CN113803647B - Pipeline leakage detection method based on fusion of knowledge features and hybrid model - Google Patents
Pipeline leakage detection method based on fusion of knowledge features and hybrid model Download PDFInfo
- Publication number
- CN113803647B CN113803647B CN202110979611.5A CN202110979611A CN113803647B CN 113803647 B CN113803647 B CN 113803647B CN 202110979611 A CN202110979611 A CN 202110979611A CN 113803647 B CN113803647 B CN 113803647B
- Authority
- CN
- China
- Prior art keywords
- pipeline
- model
- flow
- knowledge
- pressure
- 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.)
- Active
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
- F17D5/02—Preventing, monitoring, or locating loss
- F17D5/06—Preventing, monitoring, or locating loss using electric or acoustic means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
Abstract
The invention discloses a pipeline leakage detection method based on knowledge feature and mixed model fusion, which comprises the following steps: firstly, establishing a pipeline hybrid modeling method for fusing a real-time transient mechanism model and a convolutional neural network data driving model so as to accurately predict the operation condition of a pipeline; secondly, aiming at the uncertainty commonly existing in the process, a knowledge feature model of the pipeline conveying process is embedded in the mixed model, so that the robustness of the model and the prediction capability of the pipeline running state are improved; and finally, analyzing the deviation obtained by comparing the predicted output of the mixed model with the actual measured value, and judging whether the pipeline leaks or not. The method improves the precision of the model and solves the problems of false alarm, missing alarm and the like caused by the change of working conditions and small leakage characteristics.
Description
Technical Field
The invention relates to the field of pipeline leakage detection, in particular to a pipeline leakage detection method based on fusion of knowledge features and a hybrid model.
Background
Pipelines are an important component of industrial, agricultural, and municipal infrastructure, but pipelines often suffer from a number of pipeline leakage problems due to aging, corrosion, weld defects, third party damage, and the like. Pipeline leak detection therefore takes into account the hydraulic characteristics of the pipeline transport, as well as the uncertainty of the process.
In daily life and industrial production, pipeline transportation plays a vital role. The pipeline transportation has the advantages of large transportation quantity, easy control of the transportation process, low price, simple construction, no influence of factors such as ground, climate and the like, and strong environmental adaptability. However, the problems of corrosion, aging, connection and the like of the pipeline and the influence of various external forces are common, and the problem of pipeline leakage is caused, so that the resource waste and huge economic loss are caused. It is therefore particularly important to realize detection of pipe leakage. For this reason, many methods of pipeline leak detection have been proposed in the industry and academia. Modern pipeline leakage detection methods are generally based on pipeline external or internal measured variables to build detection models so that pipeline leakage detection can be performed online in real time. External methods are typically based on detecting leakage characteristics outside the pipe, such as acoustic methods, optical fiber methods. The internal methods are typically based on operating parameters such as pipeline pressure, flow, temperature, etc., such as real-time transient modeling methods, negative pressure wave methods. From the viewpoint of model building, the method based on a mechanism model and a data driving model is two main methods for detecting pipeline leakage, and each method has advantages and disadvantages. Generally, the mechanism model-based approach relies on the generation and evaluation of biases. The performance of these methods is highly dependent on the model parameters and the accuracy of the sensors, and requires extensive simulation and calibration effort. The data driven method does not require any specific in-depth pipeline hydraulics knowledge, but only learns from collected historical data, plus some statistical or pattern recognition tools. Among them, the artificial neural network (Artificial Neural Network, ANN) and support vector machine (Support Vector Machine, SVM) are most applicable.
Disclosure of Invention
The invention provides a pipeline leakage detection method based on knowledge feature and hybrid model fusion.
Firstly, establishing a pipeline hybrid modeling method for fusing a real-time transient mechanism model and a convolutional neural network data driving model so as to accurately predict the operation condition of a pipeline; secondly, aiming at the uncertainty commonly existing in the process, a knowledge feature model of the pipeline conveying process is embedded in the mixed model, so that the robustness of the model and the prediction capability of the pipeline running state are improved; and finally, analyzing the deviation obtained by comparing the predicted output of the mixed model with the actual measured value to judge whether the pipeline leaks or not. The model obtained by the method has higher precision and better leakage detection effect.
The purpose of the invention is that: due to the fact that the operation condition of the pipeline is uncertain, the related parameters of the model are inaccurate and the data matching performance is poor in the actual process, prediction of the mechanism model is biased, and therefore pipeline leakage is misreported. The invention establishes a hybrid model combining a data driving model with a pipeline transient mechanism model, wherein a convolutional neural network with knowledge features is used as a pipeline data driving model to correct the pipeline transient model, thereby improving the precision of the model and solving the problems of false alarm, false alarm and the like caused by working condition change and small leakage features.
The technical scheme of the invention is as follows: the CNN is used as a data driving model to correct the real-time transient pipeline mechanism model, so that the accuracy of the model is improved, and the problems of false alarm, false alarm and the like caused by working condition change and small leakage characteristic are solved. As the running time of the pipeline is prolonged, the resistance structure in the pipeline is changed, the output of the mixed model is also biased, and a false alarm phenomenon is generated, so that the friction resistance knowledge of the pipeline is integrated into CNN, and the robustness of the model is enhanced.
The invention effectively combines the mechanism model and the data driving model, and on the basis, the knowledge features are fused to effectively and accurately predict the running state of the pipeline. Firstly, based on the hydraulic principle, a simplified real-time transient model (Real Time Transient Model, RTTM) mechanism model is established; secondly, extracting implicit characteristic information of pipeline operation data through a convolutional neural network (Convolutional Neural Networks, CNN) data driving model, and supplementing an RTTM mechanism model; in consideration of uncertainty of a process, a learning element of domain knowledge is introduced on the basis of a CNN data driving model, and a pipeline leakage detection method integrating data, knowledge and mechanism is provided.
A pipeline leakage detection method based on knowledge feature and mixed model fusion comprises the following steps:
step 4, establishing a data driving model fusing pipeline knowledge and a hybrid model based on mechanism and data driving;
step 5, inputting a convolutional neural network in a data driving model fused with pipeline knowledge into a boundary condition input value of a pipeline fluid real-time transient pipeline mechanism model, wherein the input value is the head end pressure and the tail end flow of the pipeline in operation, and the head end flow and the tail end pressure of the pipeline in operation can be obtained through model calculation in the step 2;
step 6, obtaining pipeline operation data with real-time performance through the mechanism and data-driven hybrid model in the step 4; when the pipeline leakage occurs, the head end flow and the tail end pressure of the pipeline are obtained after calculation in the step 5, the model calculation value is used as a model calculation value, the model calculation value is compared with the actual measurement value of the sensor, and when the deviation exceeds a given threshold value, the leakage is indicated to occur, and the method is expressed as:
wherein Q is 1 For actually measuring the flow of the head end of the pipelineThe magnitude of the force is calculated,p is the calculated value of the head-end flow 2 For the actual measurement of the line end pressure, < > about->For end pressure calculations, σ 1 、σ 2 The set flow and pressure thresholds, respectively.
In step 1, a pipeline fluid real-time transient pipeline mechanism model is established, which specifically comprises the following steps:
establishing a continuity equation and a motion momentum equation for the pipeline fluid motion:
wherein, the formula (1) is a flow continuity equation, the formula (2) is a motion momentum equation, and P is the average pressure of the section of the pipeline; v is the average flow velocity of the section of the pipeline; ρ is the fluid average density; f is the hydraulic friction coefficient; and (c) is a gravitational acceleration; θ is the angle of the fluid to the horizontal axis; d is the inner diameter of the tube; a is the pressure wave propagation velocity; t is time; x is the distance along the pipeline.
