CN114263855B - Natural gas transportation pipeline leakage prediction method and application thereof - Google Patents
Natural gas transportation pipeline leakage prediction method and application thereof Download PDFInfo
- Publication number
- CN114263855B CN114263855B CN202111401274.8A CN202111401274A CN114263855B CN 114263855 B CN114263855 B CN 114263855B CN 202111401274 A CN202111401274 A CN 202111401274A CN 114263855 B CN114263855 B CN 114263855B
- Authority
- CN
- China
- Prior art keywords
- leakage
- data
- pipeline
- monitoring point
- data set
- 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
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title claims abstract description 23
- 239000003345 natural gas Substances 0.000 title claims abstract description 23
- 238000012544 monitoring process Methods 0.000 claims abstract description 80
- 238000012545 processing Methods 0.000 claims abstract description 21
- 239000011159 matrix material Substances 0.000 claims description 30
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000009434 installation Methods 0.000 claims description 2
- 230000001052 transient effect Effects 0.000 claims description 2
- 239000007789 gas Substances 0.000 claims 1
- 238000013139 quantization Methods 0.000 claims 1
- 238000001514 detection method Methods 0.000 description 5
- 239000000463 material Substances 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 208000027418 Wounds and injury Diseases 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 208000014674 injury Diseases 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Abstract
The invention relates to the technical field of pipeline transportation, and provides a method for predicting leakage of a natural gas transportation pipeline and application thereof, wherein the method for predicting leakage comprises the following steps: acquiring an original data set and a monitoring point set of a transportation pipeline; processing the original data set to obtain a reduced data set; inputting the reduced data set into a prediction model to obtain a prediction probability set of the monitoring point set; constructing a pipeline model according to the monitoring point set, and filling the prediction probability set into the pipeline model to obtain the leakage probability of a unit rectangle of a grid area in the pipeline model; and judging whether the leakage probability of the unit rectangle of the grid area is smaller than a preset threshold value. By the method for predicting the leakage of the natural gas transportation pipeline and the application thereof, disclosed by the invention, the leakage probability of the leakage point of the transportation pipeline can be predicted, the leakage can be effectively found in time, and the leakage point can be accurately positioned.
Description
Technical Field
The invention relates to the technical field of pipeline transportation, in particular to a method for predicting leakage of a natural gas transportation pipeline and application thereof.
Background
When natural gas or the like is transported, the pipeline is inevitably corroded due to the influence of operation time, operation environment and the like, and finally leakage accidents are caused, so that the position of pipeline leakage needs to be positioned. The traditional detection method is carried out by a hardware detection method, and the detection method has the problems of poor generalization capability, low positioning precision and the like.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, an object of the present invention is to provide a method for predicting leakage of a natural gas transportation pipeline and an application thereof.
To achieve the above and other related objects, the present invention provides a method for predicting leakage of a natural gas transportation pipeline, comprising the steps of:
acquiring an original data set and a monitoring point set of a transportation pipeline;
processing the original data set to obtain a reduced data set;
Inputting the reduced data set into a prediction model to obtain a prediction probability set of the monitoring point set;
constructing a pipeline model according to the monitoring point set, and filling the prediction probability set into the pipeline model to obtain the leakage probability of a unit rectangle of a grid area in the pipeline model;
Judging whether the leakage probability of the unit rectangle of the grid area is smaller than a preset threshold value or not: if yes, judging that the unit rectangle of the grid area has no leakage risk, and if not, judging that the unit rectangle of the grid area has leakage risk.
In an embodiment of the present invention, the step of processing the original data set to obtain a reduced-dimension data set includes:
Sample centralization processing is carried out on the original data set to obtain centralization data sets of different types of labels;
processing the centralized data set to obtain a corresponding characteristic value set;
Selecting the maximum characteristic value in the characteristic value set to obtain a corresponding characteristic vector set;
And processing the characteristic vector set and the centralized data set to obtain a dimensionality reduction data set.
In one embodiment of the invention, the centralized data set x' i is represented as: Where x i is the ith data in the class label,
M is the total number of data contained in a single category label.
In one embodiment of the present invention, the set of eigenvalues is expressed as: a=qΣq -1, wherein,
Sigma is a diagonal matrix, the elements on each diagonal of which are the corresponding eigenvalues,
Q is a matrix of eigenvectors of covariance matrix a,
The covariance matrix a is expressed as: a=x X T,
X is a matrix of centered data X' i,
T is the transpose.
In one embodiment of the present invention, the feature vector set W is expressed as: w= (W 1,w2,w3,…,wd), where W d is a feature vector corresponding to the d-th maximum feature value, and the dimension-reduction dataset Y is expressed as: y=w×x.
In one embodiment of the present invention, the prediction probability set P is expressed as: wherein,
N is the total number of category labels and,
M is the total number of monitoring points,
P mn is the mth monitoring point leakage probability in the nth category label.
In an embodiment of the present invention, the step of constructing a pipeline model according to the monitoring point set, filling the prediction probability set into the pipeline model, and obtaining the leakage probability of the unit rectangle of the grid area in the pipeline model includes:
constructing a pipeline model according to the monitoring point set, and carrying out gridding treatment on the pipeline model according to the positions of the monitoring points in the monitoring point set to obtain leakage probability of a single monitoring point;
and obtaining the leakage probability of the unit rectangle of the grid region in the pipeline model according to the leakage probability of the monitoring points.
In an embodiment of the present invention, the leakage probability p i of the single monitoring point is expressed as: wherein,
I is the i-th monitoring point,
C.d is a grid form of the surrounding area of the ith monitoring point,
Is a quantized coefficient of c x d rectangles centered on the ith monitored point.
In one embodiment of the present invention, the leakage probability p w of the unit rectangle is expressed as: wherein,
W is a unit rectangle within the grid region,
K is the number of active monitoring points around the unit rectangle w,
Lambda i is the weight of the influence of different monitoring points on the unit rectangle w,
The contribution of the ith monitoring point to the unit rectangle w is expressed as/>
Is the quantized coefficient of a unit rectangle w centered on the ith monitored point.
The invention also provides a system for predicting leakage of the natural gas transportation pipeline, which comprises the following steps:
the data acquisition module is used for acquiring an original data set and a monitoring point set of the transportation pipeline;
The data processing module is used for processing the original data set to obtain a reduced-dimension data set;
The prediction module is used for processing the reduced-dimension data set to obtain a prediction probability set of the monitoring point set;
The model construction module is used for constructing a pipeline model, and filling the prediction probability set into the pipeline model to obtain the leakage probability of the unit rectangle of the grid area in the pipeline model; and
And the judging module is used for judging the leakage probability of the unit rectangle of the grid area so as to judge whether the unit rectangle of the grid area has leakage risk or not.
As described above, the invention provides the prediction method for the leakage of the natural gas transportation pipeline and the application thereof, which can timely and effectively find the leakage and accurately position the leakage point, thereby reducing the overhaul range and facilitating the investigation of maintenance personnel. Meanwhile, the intelligent detection is realized, the investment of manpower and material resources can be reduced, the cost of leakage tracing is reduced, the direct contact of personnel is reduced, and the risk of injury of the personnel is reduced. The invention can give out early warning in time, prevent the position with possible leakage risk in advance, and make some protection measures in time, so as to minimize personnel and property loss and effectively protect personnel life safety.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 shows a flow chart of a method for predicting natural gas transport pipeline leakage according to the present invention.
Fig. 2 is a flowchart showing the substeps of step S2 in a method for predicting leakage of a natural gas transportation pipeline according to the present invention.
Fig. 3 is a flowchart showing the substeps of step S4 in a method for predicting leakage of a natural gas transportation pipeline according to the present invention.
Fig. 4 shows a schematic diagram of a natural gas transportation pipeline leakage prediction system according to the present invention.
In the figure: 1. a data acquisition module; 2. a data processing module; 3. a prediction module; 4. a model building module; 5. and a judging module.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the invention discloses a method for predicting leakage of a natural gas transportation pipeline, which can predict possible leakage positions on the natural gas transportation pipeline so as to predict leakage probability of leakage points of the transportation pipeline, effectively discover leakage in time and accurately position the leakage points. The method for predicting natural gas transport pipeline leakage may include the steps of:
and S1, acquiring an original data set and a monitoring point set of the transportation pipeline.
In this embodiment, the original data set may include a first data set, a second data set, and a third data set. The first data set is formed by simulating an actual working environment with multiple factors such as pressure, temperature and the like through software and the like, and a discretized transient flow model comprising a plurality of monitoring points is established, so that basic data such as pressure, temperature, flow rate, concentration and the like of the monitoring points can be obtained through calculation, each group of data corresponds to a corresponding class label, for example, a pressure data set corresponds to a pressure label, and the basic data and the class labels are collected to obtain the first data set.
The second data set refers to data acquired in real time through the sensor, for example, a plurality of pressure sensors can be installed in the pipeline, so that the pressures at different positions in the pipeline can be detected in real time, and corresponding pressure data can be obtained; the temperature sensors are arranged in the pipeline, so that the temperatures of different positions in the pipeline can be detected in real time, and corresponding temperature data are obtained; the flow velocity sensors are arranged in the pipeline, so that the flow velocity at different positions in the pipeline can be detected in real time, and corresponding flow velocity data are obtained; through installing a plurality of concentration sensors in the pipeline to can carry out real-time detection to the concentration of different positions in the pipeline, thereby obtain corresponding concentration data, can detect different kinds of labels through different sensors, thereby obtain corresponding data.
The third data set refers to the corresponding data set obtained after the pipeline is detected in advance by other companies. The monitoring point set refers to the installation position of a plurality of sensors, and corresponding data in the pipeline can be detected through different sensors.
Referring to fig. 2, in step S2, the original data set is processed to obtain a reduced-dimension data set. The step of processing the original data set to obtain a reduced-dimension data set includes:
step S21, performing sample centering processing on the original dataset to obtain a centering dataset x' i of different category labels, which is expressed as: wherein,
X i is the ith data in the class label,
M is the total number of data contained in a single category label,
The original dataset X m*n is represented as: wherein,
N is the number of category labels and,
A mn is the mth data of the nth class label,
C n is a column vector of data contained in the nth class label.
Step S22, processing the centralized data set to obtain a corresponding characteristic value set, wherein the characteristic value set is expressed as: a=qΣq -1, wherein,
Q is a matrix of eigenvectors of covariance matrix a,
Sigma is a diagonal matrix, the elements on each diagonal of which are the corresponding eigenvalues,
The covariance matrix a is expressed as: a=x× T, wherein,
X is a matrix of centered data X' i,
T is the transpose.
Step S23, selecting the maximum eigenvalue in the eigenvalue set to obtain a corresponding eigenvector set W, wherein the corresponding eigenvector set W is expressed as: w= (W 1,w2,w3,…,wd), wherein the largest eigenvalue is represented as the element with the largest value on the diagonal of each diagonal matrix Σ, and W d is the eigenvector corresponding to the d-th largest eigenvalue.
Step S24, processing the feature vector set and the centralized data set to obtain a reduced-dimension data set Y, which is expressed as: y=w×x.
In step S21, when the original data set needs to be processed, the original data set may be represented in a matrix form, so that the original data set may be processed later, where the matrix form X m*n of the original data set is represented as: wherein,
N is the number of category labels and,
A mn is the mth data of the nth class label,
C n is a column vector of data contained in the nth class label. Specifically, each column of data in the matrix form of the original dataset represents the data of the class label, for example, the first column of data may represent the data of the class label as pressure, the second column of data may represent the data of the class label as temperature, the third column of data may represent the data of the class label as flow rate, and the fourth column of data may represent the data of the class label as concentration. When the sample centralization processing is performed on the original data set, a corresponding average value can be firstly calculated for each column of data in a matrix form of the original data set, and then a difference value between specific data in each column of data and the average value of the column of data is calculated, so that centralization data sets of different column of data, namely centralization data sets x' i of different types of labels, can be obtained, and the centralization data sets are expressed as: wherein,
X i is the ith data in the class label,
M is the total number of data contained in a single category label.
In step S22, after the centralized data set x 'i of the different category labels is obtained, in order to facilitate processing thereof, the centralized data set x' i may be represented in a matrix form, which is expressed as: To obtain a corresponding set of eigenvalues, a covariance matrix a of the centered dataset may be calculated first, expressed as: a=x× T, where T is a transpose, and then performing eigenvalue decomposition on the covariance matrix a to obtain a=qΣq -1, where Q is a matrix formed by eigenvectors of the covariance matrix a, Σ is a diagonal matrix, and elements on each diagonal line are corresponding eigenvalues, so that a corresponding eigenvalue set can be obtained according to the diagonal matrix Σ.
In step S23, since there are multiple eigenvalues on each diagonal matrix Σ, the largest eigenvalue among the multiple eigenvalues of the single diagonal matrix Σ may be selected first, and then the other diagonal matrices Σ are processed accordingly, so that the largest eigenvalue among the diagonal matrix Σ may be selected to obtain a largest eigenvalue set, and thus, the eigenvector set W corresponding to the largest eigenvalue set may be expressed as: w= (W 1,w2,w3,…,wd), where W d is the feature vector corresponding to the d-th maximum feature value.
In step S24, after obtaining the feature vector set corresponding to the maximum feature value set, a dimension-reduced data set Y may be obtained according to the centralized data set and the feature vector set, which is expressed as: y=w×x.
Step S3, inputting the reduced-dimension data set into a prediction model to obtain a prediction probability set P of the monitoring point set, wherein the prediction probability set P is expressed as: Wherein p mn is the leakage probability of the mth monitoring point in the nth category label.
In this embodiment, the neural network of the prediction model may be trained using the dimension-reduced dataset, and training parameters of the neural network may be optimized to classify the predicted leak points. Therefore, different prediction probabilities can be given to different maximum characteristic values and different category labels to optimize the data monitored by the characterization monitoring points, so that a prediction probability set P of a monitoring point set can be obtained, and the prediction probability set P is expressed as: Wherein p mn is the leakage probability of the mth monitoring point in the nth category label.
Referring to fig. 3, step S4 is to construct a pipeline model according to the monitoring point set, and fill the prediction probability set into the pipeline model to obtain the leakage probability of the unit rectangle of the grid region in the pipeline model. The step of constructing a pipeline model according to the monitoring point set, filling the prediction probability set into the pipeline model, and obtaining the leakage probability of the unit rectangle of the grid region in the pipeline model comprises the following steps:
step S41, constructing a pipeline model according to the monitoring point set, and carrying out gridding treatment on the pipeline model according to the positions of the monitoring points in the monitoring point set to obtain leakage probability p i of a single monitoring point, wherein the leakage probability p i is expressed as: wherein,
I is the i-th monitoring point,
C.d is a grid form of the surrounding area of the ith monitoring point,
Is a quantized coefficient of c x d rectangles centered on the ith monitored point.
Step S42, obtaining leakage probability p w of a unit rectangle of a grid area in the pipeline model according to the leakage probability of the monitoring points, wherein the leakage probability p w is expressed as: wherein,
W is a unit rectangle within the grid region,
K is the number of active monitoring points around the unit rectangle w,
Lambda i is the weight of the influence of different monitoring points on the unit rectangle w,
The contribution of the ith monitoring point to the unit rectangle w is expressed as/>Wherein,
Is the quantized coefficient of a unit rectangle w centered on the ith monitored point.
In step S41, a pipeline model may be constructed for the transportation pipeline according to the positions of the monitoring points in the monitoring point set, and the pipeline model is subjected to gridding, so that the surrounding area of each monitoring point is represented in a grid form, and the surrounding area of each monitoring point is divided into a plurality of unit rectangles. Thus, the leakage probability p i of the single monitoring point can be obtained according to the prediction probability set of the monitoring point set, and is expressed as: wherein,
I is the i-th monitoring point,
C.d is a grid form of the surrounding area of the ith monitoring point,
Is a quantized coefficient of c x d rectangles centered on the ith monitored point.
In step S42, since the leak probability p i for each monitoring point is not greater than 1, after a plurality of unit rectangles are divided for the surrounding area of each monitoring point, the accumulation of the leak probabilities p w for all the unit rectangles is also not greater than 1. Therefore, the surrounding area of each monitoring point can be divided into a plurality of unit rectangles, the leakage probability of each monitoring point is used as a weight value and is filled into the subdivided unit rectangles, the leakage probability of the unit rectangles can be obtained, and when the area of the unit rectangles is small enough, the accurate positioning of the leakage points in the pipeline can be realized. Wherein, the leakage probability p w of the unit rectangle of the grid area in the pipeline model is expressed as: wherein,
W is a unit rectangle within the grid region,
K is the number of active monitoring points around the unit rectangle w,
Lambda i is the weight of the influence of different monitoring points on the unit rectangle w,
The contribution of the ith monitoring point to the unit rectangle w is expressed as/>Wherein,
Is the quantized coefficient of a unit rectangle w centered on the ith monitored point.
Step S5, judging whether the leakage probability of the unit rectangle of the grid area is smaller than a preset threshold value or not: if so, judging that the unit rectangle of the grid area has no leakage risk, if not, judging that the unit rectangle of the grid area has leakage risk, and closing the transportation pipeline.
In this embodiment, the specific size of the preset threshold may be set according to the actual requirement, for example, may be set to 0.3, 0.4, 0.5, etc. When the leakage probability of a certain unit rectangle in the grid area is larger than a preset threshold value, an alarm can be sent to the system at the moment, the leakage points are positioned through a pipeline model of the transportation pipeline and the specific positions of the unit rectangles in the grid area, so that workers can be reminded of the possible leakage risk of the leakage points, the transportation pipeline is closed, meanwhile, the system can record data recorded by sensors around the leakage points in a period of time, and the data are reserved, so that parameters in a neural network of the prediction model can be optimized in the later period of time, the specific length of the period of time for recording the data can be set according to actual requirements, for example, all data in the first half hour and the later half hour of leakage can be recorded, and the corresponding data can be displayed through a data statistics chart and the like when the leakage occurs.
Referring to fig. 4, the present invention also discloses a system for predicting leakage of a natural gas transportation pipeline, which may include: the system comprises a data acquisition module 1, a data processing module 2, a prediction module 3, a model construction module 4 and a judgment module 5. The data acquisition module 1 may be configured to acquire an original data set and a monitoring point set of a transportation pipeline, the data processing module 2 may be configured to process the original data set, thereby obtaining a dimension-reduced data set, the prediction module 3 may process the dimension-reduced data set, obtain a prediction probability set of the monitoring point set, the model construction module 4 may construct a pipeline model according to the monitoring point set, and fill the prediction probability set into the pipeline model, thereby obtaining a leakage probability of a unit rectangle of a grid area in the pipeline model, and the judgment module 5 may be configured to judge the leakage probability of the unit rectangle of the grid area, so as to judge whether the unit rectangle of the grid area has a leakage risk.
In summary, by the method for predicting the leakage of the natural gas transportation pipeline and the application thereof provided by the invention, the leakage can be timely and effectively found, and the leakage point can be accurately positioned, so that the overhaul range can be reduced, and the investigation by maintainers is facilitated. Meanwhile, the intelligent detection is realized, the investment of manpower and material resources can be reduced, the cost of leakage tracing is reduced, the direct contact of personnel is reduced, and the risk of injury of the personnel is reduced. The invention can give out early warning in time, prevent the position with possible leakage risk in advance, and make some protection measures in time, so as to minimize personnel and property loss and effectively protect personnel life safety.
In the description of the present specification, the descriptions of the terms "present embodiment," "example," "specific example," and the like, mean 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 invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The embodiments of the invention disclosed above are intended only to help illustrate the invention. The examples are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (6)
1. A method of predicting natural gas transportation pipeline leakage comprising the steps of:
Acquiring an original data set and a monitoring point set of a transportation pipeline, wherein the original data set comprises a first data set, a second data set and a third data set, the actual working environment is simulated through software, a discretized transient flow model comprising a plurality of monitoring points is established, basic data of pressure, temperature, flow rate and concentration of the monitoring points are obtained through calculation, each group of basic data corresponds to a corresponding class label, the data of the basic data and the class labels are collected to form the first data set, the second data set represents the data collected in real time through a sensor, the third data set represents the data obtained by detecting the pipeline in advance, the monitoring point set represents the installation positions of the plurality of sensors, and the corresponding data in the pipeline is detected through different sensors;
Sample centralization processing is carried out on the original data set to obtain centralization data sets of different types of labels;
processing the centralized data set to obtain a corresponding characteristic value set;
Selecting the maximum characteristic value in the characteristic value set to obtain a corresponding characteristic vector set;
Processing the feature vector set and the centralized data set to obtain a dimensionality reduction data set;
Inputting the reduced data set into a prediction model to obtain a prediction probability set of the monitoring point set;
constructing a pipeline model according to the monitoring point set, and carrying out gridding treatment on the pipeline model according to the positions of the monitoring points in the monitoring point set to obtain leakage probability of a single monitoring point;
Obtaining the leakage probability of a unit rectangle of a grid area in the pipeline model according to the leakage probability of the monitoring points;
judging whether the leakage probability of the unit rectangle of the grid area is smaller than a preset threshold value or not: if yes, judging that the unit rectangle of the grid area has no leakage risk, and if not, judging that the unit rectangle of the grid area has leakage risk;
Wherein, the leakage probability p i of the single monitoring point is expressed as: Wherein i is the ith monitoring point, c is the grid form of the surrounding area of the ith monitoring point,/> Quantization coefficients for c x d rectangles centered on the ith monitoring point;
The leakage probability per rectangle p w is expressed as: Wherein w is a unit rectangle in the grid area, k is the number of effective monitoring points around the unit rectangle w, lambda i is the weight of the influence of different monitoring points on the unit rectangle w, The contribution of the ith monitoring point to the unit rectangle w is expressed as/>Is the quantized coefficient of a unit rectangle w centered on the ith monitored point.
2. The method of claim 1, wherein the centralized dataset x i, expressed as: wherein,
X i is the ith data in the class label,
M is the total number of data contained in a single category label.
3. The method of claim 1, wherein the set of eigenvalues is represented as: a=qΣq -1, wherein,
Sigma is a diagonal matrix, the elements on each diagonal of which are the corresponding eigenvalues,
Q is a matrix of eigenvectors of covariance matrix a,
The covariance matrix a is expressed as: a=x X T,
X is a matrix of centered data X' i,
T is the transpose.
4. The method of predicting natural gas transportation pipeline leakage of claim 1, wherein the set of eigenvectors W is represented as: w= (W 1,w2,w3,…,wd), where W d is a feature vector corresponding to the d-th maximum feature value, and the dimension-reduction dataset Y is expressed as: y=w×x.
5. A method of predicting gas transportation pipeline leakage as claimed in claim 1, wherein the set of prediction probabilities P is expressed as: wherein,
N is the total number of category labels and,
M is the total number of monitoring points,
P mn is the mth monitoring point leakage probability in the nth category label.
6. A natural gas transportation pipeline leakage prediction system, characterized in that the natural gas transportation pipeline leakage prediction method according to any one of claims 1 to 5 is applied.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111401274.8A CN114263855B (en) | 2021-11-19 | 2021-11-19 | Natural gas transportation pipeline leakage prediction method and application thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111401274.8A CN114263855B (en) | 2021-11-19 | 2021-11-19 | Natural gas transportation pipeline leakage prediction method and application thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114263855A CN114263855A (en) | 2022-04-01 |
| CN114263855B true CN114263855B (en) | 2024-04-26 |
Family
ID=80825463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111401274.8A Active CN114263855B (en) | 2021-11-19 | 2021-11-19 | Natural gas transportation pipeline leakage prediction method and application thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114263855B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116822952A (en) * | 2023-06-26 | 2023-09-29 | 北京讯腾智慧科技股份有限公司 | Risk assessment method and device for gas pipe network |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101012913A (en) * | 2007-02-06 | 2007-08-08 | 东北大学 | Chaos analysis and micro-processor based conduit pipe micro-leakage diagnosing method and device |
| CN103914622A (en) * | 2014-04-04 | 2014-07-09 | 清华大学 | Quick chemical leakage predicating and warning emergency response decision-making method |
| CN108647470A (en) * | 2018-05-29 | 2018-10-12 | 杭州电子科技大学 | A kind of localization method at the beginning of based on the leakage loss with depth belief network is clustered |
| CN110107819A (en) * | 2019-05-17 | 2019-08-09 | 河南工业大学 | A kind of petroleum chemicals conveyance conduit leakage monitoring early warning system and method |
| CN111734961A (en) * | 2020-06-24 | 2020-10-02 | 东北石油大学 | Natural gas pipeline leakage detection method |
| CN113124328A (en) * | 2021-03-15 | 2021-07-16 | 湖南匡楚科技有限公司 | Natural gas pipeline leakage detection method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11263374B2 (en) * | 2019-10-22 | 2022-03-01 | Saudi Arabian Oil Company | Bisection numerical algorithm coupled with pressure deviation method to determine the size and location of a leak in a pipeline |
-
2021
- 2021-11-19 CN CN202111401274.8A patent/CN114263855B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101012913A (en) * | 2007-02-06 | 2007-08-08 | 东北大学 | Chaos analysis and micro-processor based conduit pipe micro-leakage diagnosing method and device |
| CN103914622A (en) * | 2014-04-04 | 2014-07-09 | 清华大学 | Quick chemical leakage predicating and warning emergency response decision-making method |
| CN108647470A (en) * | 2018-05-29 | 2018-10-12 | 杭州电子科技大学 | A kind of localization method at the beginning of based on the leakage loss with depth belief network is clustered |
| CN110107819A (en) * | 2019-05-17 | 2019-08-09 | 河南工业大学 | A kind of petroleum chemicals conveyance conduit leakage monitoring early warning system and method |
| CN111734961A (en) * | 2020-06-24 | 2020-10-02 | 东北石油大学 | Natural gas pipeline leakage detection method |
| CN113124328A (en) * | 2021-03-15 | 2021-07-16 | 湖南匡楚科技有限公司 | Natural gas pipeline leakage detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114263855A (en) | 2022-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Eftekhar Azam et al. | Damage detection in structural systems utilizing artificial neural networks and proper orthogonal decomposition | |
| Lei et al. | Mutual information based anomaly detection of monitoring data with attention mechanism and residual learning | |
| CN109583570A (en) | The method for determining bridge health monitoring system abnormal data source based on deep learning | |
| Santos et al. | Genetic‐based EM algorithm to improve the robustness of Gaussian mixture models for damage detection in bridges | |
| CN108764601A (en) | A kind of monitoring structural health conditions abnormal data diagnostic method based on computer vision and depth learning technology | |
| Giglioni et al. | Autoencoders for unsupervised real‐time bridge health assessment | |
| KR101711606B1 (en) | A similarity-based damage estimating method of offshore structures | |
| CN114263855B (en) | Natural gas transportation pipeline leakage prediction method and application thereof | |
| CN113033011B (en) | Ship mechanical health state assessment method and system | |
| CN113705875A (en) | Crop disease and pest early warning method, system, device and storage medium | |
| CN112668526A (en) | Bolt group loosening positioning monitoring method based on deep learning and piezoelectric active sensing | |
| CN113760670A (en) | Cable joint abnormity early warning method and device, electronic equipment and storage medium | |
| CN116934262A (en) | Construction safety supervision system and method based on artificial intelligence | |
| CN111931573A (en) | Helmet detection and early warning method based on YOLO evolution deep learning model | |
| CN115165725A (en) | Data-driven marine equipment corrosion monitoring and safety early warning system | |
| CN107229916A (en) | A kind of airport noise Monitoring Data restorative procedure based on depth noise reduction own coding | |
| CN113485218B (en) | Wisdom thing allies oneself with supervision platform based on 5G | |
| CN115060312A (en) | Building material safety monitoring system based on artificial intelligence | |
| CN114841518A (en) | Service state evaluation method for inland river navigation aid sign | |
| Eick et al. | Miter gate gap detection using principal component analysis | |
| CN116756505A (en) | Photovoltaic equipment intelligent management system and method based on big data | |
| Li et al. | Meteorological radar fault diagnosis based on deep learning | |
| CN116432831A (en) | Prediction method for abnormal event of chemical risk type | |
| CN115187880A (en) | Communication optical cable defect detection method and system based on image recognition and storage medium | |
| CN111967419B (en) | Dam bank dangerous case prediction method, dam bank dangerous case prediction device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |