain an intelligent maintenance decision model.
According to the technical scheme, the intelligent evaluation visualization system for the integrity of the pipeline provided by the embodiment of the invention can realize the intelligent evaluation of the integrity of the pipeline, particularly the pipeline in a refinery, a station, a gathering and transportation pipe network and the like, by interacting with the three-dimensional visualization platform of the pipeline.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic structural diagram of an intelligent evaluation system for pipeline integrity according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The intelligent evaluation visualization system for the integrity of the pipeline provided by the embodiment of the invention realizes intelligent analysis evaluation and maintenance decision of the pipeline defects of the buried pipeline, particularly the pipeline in a refinery, a gathering and transportation pipe network and an underground pipeline in a station by interacting with a three-dimensional visualization platform of the buried pipeline.
Fig. 1 shows a schematic structural diagram of a pipe integrity intelligent evaluation visualization system provided by an embodiment of the invention. As shown in fig. 1, an intelligent evaluation visualization system for pipeline integrity provided by an embodiment of the present invention includes: the system comprises a three-dimensional visualization module and a pipeline intelligent evaluation module;
the three-dimensional visualization module comprises a three-dimensional model visualization display submodule, a supporting environment submodule, an evaluation module data interface submodule, a pipeline information directory tree submodule and a pipeline engineering data management submodule;
the intelligent pipeline evaluation module comprises a basic data management submodule, a detection management submodule, a defect management submodule and an intelligent integrity evaluation and maintenance decision submodule;
the evaluation module data interface submodule is used for providing a data interface for a basic data management submodule, a detection management submodule, a defect management submodule and an integrity intelligent evaluation and maintenance decision submodule;
the integrity intelligent evaluation and maintenance decision submodule comprises an intelligent maintenance decision for the pipeline abnormal signals of which the sizes and types of the defects can not be quantized and an automatic evaluation and maintenance decision for the quantized defects, and the intelligent maintenance decision is carried out by adopting an intelligent maintenance decision model based on a signal diagram or signal data which is trained in advance through deep learning and by acquiring a detection signal diagram and a detection signal intensity value of the pipeline section from a detection management submodule for the pipeline abnormal signals of which the sizes and types of the defects can not be quantized;
aiming at automatic evaluation and maintenance decision of quantitative defects, batch and single defects of size and type information are obtained through detection, different evaluation methods are adopted for different defect types, and after the evaluation methods are selected for a certain type of defects, a built-in evaluation algorithm is called to perform batch evaluation, single evaluation and maintenance decision of the same evaluation method on defect points or a certain defect of a uniform type; and the three-dimensional model visual display submodule carries out three-dimensional dynamic interactive visual display on the intelligent evaluation and maintenance decision results.
According to the technical scheme, the intelligent evaluation visualization system for the integrity of the pipeline provided by the embodiment can realize the intelligent evaluation of the integrity of the pipeline, particularly the pipeline in a refinery, a station and a gathering and transportation pipe network, through interaction with the three-dimensional visualization platform of the pipeline.
In this embodiment, the intelligent evaluation visualization system for the integrity of the pipeline includes two parts, namely a three-dimensional visualization function and intelligent evaluation for the integrity of the pipeline. The three-dimensional visualization function mainly comprises simplified pipeline connecting equipment, devices, ground markers, visualization display of three-dimensional models of all pipelines connected with the equipment and pipeline information management; the intelligent evaluation function of the pipeline mainly comprises pipeline basic data management, detection data management, defect management, intelligent evaluation of the integrity of the pipeline and maintenance decision for realizing dynamic data interaction with the three-dimensional model.
The following is introduced with respect to the three-dimensional visualization module:
the three-dimensional visualization module comprises a three-dimensional model visualization display, a safe and reliable support environment and an application interface (pipeline basic data management, detection management, defect management and intelligent evaluation) of an integrity intelligent evaluation module. According to the man-machine interaction principle, the geographic information or the relative position of modeling objects is determined by carrying out field survey on objects such as refinery equipment and devices, ground markers (such as roads, well lids and the like) and all ground and underground pipelines connected with the equipment and the devices, the models of the objects are established in a three-dimensional model, and meanwhile, the built pipeline, device and equipment related engineering information is added, checked, deleted and modified through the three-dimensional model or a directory tree, wherein the information of the pipeline is used as a data base for evaluating the integrity of the buried pipeline.
The intelligent evaluation module for the pipeline is introduced as follows:
the intelligent pipeline evaluation module mainly comprises: and the submodules comprise basic data management, detection management, defect management, intelligent integrity evaluation for realizing dynamic data interaction with the three-dimensional model, maintenance decision and the like.
The basic data management submodule comprises basic information of a pipeline where the pipeline is located, which is acquired from the three-dimensional visualization platform, and the basic information comprises increment, check, deletion and modification of information such as pipe diameter, wall thickness, pipe materials, conveying media, connecting equipment and the like.
The detection management submodule is mainly divided into 3 functions of detected pipeline management, detection engineering management and detection data management, and mainly aims at the logical management of detection data such as detection operation, detection signals, detection results and the like. Managing a detected pipeline: the method is mainly used for managing the detection pipeline and the related detection engineering thereof, the detected pipeline is mainly used for managing the detection range of the detected pipeline, and on the basis of the built-in pipeline number naming rule, the pipeline number of the newly-added detection pipeline is automatically generated based on the pipeline where the detected pipeline is located, so that the information of the newly-added detected pipeline is as follows: detecting length, detecting location (or detecting starting point and detecting end point markers) and remark information, maintaining, modifying and checking, and meanwhile, after a detecting pipeline is newly built, highlighting and locating the detected pipeline on the three-dimensional model through a pipeline number; in addition, the detection engineering and the related information of the detected pipeline can be added, checked, deleted and modified under the added detected pipeline. After the pipeline is detected for one time, the detection range of the detection pipeline is determined by manual positioning on the pipeline to be detected through detecting information such as markers and the like on the three-dimensional model of the pipeline, the system automatically generates the number of the detection pipeline according to the pipeline where the detection pipeline is located, and meanwhile, the highlight positioning of the detection pipeline in the three-dimensional model is realized according to the number of the detection pipeline. And maintaining and modifying the information of the detected pipeline, including recording and modifying the information such as the detection length, the detection positioning point, the design coefficient and the like, and deleting the detected pipeline. And deleting the position of the defect point and the defect data of the detection project associated with the detected pipeline after the detected pipeline is deleted. Detection engineering management: after a section of pipeline carries out a detection operation, a detection project is established, namely, the detection project is newly established under the newly added detected pipeline, and maintenance management such as addition, check, deletion, modification and the like is carried out on the detection project and relevant information (detection project name, detection method, detection equipment, detection time, detection unit and the like) of the detection project. And deleting the defect data at the same time when the detection project is deleted and the position of the defect point associated with the detection project is deleted. The module can also check all detection project lists corresponding to the detected pipelines through the serial numbers of the detected pipelines. The detection data management function: management of inspection data for a certain inspection project. The functions comprise detection signal diagram management, detection signal data management and detection result data management. The detection signal diagram management mainly manages the detection signal diagram of the pipeline; the detection signal data is managed in a form of a form, and the data is imported into the system in the form of the form according to the sampling result of the detector for the strength value of the detection signal along the detection mileage; the detection result data is also managed in the form of a form import system. In addition, the method also comprises the steps of detecting signals, downloading a result template, importing data, and adding, checking, deleting and changing the data.
The defect management submodule is mainly used for managing defect data (including defect numbers, defect types, defect sizes, defect positions, defect corrosion rates, defect positioning, evaluation results and the like) obtained by analyzing signals, excavating verification or intelligent maintenance decision through experts after the pipeline is detected for one time. The defect data can realize interaction with the three-dimensional model, increase, check, delete, change and the like of the detected pipeline defect information, meanwhile, the defect information can be respectively managed according to different defect types, such as corrosion defects, manufacturing defects, deformation defects, welding seam abnormity and the like, wherein the corrosion rate of the corrosion defects can be directly calculated according to the defect depth, commissioning time and the like through a background built-in algorithm. Two ways can be adopted for newly adding defects: firstly, on the premise of known defect positioning, scale measurement can be carried out on a three-dimensional pipeline model through markers (a pipeline starting point, a road, a well cover and the like) on a detected pipeline, the position of a defect point is directly added, and then defect data is recorded; outputting the position and the maintenance grade of the defect obtained by the intelligent maintenance decision through an intelligent maintenance decision model, automatically marking an abnormal interval on a pipe section corresponding to the three-dimensional platform, automatically synchronizing the maintenance grade to the abnormal interval of the corresponding pipeline, and simultaneously storing the defect and the maintenance decision result to a defect management sub-module so as to enable the defect to enter an intelligent evaluation interface to display the maintenance decision result, wherein the defect obtained by the intelligent decision can be analyzed by an expert and verified by excavation, and can be stored simultaneously through a 'defect adding mode' field (expert analysis, excavation verification and intelligent decision); for batch defects, the batch defects obtained through intelligent decision making are subjected to batch intelligent addition and positioning in the same mode, other defects can be imported through data to realize import of batch defect data, and the method specifically comprises the following steps: downloading a defect information template, editing defect information, importing the defect information into the system, manually inputting defect positioning aiming at each piece of defect data after the defect batch data is successfully imported, displaying newly-added defects in a three-dimensional model in a correlated manner, selecting the defect to be added and positioned, adding the defect on the three-dimensional pipeline through an adding button for defect positioning, and automatically updating the defect information to a three-dimensional defect point. The deletion of defects includes three cases: deleting the defect data at the position of the defect point, automatically updating the defect data after adding the defect position with the same defect number again, deleting the defect data and keeping the position of the three-dimensional defect point, automatically associating the position defect point information with the three-dimensional defect point again, and simultaneously deleting the defect data and the defect position on the three-dimensional model pipeline. The module can realize highlight positioning of the defect points in three dimensions through the defect point numbers. On the defect management page, through a page pull-down list, the year, the device, the pipeline number and the detection project are selected, and all defect points and defect information under the detection project can be checked.
The integrity intelligent evaluation and maintenance decision module mainly comprises an intelligent maintenance decision aiming at the pipeline abnormal signal which can not quantify the size and the type of the defect and an automatic evaluation and maintenance decision aiming at the quantified defect.
Aiming at the pipeline abnormal signals which can not quantify the size and the type of the defect, a detection signal diagram and a detection signal intensity value of the pipeline section are obtained from detection data management, a signal diagram or a signal data value with a standard length is determined according to the signal diagram and the signal intensity data of the pipeline respectively by adopting a pre-trained signal diagram or signal data integrity intelligent maintenance decision model
If signal diagram or signal value length
Scaling the signal diagram and the values to a standard size; if the signal length
Starting from the left end of the signal array, establish a length of
A sliding window of each
Length clipping the signal once until the right side of the sliding window reaches the boundary of the maximum value of the signal, obtaining
The entire signal graph or signal array is (rounded down). And respectively inputting the signal diagrams or the signal arrays into an intelligent maintenance decision model for prediction, and respectively outputting the mileage interval where the abnormal signal position is located and the maintenance grade of the abnormal signal (immediate repair, planned repair and monitoring use) according to the network prediction result. If the signals comprise ultrasonic guided waves and magnetic stress signals, the results of the ultrasonic guided waves and the magnetic stress signals can be simultaneously output (the mileage interval of the abnormal signals is the union of the ultrasonic guided waves and the magnetic stress signals, the maintenance decision level is the result of a higher level, the abnormal signal mileage interval of the output pipeline and the corresponding maintenance decision result (defect type and maintenance level)) or the results of the ultrasonic guided waves and the magnetic stress signals are subjected to data fusion according to weights and then output. And then, automatically marking the abnormal interval on the pipe section in a red mode in the three-dimensional platform by the background according to the output abnormal mileage interval, automatically synchronizing the maintenance decision result to the abnormal interval of the pipe section, simultaneously storing the evaluation result, and displaying the evaluation result by entering an intelligent evaluation interface next time.
In this embodiment, it should be noted that, when the network model is applied, the length of the signal graph or the signal value is the same as that of the signal graph
Scaling the signal diagram and the values to a standard size; if the signal length
Starting from the left end of the signal array, a length of
A sliding window of each
Length clipping the signal once until the right side of the sliding window reaches the boundary of the maximum value of the signal, obtaining
The entire signal diagram or signal array is rounded down, thereby ensuring that the potential fault area can be detected by the network at least once completely and reducing the distortion of the data signal, so as to improve the network accuracy.
According to the automatic evaluation and maintenance decision of the quantified defects, batch and single defects of information such as size and type are obtained through detection, different defect type defect evaluation methods are different, and after a user selects an evaluation method for a certain type of defects respectively, the system can call a built-in evaluation algorithm to perform batch evaluation, single evaluation and maintenance decision of the same evaluation method on defect points or a certain defect of a uniform type. And after the system is successfully evaluated, the evaluation result is stored, and the evaluation page is entered again for displaying the evaluation result next time. The single defect automatic evaluation is to select a defect point to be evaluated in a single evaluation module according to the defect point with known defect type and defect size, select an evaluation method, automatically evaluate the defect point, store a defect evaluation result for the evaluated defect point, display the evaluation result when entering again, and select the evaluation method to re-evaluate; the automatic evaluation of the batch defects is to select an evaluation method for defect points needing to be evaluated after a selection device, a pipeline number, a detection project and a defect type are carried out on a batch evaluation page, to input information required for evaluation, to carry out batch evaluation on a certain type of defects through a system built-in evaluation algorithm, and to modify the evaluation method for the evaluated defect points for re-evaluation.
When excavation data is not available in the previous period, the intelligent maintenance decision model can be trained and tested through expert analysis data in the defect data submodule.