In the step 2, a characteristic line method is adopted to optimize the pipeline fluid real-time transient pipeline mechanism model in the step 1), and the method specifically comprises the following steps:
solving the nonlinear partial differential equation (1) and the equation (2) in the step 1 by adopting a characteristic line method; the characteristic line method needs to set boundary conditions and pipeline initial values, so that the pipeline steady-state model is considered, and the influence of time is not considered, and the characteristic line method can be obtained by the following formulas (1) and (2):
equations (4) and (5) are steady-state models of pipelines, are a normal differential equation set, and can be solved by adopting a fourth-order Dragon-Gregory tower method to determine initial values of a real-time transient pipeline mechanism model; and setting the input values of the boundary conditions as the head-end pressure and the tail-end flow by taking the head-end of the pipeline as the boundary conditions.
In the step 3, based on the basic principle of the pipeline, extracting knowledge features of equivalent friction coefficient in the pipeline, wherein the knowledge features specifically comprise;
based on the basic principle of the pipeline, the knowledge features of equivalent friction coefficient in the pipeline are extracted, and the method specifically comprises the following steps:
is obtainable by means of the pipeline steady-state model versions (4) and (5):
and (3) carrying out integral solution on equation sets (6) and (7) to obtain the relationship between the equivalent friction coefficient in the pipeline and the parameters in the pipeline:
wherein Q is 1 For the flow of the head end of the pipeline, Q 2 For the end flow of the pipeline, P 1 P is the pressure of the head end of the pipeline 2 For the end pressure of the pipe,for calculating the equivalent friction coefficient according to the head-end flow, < >>The equivalent friction coefficient is calculated according to the end flow.
In step 4, a data driving model fusing pipeline knowledge and a hybrid model based on mechanism and data driving are established, and the method specifically comprises the following steps:
fusing the knowledge features of the equivalent friction coefficient in the pipeline extracted in the step 3 into a one-dimensional convolutional neural network (1 DCNN), and obtaining a data driving model fused with knowledge through a full-connection layer after combining;
the mechanism model is shown in formulas (4) and (5), and the knowledge-fused data-driven model corrects the mechanism model to obtain a hybrid model based on mechanism and data driving.
Compared with the prior art, the invention has the following advantages:
according to the invention, CNN is used as a data driving model, and the real-time transient pipeline mechanism model is corrected, so that the accuracy of the model is improved, and the problems of false alarm, false alarm and the like caused by working condition change and small leakage characteristic are solved. As the running time of the pipeline is prolonged, the resistance structure in the pipeline is changed, the output of the mixed model is also biased, and a false alarm phenomenon is generated, so that the friction resistance knowledge of the pipeline is integrated into CNN, and the robustness of the model is enhanced. According to the invention, pipeline knowledge is integrated into the mixed model, and the accuracy of the corrected real-time transient pipeline mixed model is not greatly reduced along with the increase of the pipeline running time and the change of the resistance structure in the pipeline. The method improves the precision of the model and solves the problems of false alarm, missing alarm and the like caused by the change of working conditions and small leakage characteristics.
Drawings
The accompanying drawings are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention, without limitation to the invention. In the drawings:
fig. 1 is a schematic diagram of the experimental pipeline of the present invention, wherein upstream reservoir is an upstream water storage container, downstreamreservoir is a downstream water storage container, controller is a controller, pressure flow measurement is a pressure and flow measurement, leak point is a leak point water recovery pond is a water recovery tank.
FIG. 2 is a diagram of a real-time transient pipeline hybrid model incorporating a convolutional neural network.
Fig. 3 is a flow chart of the present invention.
FIG. 4 is a graph of RTTM and CNN-RTTM versus actual pressure for varying pipeline inlet pressures.
FIG. 5 is a graph of CNN-R and K-CNN-R versus actual pressure for varying the pipeline inlet pressure and coefficient of friction.
Detailed Description
The following will describe embodiments of the present invention in detail with reference to the drawings and examples, thereby solving the technical problems by applying technical means to the present invention, and realizing the technical effects can be fully understood and implemented accordingly.
As shown in fig. 1 to 3, a pipeline leakage detection method based on knowledge feature and mixed model fusion comprises the following steps:
and step 1, establishing a pipeline fluid real-time transient pipeline mechanism model.
Establishing a continuity equation, a motion momentum equation and an energy equation for the pipeline fluid motion:
wherein, the formula (1) is a flow continuity equation, the formula (2) is a motion momentum equation, the formula (3) is an energy equation, and P is the average pressure of a pipeline section; v is the average flow velocity of the section of the pipeline; ρ is the fluid average densityA degree; f is the hydraulic friction coefficient; and (c) is a gravitational acceleration; θ is the angle of the fluid to the horizontal axis; d is the inner diameter of the tube; a is the pressure wave propagation velocity; t is time; x is the distance along the pipeline; c (C) V Heat energy is liquid in the pipeline; k is the ground thermal conductivity; t (r) is a temperature function; t is the liquid temperature; r is the radial distance from the pipe diameter center at a particular location along the pipeline.
When the fluid is a liquid, the effect of temperature on the fluid flow, i.e. the energy variation in the pipe, can be neglected, and hence the original equation set (3) can be neglected.
And 2, solving the nonlinear partial differential equation set in the step 1 by adopting a characteristic line method. The characteristic line method needs to set boundary conditions and pipeline initial values, so that a pipeline steady-state model is considered, and the influence of time is not considered. From the formulae (1), (2):
equations (4) and (5) are steady-state models of pipelines, are a normal differential equation set, and can be solved by adopting a fourth-order Dragon-Gregory tower method to determine initial values of a real-time transient pipeline mechanism model. And taking the head end and the tail end of the pipeline as boundary conditions, setting input values of the boundary conditions as head end pressure and tail end flow, and obtaining the head end flow and the tail end pressure of the pipeline through model calculation.
And 3, establishing a hybrid model based on mechanism and data driving. The mechanism model is shown in formulas (4) and (5), and the convolutional neural network is used as a data driving model to correct the mechanism model. The input of the convolutional neural network is the boundary condition input value of the transient pipeline mechanism model, namely the head end pressure and the tail end flow when the pipeline runs. Because the input is the time sequence of the pressure and the flow extracted by the sensor, the time sequence is analyzed and calculated by adopting a one-dimensional convolutional neural network (1 DCNN), and the 1DCNN output is the difference between the transient pipeline mechanism model and the actual measured value.
And 4, extracting knowledge features based on the basic principle of the pipeline.
Is obtainable by means of the pipeline steady-state model patterns (4) and (5)
And (3) carrying out integral solution on equation sets (6) and (7) to obtain the relationship between the equivalent friction coefficient in the pipeline and the parameters in the pipeline:
wherein Q is 1 For the flow of the head end of the pipeline, Q 2 For the end flow of the pipeline, P 1 P is the pressure of the head end of the pipeline 2 For the end pressure of the pipe,for calculating the equivalent friction coefficient according to the head-end flow, < >>The equivalent friction coefficient is calculated according to the end flow.
And 5, fusing the pipeline knowledge extracted in the step 4 into 1DCNN, namely combining the characteristics extracted by the 1DCNN with the pipeline knowledge and then passing through the full connection layer.
And 6, correcting the real-time transient pipeline mechanism model through the model established in the step 5, so as to obtain pipeline operation data with real-time performance. When the leakage of the pipeline occurs, the head end flow and the tail end pressure of the pipeline are obtained through calculation, the model predicted value is compared with the actual measured value, and when the deviation exceeds a given threshold value, the leakage is indicated to occur and can be expressed as:
wherein Q is 1 For the flow measurement at the head end of the pipeline,p is the calculated value of the head-end flow 2 For the line end pressure measurement, +.>For end pressure calculations, σ 1 、σ 2 The set flow and pressure thresholds, respectively.
The normal flow data under different inlet pressure conditions are predicted, the inlet pressure of the pipeline is randomly changed at 30s, and the result is shown in fig. 4, so that the model obtained after correction through the convolutional neural network can be used for predicting the actual condition better. If the pipeline running time is too long and the internal parameters are changed, the original model has poor prediction effect on the actual process, and if the deviation value exceeds the set threshold value, false alarm can occur. Therefore, for the situation, firstly, knowledge, namely equivalent friction coefficient, is extracted from the pipeline, is fused with the characteristics extracted before the convolutional neural network layer, and then an accurate pipeline model can be built for a new working condition through a new full-connection layer. The inlet pressure and friction coefficient of the pipeline were randomly changed when the pipeline was operated for 30s, and it can be seen from FIG. 5 that the K-CNN-R T has a better prediction effect.
The whole pipeline has a length of 1600m, comprises two elbows, has a pipe diameter of 0.05m, has a negative pressure wave propagation speed of 1000m/s, is filled with water at 20 ℃, is provided with 4 leakage nodes, each leakage aperture is 1.2mm, and the leakage point positions are the leakage of each node controlled by ball valves. The inlet pressure of the pipeline is randomly sampled between 8bar and 25bar, the pressure and the flow of the two ends of the pipeline are sampled by using the sensors at the two ends of the pipeline, the sampling time is 30s, and the sampling frequency is 33Hz. Thus, the information collected by the sensor contains 1001 points, and the noise is added to the pressure signal at the head end and the noise added to the flow signal at the tail end of the pipeline to be connected as one sample, and each sample contains 2002 data points. And adding noise to the pipeline end pressure signal and the head end flow signal, and then storing the pipeline end pressure signal and the head end flow signal as real measurement data of pipeline output. A 400-group sample was produced for normal operation at different pressures, with 200-group tubing absolute roughness of 0.025mm and the remaining 200-group tubing absolute roughness of between 0.025mm and 0.05 mm. A500 sets of leak data are generated, with 100-200 sets of data taken for each leak location.
The real-time transient pipeline mechanism model is constructed through Matlab, the head end flow and the tail end pressure of the pipeline can be obtained through the calculation of the head end pressure and the tail end flow of the pipeline, the mechanism model is corrected through a convolutional neural network, and the neural network training is carried out by adopting 200 groups of pipeline data under different pressure flow conditions. The built convolutional neural network comprises 3 convolutional layers, each convolutional layer is provided with a pooling layer, and output is obtained through the flexible layer and the full-connection layer. After training, the accuracy of the obtained training set is 99.5%, and the accuracy of the test set is 99.2%.
When the leakage is changed, a leakage signal is transmitted to the head end and the tail end of the pipeline from the leakage point through negative pressure waves, the pressure threshold value is selected to be 0.15bar, and the flow threshold value is selected to be 0.000025m 3 And/s, when the difference between the output value of the mixed model and the measured value of the simulation software exceeds a threshold value, the leakage is considered to occur, and the leakage judgment accuracy is shown in table 1.
Table 1 leak determination accuracy
Claims (1)
1. A pipeline leakage detection method based on knowledge feature and mixed model fusion comprises the following steps:
step 1, establishing a pipeline fluid real-time transient pipeline mechanism model, which specifically comprises the following steps:
establishing a continuity equation and a motion momentum equation for the pipeline fluid motion:
wherein, the formula (1) is a flow continuity equation, the formula (2) is a motion momentum equation, and P is the average pressure of the section of the pipeline; v is the average flow velocity of the section of the pipeline; ρ is the fluid average density; f is the hydraulic friction coefficient; and (c) is a gravitational acceleration; θ is the angle of the fluid to the horizontal axis; d is the inner diameter of the tube; a is the pressure wave propagation velocity; t is time; x is the distance along the pipeline;
and 2, optimizing the pipeline fluid real-time transient pipeline mechanism model in the step 1 by adopting a characteristic line method, wherein the method specifically comprises the following steps of:
solving the nonlinear partial differential equation (1) and the equation (2) in the step 1 by adopting a characteristic line method; the characteristic line method needs to set boundary conditions and pipeline initial values, so that the pipeline steady-state model is considered, and the influence of time is not considered, and the characteristic line method can be obtained by the following formulas (1) and (2):
equations (4) and (5) are steady-state models of pipelines, are a normal differential equation set, and can be solved by adopting a fourth-order Dragon-Gregory tower method to determine initial values of a real-time transient pipeline mechanism model; taking the head end and the tail end of the pipeline as boundary conditions, and setting input values of the boundary conditions as head end pressure and tail end flow;
step 3, based on the basic principle of the pipeline, extracting knowledge features of equivalent friction coefficient in the pipeline, specifically comprising:
based on the basic principle of the pipeline, the knowledge features of equivalent friction coefficient in the pipeline are extracted, and the method specifically comprises the following steps:
is obtainable by means of the pipeline steady-state model versions (4) and (5):
and (3) carrying out integral solution on equation sets (6) and (7) to obtain the relationship between the equivalent friction coefficient in the pipeline and the parameters in the pipeline:
wherein Q is 1 For the flow of the head end of the pipeline, Q 2 For the end flow of the pipeline, P 1 P is the pressure of the head end of the pipeline 2 For the end pressure of the pipe,for calculating the equivalent friction coefficient according to the head-end flow, < >>The equivalent friction coefficient is calculated according to the end flow;
and 4, establishing a data driving model fusing pipeline knowledge and a mechanism and data driving-based hybrid model, wherein the method specifically comprises the following steps of:
fusing the knowledge features of the equivalent friction coefficient in the pipeline extracted in the step 3 into a one-dimensional convolutional neural network, and obtaining a data driving model fused with knowledge through a full-connection layer after combining;
the mechanism model is shown in formulas (4) and (5), and the knowledge-fused data-driven model corrects the mechanism model to obtain a hybrid model based on mechanism and data driving;
step 5, inputting a convolutional neural network in a data driving model fused with pipeline knowledge into a boundary condition input value of a pipeline fluid real-time transient pipeline mechanism model, wherein the input value is the head end pressure and the tail end flow of the pipeline in operation, and the head end flow and the tail end pressure of the pipeline in operation can be obtained through model calculation in the step 2;
step 6, obtaining pipeline operation data with real-time performance through the mechanism and data-driven hybrid model in the step 4; when the pipeline leakage occurs, the head end flow and the tail end pressure of the pipeline are obtained after calculation in the step 5, the model calculation value is used as a model calculation value, the model calculation value is compared with the actual measurement value of the sensor, and when the deviation exceeds a given threshold value, the leakage is indicated to occur, and the method is expressed as:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110979611.5A CN113803647B (en) | 2021-08-25 | 2021-08-25 | Pipeline leakage detection method based on fusion of knowledge features and hybrid model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110979611.5A CN113803647B (en) | 2021-08-25 | 2021-08-25 | Pipeline leakage detection method based on fusion of knowledge features and hybrid model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113803647A CN113803647A (en) | 2021-12-17 |
CN113803647B true CN113803647B (en) | 2023-07-04 |
Family
ID=78941844
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110979611.5A Active CN113803647B (en) | 2021-08-25 | 2021-08-25 | Pipeline leakage detection method based on fusion of knowledge features and hybrid model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113803647B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114636113A (en) * | 2022-02-22 | 2022-06-17 | 中国特种设备检测研究院 | Oil depot pipeline leakage detection method |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1125102A (en) * | 1997-06-30 | 1999-01-29 | Atr Chinou Eizo Tsushin Kenkyusho:Kk | Transmission method with blended knowledge and interest |
CA2598949A1 (en) * | 2006-09-01 | 2008-03-01 | Areva Np Gmbh | Collector line for leakage monitoring and leakage location |
CN101592288A (en) * | 2009-05-22 | 2009-12-02 | 重庆大学 | A kind of method for identifying pipeline leakage |
CN104750742A (en) * | 2013-12-31 | 2015-07-01 | 南京理工大学常熟研究院有限公司 | Fault diagnosis method and system for heading machine hydraulic system |
CN106015951A (en) * | 2016-07-07 | 2016-10-12 | 北京信息科技大学 | Gas transmission pipeline leakage detection system and method suitable for multiple state changes |
JP2019095292A (en) * | 2017-11-22 | 2019-06-20 | 株式会社日立製作所 | Water leakage detection system and water leakage detection method |
CN110296327A (en) * | 2019-06-19 | 2019-10-01 | 常州大学 | A kind of pipeline leakage detection method based on transient flow frequency response analysis |
CN111059477A (en) * | 2019-12-23 | 2020-04-24 | 浙江工业大学 | Double-layer framework based reverse source-seeking chemical pipeline leakage detection and positioning method |
CN111125867A (en) * | 2019-11-04 | 2020-05-08 | 浙江工业大学 | Establishment and calculation method of chemical production pipeline real-time transient model based on chaotic particle swarm |
CN112036440A (en) * | 2020-07-31 | 2020-12-04 | 山东科技大学 | Satellite attitude control system fault diagnosis and early warning method based on random forest |
CN112610903A (en) * | 2020-12-10 | 2021-04-06 | 合肥学院 | Water supply pipe network leakage positioning method based on deep neural network model |
CN113077091A (en) * | 2021-04-09 | 2021-07-06 | 中国石油大学(北京) | Machine learning-based leakage rate probability distribution prediction method and system |
Family Cites Families (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009156862A2 (en) * | 2008-06-27 | 2009-12-30 | Pii (Canada) Limited | Integrated multi-sensor non-destructive testing |
CN101761780B (en) * | 2010-01-11 | 2012-12-26 | 中国石油大学(华东) | Gas pipeline leakage detecting and positioning device and method thereof |
CN105042339B (en) * | 2015-06-03 | 2017-06-23 | 中国石化销售有限公司华东分公司 | One kind is based on nondimensional leakage of finished oil pipeline amount estimating system and method |
CN105353607B (en) * | 2015-11-26 | 2017-10-27 | 江南大学 | A kind of batch process self study dynamic optimization method driven by data difference |
US20180365555A1 (en) * | 2016-12-22 | 2018-12-20 | Naveed Aslam | Artificial intelligence based algorithm for predicting pipeline leak and corrosion detection |
US20180347763A1 (en) * | 2017-05-31 | 2018-12-06 | Schneider Electric Software, Llc | Machine learning detection of pipeline ruptures |
EP3884253B1 (en) * | 2018-11-23 | 2023-09-13 | The University of Adelaide | Method and system to analyse pipeline condition |
US11488010B2 (en) * | 2018-12-29 | 2022-11-01 | Northeastern University | Intelligent analysis system using magnetic flux leakage data in pipeline inner inspection |
CN110185939B (en) * | 2019-05-16 | 2021-04-02 | 西北工业大学 | Gas pipeline leakage identification method based on convolutional neural network |
WO2021081250A1 (en) * | 2019-10-22 | 2021-04-29 | Eog Resources, Inc. | Anomaly detection in pipelines and flowlines |
EP4066182A1 (en) * | 2019-11-26 | 2022-10-05 | Basf Se | Forecasting industrial aging processes with machine learning methods |
US11285673B2 (en) * | 2019-12-17 | 2022-03-29 | Northrop Grumman Systems Corporation | Machine-learning-based additive manufacturing using manufacturing data |
CN111915579B (en) * | 2020-07-24 | 2022-05-13 | 天津大学 | Liquid leakage detection method based on Gaussian mixture model and convolutional neural network |
-
2021
- 2021-08-25 CN CN202110979611.5A patent/CN113803647B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1125102A (en) * | 1997-06-30 | 1999-01-29 | Atr Chinou Eizo Tsushin Kenkyusho:Kk | Transmission method with blended knowledge and interest |
CA2598949A1 (en) * | 2006-09-01 | 2008-03-01 | Areva Np Gmbh | Collector line for leakage monitoring and leakage location |
CN101592288A (en) * | 2009-05-22 | 2009-12-02 | 重庆大学 | A kind of method for identifying pipeline leakage |
CN104750742A (en) * | 2013-12-31 | 2015-07-01 | 南京理工大学常熟研究院有限公司 | Fault diagnosis method and system for heading machine hydraulic system |
