A method and system for setting a key node sequence for a target pipe network structure

Abstract

The present invention relates to a method and system for setting a key node sequence for a target pipe network structure. The method is based on mean linear transformation, and expands the range of selectable sequence lengths through more balanced element division; and finally obtains a key node sequence that can be flexibly set according to the computing power of the big data platform, and can observe abnormal conditions in the gas pipe network based on the changes of the key node sequence and / or the monitoring data corresponding to the key node sequence. The present invention proposes a key node sequence to improve the stability of subsequent analysis, while also ensuring the efficiency of analysis and reducing the analysis complexity by one dimension.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban gas pipeline network operation monitoring, and particularly relates to a method and system for setting a key node sequence for a target pipeline network structure.

Background Art

[0002] With the acceleration of China's urbanization process and the continuous improvement of people's living standards, the urban central heating cause in China has developed greatly. The pipeline network system is also constantly developing and constructing. The urban pipeline network system is a very complex multi-variable control system, with characteristics such as large area, many influencing factors, strong internal correlation, long lag time, and serious non-linearity. Among them: the urban high-pressure gas pipeline network is the main artery of urban gas supply and also a lifeline project of the city. Its main function is to transport and distribute gas with sufficient pressure and flow to all users. To ensure the safe and stable operation of the pipeline network, the gas pipeline network is generally designed according to the flow rate required by users. However, due to the differences in various on-site factors, when conducting gas data discovery, the monitoring data of many nodes is incomplete or missing. In this case, the scope of problems discovered by analyzing the micro gas pipeline network starting from the nodes is limited. In addition, in the prior art, the data analysis of the gas pipeline network is often based on a single analysis target, and the transfer of pipeline network data to the big data direction has not been considered. The data analysis of the gas pipeline network is often based on traditional simulation software. Such simulation software is effective for the analysis based on an independent analysis target. However, in the case of a large amount of data, it is obviously impossible to directly use traditional simulation software for analysis. However, only by combining with new technologies can the beneficial effects of new technologies be obtained. How to combine with traditional analysis methods and utilize new technologies under the condition of downward compatibility is a technical problem to be solved. A feasible way is to select key nodes for subsequent analysis. Through the effective selection of key nodes, traditional analysis methods can be directly used. However, on the one hand, the analysis based on key nodes is not stable enough, and on the other hand, the selection of a single key node cannot reflect the hierarchical nature of the monitoring data shown by different nodes in the same community. Then, how to further improve the stability and hierarchy of the analysis is a problem to be solved. The present invention proposes a key node sequence to improve the stability of subsequent analysis, and at the same time can ensure the efficiency of the analysis, reducing the analysis complexity by one dimension;

Summary of the Invention

[0003] To solve the above problems in the prior art, the present invention proposes a method and system for setting a key node sequence for a target pipeline network structure, and the method includes:

[0004] Step S1: Construct a local pipe network structure based on the target pipe network structure; specifically: connect multiple sub-pipe network structures including the target pipe network structure, and construct a local pipe network structure based on the multiple sub-pipe network structures when the conditions for constructing the local pipe network structure are met;

[0005] Step S2: Select the key monitoring data types of the monitoring nodes in the local pipe network structure; and store the monitoring data of the key monitoring data types and the monitoring nodes in an associated manner;

[0006] Step S3: Extract the features of the local pipe network structure and construct a feature matrix; specifically, it includes the following steps:

[0007] Step S31: Number the monitoring nodes in the local pipe network structure;

[0008] Step S32: Calculate the size of the feature matrix according to the number of monitoring nodes;

[0009] Step S33: Construct a feature matrix according to the size of the feature matrix, and sequentially put the monitoring data of the monitoring nodes into the feature matrix M according to the numbers;

[0010] Step 4: Set a sliding window to perform a mean linear transformation on the feature matrix M of the local pipe network structure to obtain a mean linear transformation matrix MTQ;

[0011] Step S41: Set a sliding window W of size k×k; where: k>1;

[0012] Step S42: Perform a linear transformation on the elements of the feature matrix MT within the window to obtain a transformed window feature matrix MT‘ u ; MT u ; MT u =[mt_u i，j , MT‘ u =[mt_u‘ i,j ; where: i, j are row and column numbers, and u is the sliding window number; set the transformed window matrix MT‘ u ; MT‘ u =a u ×MT u +b u ; where: a u , b u are linear coefficients,

[0013] Step S43: Determine whether to retain the transformed window feature matrix MT‘ according to the objective function u ; retain it on the basis of meeting the objective function, and do not retain it when it does not meet;

[0014] Step S44: Determine whether the window has finished sliding. If so, proceed to step S45. At this time, each element m in the feature matrix M i，j corresponds to the corresponding elements of 1 to k×k transformed window matrices MT‘ u to form a set of feature elements D i，j corresponding to the element m i，j ={D_d i，j}; Otherwise, continue to slide the window, set u = u + 1, and return to step S42;

[0015] Step S45: Calculate the mean of the set of feature elements D i，j corresponding to each element m in the feature matrix M i，j to obtain the mean linear transformation matrix MTQ = [mtq i，j ;

[0016] where: D_d i，j is the d-th element in the set of feature elements D i，j ; ||D|| is the number of elements in D;

[0017] Step S5: Divide the elements in the mean linear transformation matrix into X intervals, select significant elements from the intervals, and arrange the monitoring nodes corresponding to the significant elements in ascending order of their numbers to form a key node sequence.

[0018] Furthermore, the target pipe network structure is a gas pipe network structure.

[0019] Furthermore, the target pipe network structure is a pipe network structure including specific nodes.

[0020] Furthermore, the method further includes step S6: observing abnormal conditions in the gas pipe network based on the change of the key node sequence and / or the monitoring data corresponding to the key node sequence.

[0021] Furthermore, the target pipe network structure is a pipe network structure of a key concern section.

[0022] A system for setting a key node sequence for a target pipe network structure, the system includes: a collection device, a setting node;

[0023] The collection device is used to collect the monitoring data of the monitoring nodes and send the collected monitoring data to the setting node;

[0024] The big data computing unit is used to execute the above method for setting the key node sequence for the target pipe network structure.

[0025] Furthermore, the collected data is a pressure parameter value.

[0026] A processor for running a program, wherein when the program runs, it executes the method for setting the key node sequence for the target pipe network structure as described above.

[0027] A computer-readable storage medium includes a program that, when run on a computer, causes the computer to execute the method for setting the key node sequence for the target pipe network structure as described above.

[0028] An execution device includes a processor coupled to a memory that stores program instructions, and when the program instructions stored in the memory are executed by the processor, the method for setting the key node sequence for the target pipe network structure as described above is implemented.

[0029] Based on the same inventive concept, the present invention further provides a processor for running a program, wherein when the program runs, it executes the method for setting the key node sequence for the target pipe network structure as described above.

[0030] The beneficial effects of the present invention include:

[0031] (1) A key node sequence is proposed to improve the stability of subsequent analysis, while also ensuring the efficiency of analysis, reducing the analysis complexity by one dimension; (3) Through a sliding window mechanism, based on considering the characteristics of global data values, the discrete points of the feature matrix are mapped into a numerical space related to the elements of the feature matrix and the elements within the window, thereby forming better discrimination; (3) Based on mean linear transformation, the range of selectable sequence lengths is extended through more balanced element division; enabling the key node sequence to be set in length as needed, so that subsequently, the length of the key node sequence can be flexibly set according to the computing power of the big data platform, ultimately ensuring the computing efficiency of the big data platform.

BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings described herein are used to provide a further understanding of the present invention, form a part of this application, but do not constitute an improper limitation to the present invention. In the drawings:

[0033] Figure 1 It is a schematic diagram of the method for setting the key node sequence for the target pipe network structure of the present invention.

DETAILED DESCRIPTION

[0034] The present invention will be described in detail below in conjunction with the drawings and specific embodiments, wherein the illustrative embodiments and descriptions are only used to explain the present invention, but do not limit the present invention.

[0035] As shown in the attached Figure 1 figures, the present invention proposes a method for setting the key node sequence for the target pipe network structure, and the method includes the following steps:

[0036] Step S1: Construct a local pipe network structure based on the target pipe network structure. Specifically: Connect multiple sub-pipe network structures including the target pipe network structure, and construct a local pipe network structure based on the multiple sub-pipe network structures when the conditions for constructing the local pipe network structure are met. Traditional simulation software analysis often targets sub-pipe network structures, and a pipe network structure often involves dozens or even hundreds of monitoring nodes. When conducting pipe network analysis based on big data, it is necessary to construct a local pipe network structure that meets the analysis requirements to provide support for effective analysis. Of course, through the connection of the local pipe network structure, more comprehensive gas pipe network data analysis can be carried out.

[0037] Preferably: The pipe network structure is a gas pipe network structure.

[0038] Preferably; The conditions for constructing the local pipe network structure are preset conditions, such as: conditions with a minimum limit on the monitoring nodes included therein, etc.

[0039] The specific steps of step S1 are as follows:

[0040] Step S11: Obtain the target pipe network structure; Set the local pipe network structure equal to the target pipe network structure.

[0041] Preferably: The target pipe network structure is the pipe network structure of a specified area or section; For example; It is the pipe network structure of a key concern section.

[0042] Preferably: The target pipe network structure is a pipe network structure containing specific nodes; Among them: The specific nodes are one or more.

[0043] Preferably: The target pipe network structure is manually specified.

[0044] Step S12: Obtain an unprocessed adjacent sub-pipe network structure of the target pipe network structure; And judge whether the distribution of the monitoring data of its adjacent sub-pipe network structure is the same as that of the target pipe network structure. If so, merge the local pipe network structure and the adjacent sub-pipe network structure to form a new local pipe network structure.

[0045] Preferably: Judging whether the distribution of the monitoring data of the adjacent sub-pipe network structure is the same as that of the target pipe network structure specifically means: If the mean value and the average deviation value of the monitoring data of both are within the preset range, it is considered that the data distributions of the two are the same; Of course, more complex judgment methods can also be used, which will not be listed one by one here.

[0046] The sub-pipe network structure can perform data supplementation. The number of sub-pipe network structures itself is often too small to have the significance of analysis, and it is usually not the object of concern either. Therefore, the structural integration of adjacent sub-pipe network structures is actually an operation that is beneficial to both parties.

[0047] Step S13: Determine whether all adjacent sub - pipeline network structures have been processed. If so, this step ends; otherwise, return to step S12;

[0048] "Adjacent" in the adjacent sub - pipeline network structure means that there are more than a preset number of shared monitoring nodes between the sub - pipeline network structure and the target pipeline network structure. The monitoring node is a node in an extended sense. As long as it is a user, it can be regarded as a monitoring node because at least the natural gas consumption data of the user can be obtained. Then each user has the attribute of a monitoring node in an extended sense. Through the above iterative processing, a local pipeline network structure centered on the target pipeline network structure can be obtained for effective analysis. Here, the local pipeline network structure is the analysis basis or a component of the global pipeline network structure;

[0049] Step S2: Select the key monitoring data types of the monitoring nodes in the local pipeline network structure; and store the monitoring data of the key monitoring data types in association with the monitoring nodes;

[0050] Preferably: The key monitoring data type is the pressure parameter value of the monitoring node;

[0051] Preferably: The key monitoring data type is set according to experience;

[0052] Preferably: The key monitoring data type is set according to data sensitivity comparison;

[0053] Preferably: The monitoring data type and the key monitoring data type are one or more; when there are multiple key monitoring data types, the analysis methods involved in the present invention can be used to perform multiple parallel analyses, and auxiliary analysis and final decision - making can be based on the results of multiple analyses;

[0054] Preferably: The monitoring data types include: pressure, flow rate, velocity, price, temperature, image, sound, etc.;

[0055] Step S3: Extract the features of the local pipeline network structure and construct a feature matrix; specifically, it includes the following steps:

[0056] Step S31: Number the monitoring nodes in the local pipeline network structure;

[0057] Step S32: Calculate the size of the feature matrix according to the number of monitoring nodes; specifically: Obtain the size of the feature matrix by taking the square root and rounding up;

[0058] Step S33: Construct a feature matrix according to the size of the feature matrix, and sequentially put the monitoring data of the monitoring nodes into the feature matrix M according to the numbers;

[0059] Preferably: The way of putting them in sequence is to put them in row - first and column - later order;

[0060] Preferably, positions in the feature matrix that are more than the number of monitoring nodes are set to default values;

[0061] Preferably, the default value is 0 or an invalid value;

[0062] Step 4: Set a sliding window to perform a mean linear transformation on the feature matrix M of the local pipe network structure to obtain a mean linear transformation matrix MTQ;

[0063] Step S41: Set a sliding window W of size k×k; where: k>1;

[0064] Preferably, k = 3; set the window number U = 1;

[0065] Step S42: Perform a linear transformation on the elements of the feature matrix MT within the window to obtain a transformed window feature matrix MT‘ u ; MT u ; MT u = [mt_u i，j ; MT‘ u = [mt_u‘ i,j where: i, j are row and column numbers, and u is the sliding window number;

[0066] Set the transformed window matrix MT‘ u ; MT‘ u = a u ×MT u + b u ; where: a u , b u are linear coefficients, where: b u is the variance of the element values within MT u ; is the mean of the feature matrix elements; The mean of the feature matrix elements within the sliding window; Through linear transformation, the discrete points of the feature matrix are mapped into a numerical space related to the feature matrix elements and the elements within the window, thereby forming better discrimination;

[0067] Step S43: Determine whether to retain the transformed window feature matrix MT‘ according to the objective function u ; Retain it on the basis of meeting the objective function, and do not retain it in case of non - compliance;

[0068] Preferably, the objective function is related to the key detection data type;

[0069] Preferably, set the objective function as:

[0070] Judge in the objective function If the value then the transformed window matrix MT' is not retained u ; otherwise, the transformed window matrix MT' is retained u ; where: α, β are correlation coefficients related to the critical detection data type;

[0071] Preferably: The non - retention method is deletion;

[0072] Preferably: Set α = 1, β = 1;

[0073] Step S44: Determine whether the window has finished sliding. If so, go to step S45; otherwise, continue window sliding, set u = u + 1, and return to step S42;

[0074] After step S44 ends normally, each element m in the feature matrix M i，j corresponds to 1 to k×k transformed window matrices MT' u whose corresponding elements form a set of characteristic elements D i，j corresponding to the element m i，j ={D_d i，j}; Some sets have fewer than k×k elements because they are filtered out in step S43; Of course, a mechanism can be set to ensure that not all elements are filtered out; For example: When all the transformed window feature matrices are filtered out, set MT' u = MT u ; At this time, the (i, j) - th element m in the feature matrix i，j corresponds to ||D|| values of the elements of the transformed window matrix; || || is to obtain the number of elements in the set;

[0075] Step S45: Calculate the mean of the set of characteristic elements corresponding to each element in the feature matrix M to obtain the mean linear transformation matrix MTQ = [mtq i，j ;

[0076] where: D_d i，j is the d - th element in the set of characteristic elements D i，j ; ||D|| is the number of elements in D;

[0077] Step S5: Divide the elements in the mean linear transformation matrix MTQ into X intervals, select significant elements from the intervals, and arrange the monitoring nodes corresponding to the significant elements in descending order of their numbers to form a critical node sequence;

[0078] The elements in the mean linear transformation matrix MTQ are divided into X intervals, specifically: the elements in MTQ are divided into X intervals such that the number of elements in each interval is the same or differs by one element, and the data values of the elements within the interval are the closest; where: N_ALL is the number of all monitoring nodes in the local pipe network structure; is the floor function; after the element division, the elements in the mean linear transformation matrix are divided into X intervals in a basically average manner. Since the mean linear transformation has been performed, more balanced element division can be carried out, thus expanding the range of selectable sequence lengths; if a traditional clustering method is used, for many data types, the pressure data of a small area may be clustered into 1 cluster, obviously such a clustering method is meaningless. If the data is simply enlarged to understand the pressure data, the relationship between the data is ignored; while the present invention is based on the mean linear transformation, which can perform more balanced element division, thus expanding the range of selectable sequence lengths; enabling the key node sequence to be set in length according to needs, and then the length of the key node sequence can be flexibly set according to the computing power of the big data platform;

[0079] The selection of significant elements from the intervals is specifically: select an element with the smallest difference from the mean of the element data values within each interval as the significant element;

[0080] After arranging the monitoring nodes corresponding to the significant elements in ascending order of their numbers to form a key node sequence, the maximum length of the key node sequence can be equal to the number of intervals, that is, the length is X. The data to be analyzed with one dimension reduced with a length of X can further shorten the length of the key node sequence through key node selection;

[0081] Preferably: flexibly set the length of the key node sequence according to the computing power of the big data platform; and further shorten the length of the key node sequence by means of truncation; for example: truncating from the beginning, truncating from the end and other methods;

[0082] Preferably: select 3 nodes from the key node sequence to form the final key node sequence, and the 3 nodes include: the median node, the maximum value node and the minimum value node in the key node sequence;

[0083] Preferably: the method further includes step S6: observing data anomalies in the gas pipeline network based on the key node sequence and / or the changes in the monitoring data corresponding to the key node sequence;

[0084] Based on the same inventive concept, the present invention proposes a key node sequence setting system for a target pipe network structure, and the system includes: a collection device, a setting node;

[0085] The acquisition device is used to acquire the monitoring data of the monitoring nodes and send the acquired monitoring data to the setting node;

[0086] The setting node is used to execute the above-mentioned key node sequence setting method for the target pipe network structure.

[0087] Preferably: The acquisition device is used to acquire pressure parameter values;

[0088] A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including assembly or interpreted languages, declarative or procedural languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program may or may not correspond to a file in a file system. The program can be stored as part of a file that holds other programs or data (such as one or more scripts in a markup language document), in a single file dedicated to the program, or in multiple cooperating files (such as files that store one or more modules, subroutines, or code portions). A computer program can be deployed to execute on one computer or on multiple computers located at one site or distributed across multiple sites and interconnected by a communication network.

[0089] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0090] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing device generate means for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0091] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the processes Figure 1 or blocks. Figure 1

[0092] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the processes Figure 1 or blocks. Figure 1

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the specific embodiments of the present invention can still be modified or equivalently replaced, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.​​

Claims

1. A method for setting a key node sequence for a target pipe network structure, characterized in that, Including: Step S1: Construct a local pipe network structure based on the target pipe network structure; Step S2: Select the key monitoring data types of the monitoring nodes in the local pipe network structure; Step S3: Extract the features of the local pipe network structure and construct a feature matrix; Step 4: Set a sliding window to perform a mean linear transformation on the feature matrix M of the local pipe network structure to obtain a mean linear transformation matrix MTQ; Step S41: Set a sliding window W of size k×k; where: k>1; Step S42: Generate the feature matrix MT within the window u After linear transformation of the elements, the transformed window feature matrix MT' is obtained u ; MT u = [mt_u i，j , MT' u = [mt_u' i,j ; where: i, j are row and column numbers, and u is the sliding window number; Set the transformed window matrix MT' u ; MT' u = a u × MT u + b u ; where: a u , b u are linear coefficients Step S43: Determine whether to retain the transformed window feature matrix MT' according to the objective function u ; retain it on the basis of meeting the objective function, and do not retain it if it does not meet the requirements; Step S44: Determine whether the window has finished sliding. If so, proceed to step S45. At this time, each element m in the feature matrix M i，j corresponds to the corresponding elements of 1 to k×k transformed window matrices MT‘ u to form a set of feature elements D i，j corresponding to the element m i，j ={D_d i，j}; Otherwise, continue to slide the window, set u = u + 1, and return to step S42; Step S45: For each element m in the feature matrix M i，j The corresponding set of feature elements D i，j Calculate the mean to obtain the mean linear transformation matrix MTQ = [mtq i，j ; Where: D_d i，j is the d-th element of the set of characteristic elements D i，j ; ||D|| is the number of elements in D; Step S5: Divide the elements in the mean linear transformation matrix into X intervals so that the number of elements in each interval is the same or differs by one element, and the data values of the elements in the interval are the closest. Select significant elements from the intervals, and arrange the monitoring nodes corresponding to the significant elements in ascending order of numbers to form a key node sequence.

2. The method for setting the key node sequence for the target pipe network structure according to claim 1, wherein The target pipe network structure is a gas pipe network structure.

3. The method for setting the key node sequence for the target pipe network structure according to claim 2, wherein, The target pipe network structure is a pipe network structure including specific nodes.

4. The method for setting the key node sequence for the target pipe network structure according to claim 3, wherein, The method further includes step S6: Observe abnormal conditions in the gas pipe network based on the key node sequence and / or the changes in the monitoring data corresponding to the key node sequence.

5. The method for setting the key node sequence for the target pipe network structure according to claim 4, wherein The target pipe network structure is a pipe network structure in a key attention section.

6. A key node sequence setting system for a target pipe network structure, characterized in that, The system includes: a collection device, a setting node; The collection device is used to collect the monitoring data of the inspection monitoring nodes and send the collected monitoring data to the setting node; The big data calculation node is used to execute the method for setting the key node sequence for the target pipe network structure according to any one of claims 1-5 above.

7. The key node sequence setting system for a target pipe network structure according to claim 6, characterized in that The collected data is a pressure parameter value.

8. A processor, characterized in that, The processor is used to run a program, wherein when the program runs, it executes the method for setting the key node sequence for the target pipe network structure according to any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, Including a program, when it runs on a computer, it causes the computer to execute the method for setting the key node sequence for the target pipe network structure according to any one of claims 1-5.

10. An execution device, characterized in that, Including a processor, the processor is coupled to a memory, and the memory stores program instructions. When the program instructions stored in the memory are executed by the processor, the method for setting the key node sequence for the target pipe network structure according to any one of claims 1-5 is implemented.

Images