In this embodiment, the training process of the intelligent maintenance decision model based on the detection signal diagram includes:
acquiring a sample magnetic stress picture from expert analysis data in a defect management submodule; wherein, various defect pixel positions and various maintenance grades need to be covered in the sample magnetic stress picture;
acquiring the mileage position and the maintenance grade of the abnormal area of the pipeline of the sample magnetic stress picture from expert analysis data in a defect management submodule;
taking a sample magnetic stress picture with a predetermined defect pixel position and a predetermined maintenance grade as the input of a model, taking the mileage position and the maintenance grade of the abnormal area of the pipeline in the sample magnetic stress picture as the output of the model, and training and testing a neural network model based on an intelligent learning algorithm to obtain an intelligent maintenance decision model;
therefore, the embodiment is based on the magnetic induction signal picture data, and carries out an intelligent maintenance decision which is strong in adaptability and not influenced by artificial subjective factors, so that the analytic accuracy of the detection result can be effectively improved, blind maintenance is avoided, the economy can be greatly improved, the labor force is reduced, and the safety of the pipeline is improved.
Based on the content of the above embodiment, in this embodiment, after the sample magnetic stress picture is obtained from the expert analytic data in the defect management submodule, manual and intelligent picture processing is further included. Wherein, the manual picture processing process comprises the following steps:
carrying out unified scale adjustment on the sample magnetic stress picture, and specifically comprising:
firstly, counting the detection mileage X of each detected pipe section picture max Average detected mileage
Pixel length L src And average original pixel length
Average width
Rounded to an integer;
to ensure that the mileage detection ratio of all pictures is uniform, the pixel length L of all pictures is first determined src Adjusting the length of the sample picture after adjustment
Counting the length L of the picture 0 Pixel average value of
Rounding to an integer, and then performing the next zoom adjustment on the picture:
case 1: for detecting mileage
The original pixel length L of the sample magnetic stress picture 0 Scaled to a length of
Original width W 0 Is scaled to
Obtaining the coordinates of the pixel points of the zoomed picture in a coordinate interpolation mode:
wherein the content of the first and second substances,
to transform the coordinates of the pixel points, L 0 ,W 0 The length and width of the sample image;
the coordinates (x, y) of the transformed pixel point belong to a pixel point coordinate set P:
wherein, P ij ＝(x i ,y j )；i＝1,2,…,m,j＝1,2,…,n；
When the pixel point coordinates (x, y) are integers, the pixel value f (x, y) of the processed point is the same as the pixel value of the corresponding point in the original image; when the pixel point coordinates (x, y) are not integers, performing bilinear interpolation on the pixels of four points surrounding the point in the original image, and then rounding to obtain a pixel value f (x, y) of the processed point;
case 2: for detecting mileage
Copying and splicing the sample magnetic stress picture until the sum of the detection mileage of the spliced picture is within the range
Within the interval, then executing case 1;
case 3: for detecting mileage
The sample magnetic stress picture is split intelligently through manual splitting or through a computer, and the manual splitting principle is as follows: ensure the split picture detection mileage to be in
And the abnormal signal area is not split into two pictures, and the condition 1 is executed on the split pictures.
Wherein, carry out intelligent split process through the computer and include:
processing case 3 for detection mileage
The original magnetic stress picture of the abnormal-free area is divided from the left end point and the right end point of the original magnetic stress picture to the opposite side respectively
Corresponding pixel length
Obtain two pixel lengths of
The target magnetic stress picture of (1).
Processing case 4 for detection mileage
The original magnetic stress picture of the abnormal-free area starts from the left end point of the original magnetic stress picture according to the pixel length
Cutting, and taking out the mileage of the remaining part
Finishing cutting, wherein the obtained target magnetic stress pictures have the pixel length of
The remaining portions continue to be processed in the manner described above for cases 1-2, and above for case 3.
Processing case 5 for detection mileage
The original magnetic stress picture of the signal with the abnormal area has a pixel abscissa of [0, L]Then the abnormal region set G is expressed as
G＝{[g i ,g i +D i ],i＝1,2,…,q}
Wherein q is the number of abnormal regions in an original magnetic stress picture, g i Is the origin of the abscissa pixel of the i-th anomaly, D i Is the length of the pixel for the i-th anomaly;
for each abnormal region, dividing the abnormal region from the start point and the end point to two sides respectively, if
Or
If the interval completely contains other abnormal regions, merging the abnormal regions contained in the original magnetic stress picture, specifically comprising:
when abnormal area
Combining the abnormal region with the same target magnetic stress picture and cutting, and the same way is adopted
Combining the abnormal regions into the same target magnetic stress picture and cutting, wherein the cut target magnetic stress picture has the pixel length of
The standard target magnetic stress picture of (1); if it is
Or
The section does not completely contain other abnormal regions, and the abnormal region [ g ] i ,g i +D i ]Respectively extend to both sides to obtain a length of
Interval of (2)
If the interval contains the end points of other abnormal areas, the end point is used as a cutting point and extends to the other side for cutting, and the length of the target magnetic stress picture containing the abnormal area is ensured to be equal to
Wherein the incomplete inclusion comprises no or half inclusion;
for abnormal area
The part of abnormal region is cut into a target magnetic stress picture, and then the residual target magnetic stress picture without abnormal region after cutting is subjected to the cycle execution of the case 1, the case 2, the processing case 3 and the processing case 4.
When the pixel point coordinates (x, y) are not integers, performing interpolation and rounding on pixels of four points surrounding the point in the original image to obtain a pixel value f (x, y) of the point after processing, wherein the method comprises the following steps:
the bilinear interpolation is:
wherein Q 11 ，Q 12 ，Q 21 ，Q 22 Four pixel points of upper, lower, left and right surrounding the point in the original picture, the coordinates of the pixel points are (x) 1 ,y 1 )，(x 1 ,y 2 )，(x 2 ,y 1 )，(x 2 ,y 2 ) The pixel values of the original picture are f (Q) 11 )，f(Q 12 )，f(Q 21 )，f(Q 22 )。
It can be understood that the magnetic induction detection signal diagram is subjected to unified scale adjustment, so that the training process of the intelligent maintenance decision model based on the detection signal diagram is more accurate and effective, and the problem that the obtained magnetic stress image cannot be directly utilized to carry out intelligent identification of the pipeline maintenance grade based on the neural network due to different measured mileage and different abscissa scales of the obtained magnetic stress image at each time can be solved.
In the embodiment, the problem that the obtained magnetic stress picture cannot be directly utilized to carry out intelligent identification of the pipeline maintenance level based on the neural network due to the fact that the abscissa scale of the magnetic stress picture obtained each time is different because the mileage measured each time is different can be solved by determining the relation between the detection mileage of the magnetic stress picture and the average detection mileage of the magnetic stress picture of the detected pipe section and then carrying out unified scale adjustment on the magnetic stress picture.
And after the excavation data are accumulated to a certain degree, excavation defect data can be obtained from the defect management submodule, and the intelligent maintenance decision model is optimized through data migration.
Based on the content of the foregoing embodiment, in this embodiment, when making an intelligent maintenance decision for a pipeline, the method further includes: adjusting and transferring the intelligent maintenance decision model;
wherein, the process of adjusting and migrating the intelligent maintenance decision model comprises:
after the pipeline excavation data are accumulated, performing thermal One-Hot coding or continuous coding on the ambient environmental factors influencing magnetic induction, including metal, tree and pipeline attributes to obtain a first feature vector; convolving the target region selected by the region generating network RPNSplicing the corresponding feature vector on the feature map obtained by the neural network with the first feature vector to obtain a second feature vector; calculating the second characteristic vector through a full-connection neural network layer and then calculating through a Softmax function to obtain the probability L that the sample magnetic stress picture belongs to each maintenance grade i Training a fully-connected network by using a random gradient descent (SGD) method, and outputting L by the network i Among the results, the maximum L i Where is the most likely repair level for the anomaly;
wherein the probability L that a signal sample belongs to each maintenance class i The calculation method is as follows:
wherein p is i Is the output of the fully connected network layer and N represents the number of repair levels.
In the embodiment, through the processing, the maintenance grade finally predicted by the intelligent maintenance decision model is more accurate, and the referential significance is larger, so that the problems that due to the fact that the personnel level difference is large and the workload is huge, the accuracy of manual troubleshooting interference signals is low, the working efficiency is low, misjudgment and missing judgment of the signals cause blind maintenance of enterprises, and economic loss and social influence are caused due to shutdown are solved.
Based on the content of the foregoing embodiment, in this embodiment, when making an intelligent maintenance decision for a pipeline, the method further includes: optimizing the intelligent maintenance decision model to enable the intelligent maintenance decision model to output a defect type;
adding a layer of multi-label Sigmod loss function behind a full connection layer of the network to obtain the probability of predicting the sample magnetic stress picture into each type of defect
Wherein the content of the first and second substances,
the column vector output for the last fully connected layer of the network, M being the number of defect types,
adopting cross entropy loss function, using random gradient descent method SGD training network, network output
All defect types corresponding to positions larger than the threshold value in the vector;
after the above processing is completed, the network outputs the defective pixel position, the defect type, and the corresponding maintenance grade in prediction.
In this embodiment, after the data is further accumulated, the output of the network may also migrate to a more accurate defect type, such as a recess, metal corrosion, weld defect, or the like.
According to the two embodiments, the accumulated actual defect maintenance decision results and the corresponding data signals are substituted into the training module to carry out model tuning and further accurately predict the defect type by collecting the defect data of pipeline excavation, the surrounding environment data and the maintenance decision data obtained through an empirical formula.
It can be understood that, the defect data verified by pipeline excavation can be further used, the maintenance decision of the excavated defects is carried out by adopting a standard method, the accumulated actual defect maintenance decision results and the corresponding picture signals are substituted into the training module for tuning, and when the data amount is accumulated to a certain degree, the model is directly migrated, so that the precision of the network is further improved through the actual data. In addition, the model can be optimized by adding a layer of multi-label Sigmod loss function, so that the network can migrate to more accurate defect types, such as pits, metal corrosion, weld defects and the like.
The training and optimization process of the above model is explained below by specific examples:
firstly, normalizing the picture into vectors with the same size according to pixel values, and extracting features of the vectors through a gray VGG convolutional neural network to obtain feature map vectors. The feature map is extracted by a convolution header, then sent to an RPN (Region-generated Network) to obtain a target Region, and then the target Region is predicted.
According to the position of an image abnormal region (which can be understood as a defective pixel position) marked manually and the type of pipeline signal abnormality (which can be understood as a maintenance level), the proportion of signal abnormality and normal quantity is counted, the existing data is randomly divided into a training set and a testing set according to the proportion value, a network model is trained by the training set, wherein the input of the model is a normalized image, the output of the model is the defective pixel position and the maintenance level, and the maintenance level of the magnetic stress abnormality signal is divided into immediate repair, planned repair and monitoring use of 3 conditions. And calculating the loss function of the sample in the network. In order to reduce the loss function, a SGD (Stochastic Gradient Descent) method is adopted to update network parameters and train a target detection network, and in order to accelerate the network training speed, only an anchor of one size can be selected according to the shape characteristics of a magnetic induction signal image in the aspect of anchor selection, wherein the aspect ratio is 1: 2. and testing the network effect by adopting the test set. After further data acquisition and accumulation, in order to improve the precision of the network, newly added data are substituted into the training module for tuning, so that the network detection accuracy is further improved.
After the signal picture data and the pipeline excavation data are further accumulated, One-Hot encoding can be carried out on pipeline influencing factors such as variables of metal, trees, pipeline attributes and the like, or continuous numerical type encoding is carried out vectorization, namely, factors possibly influencing magnetic stress such as trees, electric wires, pipeline attribute data and the like around the pipeline are counted during excavation, the data of the factors are recorded, and a new feature vector is obtained. If the pipeline attribute data (pipeline material, pipe diameter and wall thickness) are classified to form One-Hot codes and continuous codes, the form is as follows:
wherein m represents the pipe type, n represents the pipe diameter type, and r represents different thickness values of the pipeline. Splicing the characteristic diagram of the special signal area selected by the RPN with the characteristic vector of the surrounding environment factors, calculating the spliced characteristic vector through a full-connection neural network layer, and calculating through a Softmax function to obtain the probability L that the signal sample belongs to each maintenance level i Training the network, network output L, also using the SGD method i Among the results, the maximum L i The site is the most likely repair level for the site anomaly. Probability L of signal sample belonging to each maintenance class i The calculation is as follows:
in the formula, p i Is the output of the fully connected network layer, and N represents the number of maintenance levels, such as 3 levels, immediate repair, planned repair, monitoring use.
After the data is further accumulated, the output of the network can also migrate to more accurate defect types, such as pits, metal corrosion, weld defects, and the like. Adding a layer of multi-label Sigmod loss function behind a full connection layer of the network to obtain the probability of predicting the signal sample as each type of defect
Wherein the content of the first and second substances,
the column vector output by the last full connection layer of the network, M is the number of abnormal types,
training network by adopting cross entropy loss function and SGD method and network output
All locations in the vector that are larger than a threshold (typically 0.5) correspond to a defect type.
After the steps are completed, the network can output the defect type and the corresponding maintenance grade during prediction, such as 'corrosion defect, monitoring and using'.
In this embodiment, the training process of the intelligent maintenance decision model based on the detected signal strength value includes:
acquiring a sample magnetic stress detection signal intensity value from expert analysis data in a defect management submodule; wherein, the sample magnetic stress detection signal intensity value needs to cover various defect pixel positions and various maintenance grades;
acquiring the position and the maintenance grade of the abnormal area of the pipeline corresponding to the strength value of the sample magnetic stress detection signal from expert analysis data in a defect management submodule; wherein, the sample data is divided into a training set and a test set;
the method comprises the steps of taking a sample magnetic stress detection signal intensity value with a predetermined defect pixel position and a predetermined maintenance grade as an input of a model, taking a pipeline abnormal region mileage position and a maintenance grade corresponding to the sample magnetic stress detection signal intensity value as an output of the model, and training and testing a neural network model based on an intelligent learning algorithm to obtain an intelligent maintenance decision model.
Therefore, the embodiment carries out an intelligent maintenance decision which is strong in adaptability and not influenced by artificial subjective factors based on magnetic induction detection signal intensity data, so that the analytic accuracy of the detection result can be effectively improved, blind maintenance is avoided, the economy can be greatly improved, the labor force is reduced, and the safety of the pipeline is improved.
Here, it should be noted that the data of the pipeline detection signal is generally a data pattern of signal strength corresponding to the pipeline mileage, and such data is usually displayed and analyzed in the form of a signal diagram. For example, a corresponding magnetic stress signal picture can be drawn according to the obtained intensity value of the magnetic stress signal, and then the severity of the pipeline defect can be predicted by adopting an intelligent algorithm according to the signal picture. However, since the digital signal loses a part of precision when being converted into a picture, the accuracy of the algorithm has a certain limitation, and processing the computed picture requires a large amount of computing resources, the computing speed is slow, and the requirement on computer hardware is very high. Therefore, the embodiment provides an intelligent pipeline maintenance decision-making mode directly aiming at data signals, the mode is not only suitable for intelligent evaluation of integrity of signal data pipelines such as magnetic stress detection in pipeline non-contact detection, but also can be used for direct intelligent evaluation of other types of detection data, on one hand, the mode has a strong applicable space, on the other hand, the labor cost is reduced, the economic benefit and the pipeline safety are improved while the evaluation accuracy and the evaluation efficiency of the integrity of the pipeline are improved, and the technical blank of the existing intelligent evaluation of the integrity of the pipeline is filled.
Based on the content of the foregoing embodiment, in this embodiment, after obtaining the sample magnetic stress detection signal strength value from the expert analytic data in the defect management sub-module, the method further includes the following steps:
carry out length unified processing to sample magnetic stress detection signal intensity numerical value, specifically include:
the magnetic induction detection signal of each section of pipeline is usually a two-dimensional array, and the form is as follows:
wherein m is the number of sampling times, and the magnetic induction detection signal obtained at each sampling point:
wherein n is the magnetic induction detection signal dimension, h ij A value representing a jth signal characteristic of the magnetic induction detection signal;
sampling frequency of the magnetic induction detection signal data is uniform, the length of H represents the length of the detected pipe section, and the magnetic induction detection signal data is subjected to standardization processing to enable the length of each section of detection signal vector to be the same, so that network training is facilitated;
firstly, the average length of magnetic induction detection signal data of all detected pipe sections is counted
Rounded to an integer, then the following cases 1 to 5 are performed:
case 1: for array length
By interpolating to scale the array length to
Specifically, the position of the new array of sampling points in the original numerical coordinates is obtained:
in the formula, X is a position information set of the new array after transformation;
calculating the detection signal at the new position by linear interpolation based on the new position information X
The interpolation method is as follows:
wherein j ═ i]Is the integer part of the number i,
new magnetic induction detection signal data H obtained for the magnetic induction detection signal data of the corresponding position in the original data * Has a vector length of
Case 2: for array length
Signal data of, pair
The array is spliced to obtain a new signal:
the signal length becomes N x l until
Then, as in case 1, the length is scaled to
Case 3: for array length
The abnormal signal-free data are divided from the initial position and the end position of the array to the opposite side respectively
Length, two new data are obtained, namely:
case 4: for array length
Of abnormal-free signal data H, slave array
Starting point position is pressed
The length is cut to obtain the length of
Until the length of the remaining part
Then changing the length of the residual part into the length of the residual part according to the method of the case 1-the case 3
Case 5: for array length
Has abnormal signal data H, the abnormal region set G is expressed as
G＝{[g i ,g i +D i ],i＝1,2,…,q}
Wherein q is the number of abnormal regions in H, g i Is the starting position of the i-th abnormality, D i For the position length of the i-th abnormality, first, each abnormal region is divided into two regions from the start and end points of the abnormal signal
Length if
Or
And if the interval completely contains other signal abnormal intervals, combining the abnormal intervals and clipping, specifically: when the temperature is higher than the set temperature
When the abnormal regions are combined into one signal and cut off, the same principle is used
Then, the abnormal regions are merged into the same signal and cut, and the length of the array after cutting is
If it is
Or
The interval does not completely contain other abnormal signals, i.e. the abnormal signal [ g i ,g u +D u ]Respectively extend to both sides to obtain a length of
Interval of (2)
If the interval contains the end points of other abnormal signals, the end point is used as a cutting point and then extends to the other side for cutting, and the length of the part containing the signal abnormality is ensured to be
To include an endpoint g u-1 +D i-1 For example, a clipping interval is obtained
If the right end point is in other abnormal interval, then the left side is contracted to the abnormal position to obtain the final cutting interval g i-1 +D i-1 ,g i+1 ]Repeating case 1-case 2 for the final clipped signal array; for the
The abnormal signal area of the part is directly cut into a signal array, and the rest part is processed by the steps; and circularly executing the cases 1-4 for the no-signal area remained after the clipping.
It can be understood that the magnetic induction detection signal data are subjected to unified scale adjustment, so that the training process of the intelligent maintenance decision model based on the detection signal intensity value is more accurate and effective, and the problem that the pipeline maintenance grade intelligent identification based on the neural network cannot be directly carried out by utilizing the obtained magnetic induction detection signal data because the abscissa scales of the magnetic induction detection signal data obtained at each time are different due to different measuring mileage at each time can be solved. In this embodiment, by first performing a uniform scale adjustment on the magnetic induction detection signal data according to the relationship between the array length of the magnetic induction detection signal data and the average array length of the detected pipe section, the problem that the obtained magnetic induction detection signal data cannot be directly utilized for intelligent identification of the pipeline maintenance level based on the neural network due to different scales of the magnetic induction detection signal data obtained at each time due to different measurement mileage can be solved.
The above training and optimization process of the data signal-based model is described in detail below by way of specific examples:
and (3) regarding the magnetic induction detection signal data as a two-dimensional array, vectorizing the two-dimensional array, and extracting features of the vector through a gray VGG convolutional neural network to obtain a feature map vector. And (4) extracting the feature map through a convolution header, sending the feature map into an RPN (Region pro-apparent Network) to obtain a target Region, and predicting the target Region.
According to the position of the abnormal area of the pipeline and the maintenance grade of the abnormal signal corresponding to the magnetic induction detection signal data (the maintenance grade of the magnetic induction abnormal signal is divided into immediate repair, planned repair and monitoring use 3 conditions) determined by a signal analysis expert of a detection company by experience or excavation, the proportion of the abnormal number and the normal number of the signal is counted, the existing data is randomly divided into a training set and a testing set according to the proportion value, a network model is trained by the training set, wherein the input of the model is normalized signal data, and the output of the model is the starting position and the ending position (namely the column number of a two-dimensional array) of the defect position on the original signal data and the corresponding maintenance grade. And calculating the loss function of the sample in the network. In order to reduce the loss function, an SGD (Stochastic Gradient Descent) method is adopted to update network parameters and train a target detection network, and in order to accelerate the network training speed, only an anchor of one size can be selected according to the characteristics of magnetic induction signals in the selection of the anchor, and the length-width ratio is 1: 2. and testing the network effect by adopting the test set. After further data acquisition and accumulation, in order to improve the precision of the network, newly added data are substituted into the training module for tuning, so that the network detection accuracy is further improved.
When the network model is applied, for a section of magnetic induction signal, if the signal length
Scaling the signal array to a standard size according to the method of part 1; if the signal length
Ensuring that the potentially failing region can be detected by the network completely at least once and reducing distortion of the data signal to improve network accuracyAnd (4) degree. At this point no further scaling is performed but starting from the left end of the signal array, a length of is established
A sliding window of each
Length clipping the signal once until the right side of the sliding window reaches the boundary of the signal array to obtain
(round down) the entire signal array. And inputting the signal arrays into a network for prediction, and outputting the interval of the abnormal signal position on the data and the maintenance grade of the abnormal signal according to the network prediction result.
After the signal data and the pipeline excavation data are further accumulated, One-Hot encoding can be carried out on pipeline influencing factors such as variables of metal, trees, pipeline attributes and the like, or continuous numerical type encoding is carried out vectorization, namely, factors possibly influencing magnetic induction such as trees, electric wires, pipeline attribute data and the like around the pipeline are counted during excavation, data of the factors are recorded, and a new feature vector is obtained. If the pipeline attribute data (pipeline material, pipe diameter and wall thickness) are classified to form One-Hot codes and continuous codes, the form is as follows:
wherein m represents the pipe type, n represents the pipe diameter type, and r represents different thickness values of the pipe. Splicing the characteristic diagram of the special signal area selected by the RPN with the characteristic vector of the surrounding environment factors, calculating the spliced characteristic vector through a full-connection neural network layer, and calculating through a Softmax function to obtain the probability L that the signal sample belongs to each maintenance level i Training the network, network output L, also using the SGD method i Among the results, the maximum L i The site is the most likely repair level for the site anomaly. The signal samples belonging to each dimensionProbability of class repair L i The calculation is as follows:
in the formula, p i Is the output of the fully connected network layer, and N represents the number of maintenance levels, such as 3 levels, immediate repair, planned repair, monitoring use.
After the data is further accumulated, the output of the network can also migrate to more accurate defect types, such as pits, metal corrosion, weld types, and the like. Adding a layer of multi-label Sigmod loss function behind a full connection layer of the network to obtain the probability of predicting the signal sample as each type of defect
Wherein the content of the first and second substances,
the column vector output by the last full connection layer of the network, M is the number of abnormal types,
adopt cross entropy loss function, SGD method training network, network output
All locations in the vector that are larger than a threshold (typically 0.5) correspond to defect types.
After the steps are completed, the network can output the defect types and corresponding maintenance grades such as 'weld defects, monitoring and using' during prediction.
Therefore, in order to solve the technical blind point of the existing method, the embodiment provides the intelligent evaluation method for the integrity of the pipeline based on the data signals. Specifically, the technical solution adopted in this embodiment includes: collecting magnetic induction detection signal data, abnormal positions in the signal data and maintenance decision levels of the abnormal positions determined by experts according to experience or excavation; taking abnormal points in the classified signal data with the maintenance grade as labels; constructing a neural network model, carrying out standardized preprocessing on the collected data to a uniform size, then randomly dividing the data into a training set and a test set in proportion, training the model through the training set data, testing the model through the test set data, and outputting an intelligent integrity evaluation result; and further acquiring defect data and surrounding environment data of pipeline excavation and maintenance decision data obtained through an empirical formula, substituting the accumulated intelligent evaluation result of the integrity of the actual defects and the corresponding data signals into a training module to perform model tuning and further accurately predicting the types of the defects.
Therefore, the intelligent evaluation method based on the magnetic induction detection signal data can objectively carry out intelligent evaluation on the integrity of the pipeline, so that the analysis accuracy of the detection result can be improved, blind maintenance is avoided, the economy and the working efficiency are greatly improved, and the labor force is reduced. In the embodiment, the magnetic induction detection signal data accumulated historically is preprocessed, the target detection convolutional neural network is trained to make a decision on the processing result, and the ambient environment factors are added during the decision making so as to continuously improve the identification accuracy.
In the present disclosure, reference to the description of "one embodiment," "some embodiments," "an example," "a specific example," or "some examples" or the like means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, and not to limit it; although the present invention 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 technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. An intelligent evaluation visualization system for pipeline integrity, comprising: the system comprises a three-dimensional visualization module and a pipeline intelligent evaluation module;
the three-dimensional visualization module comprises a three-dimensional model visualization display submodule, a supporting environment submodule, an evaluation module data interface submodule, a pipeline information directory tree submodule and a pipeline engineering data management submodule;
the intelligent pipeline evaluation module comprises a basic data management submodule, a detection management submodule, a defect management submodule and an intelligent integrity evaluation and maintenance decision submodule;
the evaluation module data interface submodule is used for providing data interfaces for the basic data management submodule, the detection management submodule, the defect management submodule and the integrity intelligent evaluation and maintenance decision submodule;
the integrity intelligent evaluation and maintenance decision submodule comprises an intelligent maintenance decision for the pipeline abnormal signals of which the sizes and types of the defects can not be quantized and an automatic evaluation and maintenance decision for the quantized defects, and the intelligent maintenance decision is carried out by adopting an intelligent maintenance decision model based on a signal diagram or signal data which is trained in advance through deep learning and by acquiring a detection signal diagram and a detection signal intensity value of the pipeline section from a detection management submodule for the pipeline abnormal signals of which the sizes and types of the defects can not be quantized;
aiming at automatic evaluation and maintenance decision of quantitative defects, batch and single defects of size and type information are obtained through detection, different evaluation methods are adopted for different defect types, and after the evaluation methods are selected for a certain type of defects, a built-in evaluation algorithm is called to perform batch evaluation, single evaluation and maintenance decision of the same evaluation method on defect points or a certain defect of a uniform type; and the three-dimensional model visual display submodule performs three-dimensional dynamic interactive visual display on the intelligent evaluation and maintenance decision results.
2. The intelligent evaluation visualization system for the integrity of the pipeline according to claim 1, wherein the three-dimensional visualization module is based on the principle of human-computer interaction, and is used for surveying the equipment and the device of the refinery, the above-ground marker and the above-ground and underground pipeline objects connected with the equipment and the device in the field, determining the geographic information or the relative position of the modeling object, establishing a model of the object in the three-dimensional model, and adding, checking, deleting and modifying the established engineering information related to the pipeline, the device and the equipment through the three-dimensional model or a directory tree, wherein the information of the pipeline is used as the data base for the integrity evaluation of the buried pipeline.
3. The intelligent evaluation visualization system for the integrity of the pipeline according to claim 1, wherein the basic data management submodule is used for acquiring basic information of the pipeline where the pipeline is located from the three-dimensional visualization module, and is used for increasing, checking, deleting and modifying information of pipe diameter, wall thickness, pipe material, conveying medium and connecting equipment.
4. The visualization system for intelligent evaluation of pipeline integrity according to claim 1, wherein the detection management submodule comprises three functions of detected pipeline management, detection engineering management and detection data management, and is used for logical management of detection operation, detection signals and detection results;
wherein, the detected pipeline management: the method is used for managing the detection pipeline and the related detection engineering thereof, the detection pipeline manages the detection range of the detected pipeline, and on the basis of the built-in pipeline number naming rule, the pipeline number of the newly-added detection pipeline is automatically generated based on the pipeline where the detected pipeline is located, and the information of the newly-added detection pipeline comprises the following steps: detecting length, detecting and positioning, maintaining, modifying and checking remark information, and meanwhile, after a detection pipeline is newly built, highlighting and positioning the detected pipeline on the three-dimensional model through a pipeline number; in addition, the detection engineering and the related information of the detected pipeline can be added, checked, deleted and modified under the added detected pipeline; after the pipeline is detected for the first time, on the three-dimensional model of the pipeline, the detection range of the detection pipeline is determined by manual positioning through detecting marker information and clicking on the pipeline to be detected, a system automatically generates a detection pipeline number according to the pipeline where the detection pipeline is located, and meanwhile, according to the detection pipeline number, highlight positioning with the detected pipeline in the three-dimensional model is realized; maintaining and modifying the information of the detected pipeline, including detection length, detection positioning point, input and modification of design coefficient information and deletion of the detected pipeline; after the detected pipeline is deleted, the detection project, the position of the defect point and the defect data which are related to the detected pipeline are deleted at the same time;
detection engineering management: the method is used for establishing a detection project after one section of pipeline carries out detection operation, the detection project is newly established under the newly added detected pipeline, and maintenance management is carried out on the detection project and relevant information thereof, including detection project name, detection method, detection equipment, detection time and detection unit, such as addition, search, deletion and modification; detecting the position of a defect point associated with the project after the project is deleted, and deleting the defect data at the same time; the module also checks all detection project lists under the corresponding detected pipelines according to the serial numbers of the detected pipelines;
and (3) detection data management: the system is used for managing detection data of a certain detection project, and the functions comprise detection signal diagram management, detection signal data management and detection result data management; the detection signal diagram management is used for managing the detection signal diagram of the pipeline; the detection signal data is managed in a form, and the data is imported into the system in the form of the form for the strength value of the detection signal along the detection mileage according to the sampling result of the detector; the detection result data is managed in a form of a form import system; the method also comprises the steps of detecting signals, downloading a result template, importing data, and adding, searching, deleting and modifying the data.
5. The intelligent pipe integrity evaluation visualization system according to claim 1, wherein the defect management submodule is configured to manage defect data obtained by analyzing a signal, excavation verification, or an intelligent maintenance decision by an expert after a pipe is detected for one time; the defect data can interact with the three-dimensional model, so that the defect information of the detected pipeline can be increased, checked, deleted and changed, and meanwhile, the defect information is managed respectively according to different defect types, wherein the corrosion rate of corrosion defects is directly calculated according to the defect depth and the commissioning time through a background built-in algorithm;
wherein, newly-increased defects adopt different processing modes:
for single defects obtained by excavation verification and expert analysis, on the premise of known defect positioning, scale measurement is carried out on a three-dimensional pipeline model through markers on a detected pipeline, the positions of the defects are directly added, and then defect data are recorded;
for the defects obtained by intelligent maintenance decision, outputting the mileage position and the maintenance grade of the abnormal area of the pipeline through an intelligent maintenance decision model, automatically marking an abnormal interval on the pipeline section corresponding to the three-dimensional platform, automatically synchronizing the maintenance grade to the abnormal interval of the corresponding pipeline, and simultaneously storing the defects and the maintenance decision result to a defect management sub-module so as to enable the defects to enter an intelligent evaluation interface to display the maintenance decision result, wherein the defects obtained by intelligent decision can be analyzed by experts and verified by excavation, and are simultaneously stored through a field of a defect adding mode; wherein, the defect addition mode includes: expert analysis, excavation verification or intelligent decision making;
for batch defects, batch defects obtained through intelligent decision are subjected to batch intelligent addition and positioning in the same mode, other defects are imported through data, and the method is specifically operated as follows: downloading a defect information template, editing defect information, importing the defect information into a system, manually inputting defect positioning aiming at each piece of defect data after the defect batch data are successfully imported, displaying newly-added defects in a three-dimensional model in a correlated mode, or selecting defects to be added and positioned, adding the defects on the three-dimensional pipeline through an adding button for defect positioning, and automatically updating the defect information to a three-dimensional defect point.
6. The visualization system for intelligent evaluation of the integrity of pipelines according to claim 5, wherein the deletion of defects comprises three cases:
A. deleting the position of the defect point, reserving defect data, and automatically updating the defect data after adding the mileage position of the abnormal area of the pipeline with the same defect number again;
B. deleting the defect data and reserving the position of the three-dimensional defect point, and automatically associating the position defect point information to the three-dimensional defect point after newly adding the position defect point information;
C. deleting the defect data and the mileage position of the abnormal area of the pipeline on the three-dimensional model pipeline simultaneously;
the defect management submodule realizes highlight positioning of the defect point in three dimensions through the defect point number; on the defect management page, through a page pull-down list, the year, the device, the pipeline number and the detection project are selected, and all defect points and defect information under the detection project can be checked.
7. The intelligent pipe integrity evaluation visualization system according to claim 1, wherein the intelligent integrity evaluation and maintenance decision submodule comprises an intelligent maintenance decision for pipe anomaly signals that fail to quantify defect size and type, and an automatic evaluation and maintenance decision for quantified defects;
aiming at the pipeline abnormal signal which can not quantify the size and the type of the defect, the detection of the pipeline section is firstly obtained from the detection management submoduleThe method comprises the steps of determining a signal diagram or signal data value with standard length according to the signal diagram and the signal intensity data of a pipeline by respectively adopting a pre-trained intelligent maintenance decision model based on a detection signal diagram and an intelligent maintenance decision model based on a detection signal intensity value aiming at the detection signal diagram and the detection signal intensity value
If the signal diagram or signal value length
Scaling the signal diagram and the values to a standard size; if the signal length
Starting from the left end of the signal array, establish a length of
A sliding window of each
Length clipping the signal once until the right side of the sliding window reaches the boundary of the maximum value of the signal, obtaining
A signal graph or signal array; respectively inputting the signal diagrams or the signal arrays into an intelligent maintenance decision model for prediction, and respectively outputting the mileage interval where the abnormal signal position is located and the maintenance grade of the abnormal signal according to the prediction result;
if the signals comprise ultrasonic guided waves and magnetic stress signals, the results of the ultrasonic guided waves and the magnetic stress signals are simultaneously output, wherein the mileage interval of the abnormal signals is the union of the two results, the maintenance decision level is the result of a higher level, and the abnormal signal mileage interval of the pipeline and the corresponding maintenance decision result are output; or the results of the two are output after data fusion is carried out according to the weight;
and the background automatically marks the abnormal interval on the pipe section in a red form in the three-dimensional platform according to the output abnormal mileage interval, automatically synchronizes the maintenance decision result to the abnormal interval of the pipe section, simultaneously stores the evaluation result, and can display the evaluation result when entering the intelligent evaluation interface next time.
8. The intelligent visual pipe integrity evaluation system according to claim 7, wherein for automatic evaluation and maintenance decision of quantitative defects, batch and single defects of size and type information are obtained by detection, different defect type defect evaluation methods are different, and for a certain type of defects respectively, after the evaluation method is selected, the system can call a built-in evaluation algorithm to perform batch evaluation, single evaluation and maintenance decision of the same evaluation method on defect points of a uniform type or a certain defect;
after the system is successfully evaluated, the evaluation result is stored, and the evaluation page is entered for displaying the evaluation result next time; the single defect automatic evaluation is to select a defect point to be evaluated in a single evaluation module according to the defect point with known defect type and defect size, select an evaluation method, automatically evaluate the defect point, store a defect evaluation result for the evaluated defect point, display the evaluation result when entering again, and select the evaluation method to re-evaluate; the automatic evaluation of the batch defects is to evaluate a certain type of defects in batch through a built-in evaluation algorithm of a system after selecting a device, numbering pipelines, detecting engineering and defect types on a batch evaluation page and inputting information required for evaluation, and to modify and evaluate the evaluated defect points again.
9. The intelligent pipe integrity evaluation visualization system according to claim 7, wherein when excavation data is not available in the previous period, the intelligent maintenance decision model is trained and tested by expert analysis data in the defect management submodule;
and after the accumulation degree of the excavation data meets the preset condition, acquiring excavation defect data from the defect management submodule, and optimizing the intelligent maintenance decision model through data migration.
10. The system for visualizing intelligent evaluation of pipe integrity as recited in claim 9, wherein said training process of said intelligent maintenance decision model based on a detection signal map comprises:
acquiring a sample magnetic stress picture from expert analysis data in a defect management submodule; wherein, various defect pixel positions and various maintenance grades need to be covered in the sample magnetic stress picture;
acquiring the mileage position and the maintenance grade of the abnormal area of the pipeline of the sample magnetic stress picture from expert analysis data in a defect management submodule;
taking a sample magnetic stress picture with a predetermined defect pixel position and a predetermined maintenance grade as the input of a model, taking the mileage position and the maintenance grade of the abnormal area of the pipeline in the sample magnetic stress picture as the output of the model, and training and testing a neural network model based on an intelligent learning algorithm to obtain an intelligent maintenance decision model;
the training process of the intelligent maintenance decision model based on the detection signal strength value comprises the following steps:
acquiring a sample magnetic stress detection signal intensity value from expert analysis data in a defect management submodule; wherein, the sample magnetic stress detection signal intensity value needs to cover various defect pixel positions and various maintenance grades;
acquiring the position and the maintenance grade of the abnormal area of the pipeline corresponding to the strength value of the sample magnetic stress detection signal from expert analysis data in a defect management submodule; wherein, the sample data is divided into a training set and a testing set;
the method comprises the steps of taking a sample magnetic stress detection signal intensity value with a predetermined defect pixel position and a predetermined maintenance grade as an input of a model, taking a pipeline abnormal region mileage position and a maintenance grade corresponding to the sample magnetic stress detection signal intensity value as an output of the model, and training and testing a neural network model based on an intelligent learning algorithm to obtain an intelligent maintenance decision model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110252348.XA CN115034250A (en) | 2021-03-08 | 2021-03-08 | Intelligent evaluation visualization system for pipeline integrity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110252348.XA CN115034250A (en) | 2021-03-08 | 2021-03-08 | Intelligent evaluation visualization system for pipeline integrity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115034250A true CN115034250A (en) | 2022-09-09 |
Family
ID=83118179
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110252348.XA Pending CN115034250A (en) | 2021-03-08 | 2021-03-08 | Intelligent evaluation visualization system for pipeline integrity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115034250A (en) |
-
2021
- 2021-03-08 CN CN202110252348.XA patent/CN115034250A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111241611B (en) | Method for assisting foundation pit implementation | |
| JP6682021B2 (en) | Road collapse risk evaluation device, road collapse risk evaluation method, and computer program for road collapse risk evaluation | |
| JP5423983B2 (en) | Computer system and risk diagnosis method | |
| CN112579845A (en) | Industrial big data display geographic information system platform | |
| CN116090322A (en) | Intelligent decision visualization system for pipeline integrity | |
| CN112685820A (en) | Digital power station delivery method, medium and equipment based on BIM and GIS | |
| JP4601622B2 (en) | Long object management system and program | |
| CN117128448A (en) | LNG storage tank structure intelligent monitoring system | |
| Dayan et al. | A scoping review of information-modeling development in bridge management systems | |
| CN111626358A (en) | Tunnel surrounding rock grading method based on BIM picture recognition | |
| CN113780475B (en) | Mountain tunnel model fusion method based on GIS environment | |
| CN115063337A (en) | Intelligent maintenance decision-making method and device for buried pipeline | |
| KR102229423B1 (en) | Appearance survey network construction system and coded expression method with the characteristics of the appearance network damage, 3D position and size information as parameters grafted with the 3D BIM model | |
| CN115034250A (en) | Intelligent evaluation visualization system for pipeline integrity | |
| US10783680B2 (en) | Server-based management of robotic pipe inspection data and analysis reports | |
| KR102530194B1 (en) | Integration system for sewerage pipe asset management utilizing web applications | |
| CN116342059A (en) | Urban rail transit underground structure intelligent monitoring system based on Internet of things | |
| CN111538297A (en) | Web-based interactive three-dimensional visualization platform | |
| CN111208582A (en) | Underground pipe network system | |
| CN113762653A (en) | Underground pipeline integrity evaluation system and method | |
| KR20210076883A (en) | State analysis and risk analysis and modelling method of sewerage pipe | |
| CN111858808B (en) | Automatic evaluation method for feature and ground object precision of topographic map based on massive actual measurement points | |
| KR101585950B1 (en) | Drawing conversion automatic creation | |
| KR102530193B1 (en) | Integration system for waterworks pipe asset management utilizing web applications | |
| Gaafar et al. | Digital transformation solution for identification of geotechnical parameters using statistical data analysis |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240111 Address after: 100728 No. 22 North Main Street, Chaoyang District, Beijing, Chaoyangmen Applicant after: CHINA PETROLEUM & CHEMICAL Corp. Applicant after: Sinopec (Dalian) Petrochemical Research Institute Co.,Ltd. Address before: 100728 No. 22 North Main Street, Chaoyang District, Beijing, Chaoyangmen Applicant before: CHINA PETROLEUM & CHEMICAL Corp. Applicant before: DALIAN RESEARCH INSTITUTE OF PETROLEUM AND PETROCHEMICALS, SINOPEC Corp. |
|
| TA01 | Transfer of patent application right |