CN106015951A (en) * | 2016-07-07 | 2016-10-12 | 北京信息科技大学 | Gas transmission pipeline leakage detection system and method suitable for multiple state changes |
JP2019095292A (en) * | 2017-11-22 | 2019-06-20 | 株式会社日立製作所 | Water leakage detection system and water leakage detection method |
CN110296327A (en) * | 2019-06-19 | 2019-10-01 | 常州大学 | A kind of pipeline leakage detection method based on transient flow frequency response analysis |
CN111125867A (en) * | 2019-11-04 | 2020-05-08 | 浙江工业大学 | Establishment and calculation method of chemical production pipeline real-time transient model based on chaotic particle swarm |
CN111059477A (en) * | 2019-12-23 | 2020-04-24 | 浙江工业大学 | Double-layer framework based reverse source-seeking chemical pipeline leakage detection and positioning method |
CN112036440A (en) * | 2020-07-31 | 2020-12-04 | 山东科技大学 | Satellite attitude control system fault diagnosis and early warning method based on random forest |
CN112610903A (en) * | 2020-12-10 | 2021-04-06 | 合肥学院 | Water supply pipe network leakage positioning method based on deep neural network model |
CN113077091A (en) * | 2021-04-09 | 2021-07-06 | 中国石油大学(北京) | Machine learning-based leakage rate probability distribution prediction method and system |
Non-Patent Citations (3)
Title |
---|
基于模型的输气管道泄漏检测与定位系统;王军茹;刘丽华;王巧玲;;科技视界(第27期);全文 * |
管道泄漏检测技术研究进展;李平;张园园;;管道技术与设备(第02期);全文 * |
长距离流体输送管道泄漏检测与定位技术的现状与展望;王占山,张化光,冯健,伦淑娴;化工自动化及仪表(第05期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113803647A (en) | 2021-12-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111425768B (en) | Oil leakage point and oil leakage rate detection method for oil pipeline | |
CN113239504B (en) | Pipeline corrosion defect prediction method based on optimized neural network | |
CN104063588B (en) | Multi-source data fusion-based method for predicting pipeline corrosion defect size | |
CN102853261A (en) | Method and device for determining leakage amount of fluid in conveying pipeline | |
CN104180166A (en) | Pipeline leakage detection method based on pipeline pressure data | |
CN109469834B (en) | Liquid pipeline leakage detection method, device and system | |
CN113803647B (en) | Pipeline leakage detection method based on fusion of knowledge features and hybrid model | |
CN111339701B (en) | Pipeline leakage characteristic Godunov simulation method based on Brunone dynamic friction resistance | |
CN111684238A (en) | System and method for real-time discovery and resolution of wet gas venturi meter problems | |
CN113551744A (en) | Ultrasonic flowmeter performance online monitoring method and system | |
CN111102476B (en) | Pipe burst searching method based on three-dimensional underground pipeline | |
CN112628613A (en) | Method and system for monitoring pipeline leakage, positioning leakage and calculating leakage amount | |
US20170191862A1 (en) | System and method for identifying and recovering from a temporary sensor failure | |
CN115654381A (en) | Water supply pipeline leakage detection method based on graph neural network | |
CN114741815A (en) | Seabed petroleum pipeline health management method based on digital twinning | |
CN111365623A (en) | Negative pressure wave inflection point identification method based on linear fitting | |
Miao et al. | Regulating control of in-pipe intelligent isolation plugging tool based on adaptive dynamic programming | |
CN100383510C (en) | Detection of oil and gas pipeline leakage by additional dynamic micro-pressure signal | |
US20240003501A1 (en) | Methods and smart gas internet of things (iot) systems for remote control of ultrasonic metering devices | |
CN110807511A (en) | Coarse set theory-cyclic neural network model structure | |
CN116164241A (en) | Intelligent detection method for leakage faults of gas extraction pipe network | |
CN111125867B (en) | Method for establishing and calculating real-time transient model of chemical production pipeline based on chaotic particle swarm | |
WO2009022148A1 (en) | Method and apparatus for operating a pipeline | |
CN110705186B (en) | Real-time online instrument checksum diagnosis method through RBF particle swarm optimization algorithm | |
CN115705512A (en) | Method and system for identifying corrosion factors in pipe and predicting residual life |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |