A water supply network topology verification method, device, equipment and medium

Abstract

The application discloses a water supply network topology verification method, device, equipment and medium, the water supply network includes a plurality of topology nodes and the topology edge connected at least two topology nodes, the water supply network topology verification method includes obtaining the pipe network topology data of the water supply network of different data sources respectively, and the operation data of the topology node in the pipe network topology data; the consistency of the pipe network topology data of different data sources is verified, and the preliminary verification result is generated; the feature extraction is carried out to the pipe network topology data, and the first embedding vector of the topology node is determined; the second embedding vector of the topology node is determined according to the operation data, and the first embedding vector is combined with the second embedding vector, and the fusion feature vector of the topology node is obtained; the correlation index of two topology nodes is determined according to the fusion feature vector of each. Through the above setting, the artificial cost is reduced, and the accuracy, reliability and flexibility of the water supply network topology verification are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water management, and particularly relates to a water supply network topology verification method, device, equipment and medium. BACKGROUND

[0002] Water supply network topology verification is a process of ensuring that the spatial topology structure of the water supply network is consistent with the actual operation state through logical verification of the topology nodes, topology edges and connection relationship of the topology nodes and topology edges of the network. Accurate water supply network topology verification is of great significance to the management, hydraulic simulation, optimal scheduling and the like of the water supply network.

[0003] In the process of water supply network topology verification, if the network topology data is obtained and maintained by manual operation, the labor cost is high, and the network topology data obtained by manual operation may not be updated in time, which may result in inconsistency between the network topology data and the actual situation, low accuracy and reliability of the water supply network topology verification, and over-reliance on the integrity of the model, which leads to insufficient flexibility of the water supply network topology verification based on the hydraulic model. SUMMARY

[0004] In order to solve the problems of the prior art, the present application provides a water supply network topology verification method, device, equipment and medium, which can reduce labor cost and improve the accuracy, reliability and flexibility of the water supply network topology verification.

[0005] To achieve the above object, the present application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a water supply network topology verification method, the water supply network comprising a plurality of topology nodes and topology edges connecting at least two topology nodes, and the water supply network topology verification method comprising:

[0007] obtaining network topology data of the water supply network from different data sources, and operation data of the topology nodes in the network topology data;

[0008] performing consistency verification on the network topology data from different data sources to generate a preliminary verification result;

[0009] if the preliminary verification result is within a preset verification range, performing feature extraction on the network topology data to obtain structured features representing the connection relationship between the topology nodes and the topology edges, and / or the position of the topology nodes, and determining a first embedding vector of the topology nodes according to the structured features;

[0010] obtaining time series features representing the behavior mode of the topology nodes according to the operation data, determining a second embedding vector of the topology nodes according to the time series features, and combining the first embedding vector and the second embedding vector to obtain a fusion feature vector of the topology nodes;

[0011] determining a correlation index of the two topological nodes according to the fusion feature vectors of the two topological nodes respectively, and determining an inconsistency score of a topological edge connecting the two topological nodes based on the correlation index of the two topological nodes;

[0012] determining a deviation of the connection relationship of the topological edge in the pipe network topology data from the connection relationship in the actual scene based on the inconsistency score.

[0013] In some possible implementation manners, the consistency verification of the pipe network topology data of different data sources comprises:

[0014] determining the same topological node in different data sources according to the unique identification code of the topological node;

[0015] performing consistency comparison on the connection relationship, the position attribute, the structure attribute and / or the classification attribute of the same topological node in different data sources; the structure attribute comprises pipe diameter, length and elevation, and the classification attribute comprises material and current state.

[0016] In some possible implementation manners, the time series feature representing the behavior pattern of the topological node is obtained according to the operation data, comprising:

[0017] obtaining operation data of the topological node at multiple time nodes in a time period;

[0018] obtaining the behavior pattern of the topological node in the time period based on the operation data, the behavior pattern being a pressure change and / or a flow change of the topological node in the time period;

[0019] obtaining basic features and time series features of the pressure change and the flow change respectively, the basic features comprising pressure mean value, pressure variance, pressure maximum value and pressure minimum value of the topological node at each time node, and flow mean value, flow variance, flow maximum value and flow minimum value; the time series features comprising Fourier transform coefficients and autocorrelation coefficients;

[0020] performing normalization processing on the basic features and the time series features to obtain the time series features.

[0021] In some possible implementation manners, the fusion feature vector of the topological node is obtained by combining the first embedding vector and the second embedding vector, comprising:

[0022] splicing the first embedding vector and the second embedding vector to obtain a preprocessed vector;

[0023] performing calculation on the preprocessed vector based on a linear layer to adjust the dimension of the preprocessed vector to obtain the fusion feature vector.

[0024] In some possible implementation manners, the water supply pipe network topology verification method further comprises:

[0025] obtain a topology edge with an inconsistency score not less than a preset score threshold;

[0026] transmit the association information of the topology edge to the terminal device, the association information including the inconsistency score, attribute information, and the preliminary verification result; wherein the attribute information includes a connection relationship, a location attribute, a structure attribute, and / or a classification attribute of the topology edge, the structure attribute including a pipe diameter, a length, and an elevation, and the classification attribute including a material and a current state;

[0027] compare the association information of the topology edge with a database pre-stored in the terminal device to generate a final verification result.

[0028] In some possible implementation manners, the correlation index includes a Pearson correlation coefficient index of the running data of the two topology nodes and a difference index of the fusion feature vectors of the two topology nodes.

[0029] In some possible implementation manners, the data source includes a GIS geographic information data source, a CAD computer-aided design data source, and a device list data source, and the running data includes pressure and flow.

[0030] In a second aspect, the present application further provides a water supply network topology verification device, the water supply network including a plurality of topology nodes and topology edges connecting at least two topology nodes, characterized in that the water supply network topology verification device includes an acquisition module, a first verification module, a second verification module, and a calculation module, the acquisition module being configured to acquire network topology data of the water supply network from different data sources and running data of the topology nodes in the network topology data, the first verification module being configured to perform consistency verification on the network topology data from different data sources and output a preliminary verification result, the second verification module being configured to, in a case where the preliminary verification result is within a preset verification range, perform feature extraction on the network topology data to obtain structured features representing a connection relationship between the topology nodes and the topology edges and / or a location of the topology nodes, determine a first embedding vector of the topology node according to the structured features, obtain time series features representing a behavior pattern of the topology node according to the running data, determine a second embedding vector of the topology node according to the time series features, and combine the first embedding vector and the second embedding vector to obtain a feature vector of the topology node, and the calculation module being configured to determine a correlation index of two topology nodes according to the feature vectors of the two topology nodes, determine an inconsistency score of a topology edge connecting the two topology nodes based on the correlation index of the two topology nodes, and determine a deviation of the connection relationship of the topology edge in the network topology data from the connection relationship in an actual scenario based on the inconsistency score.

[0031] In a third aspect, the present application provides a computer device, comprising a memory and a processor, the memory storing a computer program, and the computer program being executed by the processor to enable the processor to perform any of the water supply network topology verification methods.

[0032] In a fourth aspect, the present application provides a computer readable storage medium, storing a computer program, and the computer program being executed to implement the water supply network topology verification method.

[0033] The water supply network topology verification method does not rely on a complete hydraulic model, improves the flexibility of water supply network topology verification, reduces the labor cost of water supply network topology verification, integrates network topology data and operation data of different data sources, avoids inconsistency of network topology data and operation data, and improves the accuracy and reliability of water supply network topology verification. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 A first flowchart of the water supply network topology verification in the embodiments of the present application;

[0035] Figure 2 A schematic diagram of the water supply network topology verification process in the embodiments of the present application;

[0036] Figure 3 A second flowchart of the water supply network topology verification in the embodiments of the present application;

[0037] Figure 4 A schematic diagram of the water supply network topology verification device in the embodiments of the present application;

[0038] Figure 5 A schematic diagram of the computer device in the embodiments of the present application. DETAILED DESCRIPTION

[0039] In order to enable those skilled in the art to better understand the present application, the technical solutions in the specific embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0040] It should be noted that the terms "first", "second" and similar terms used in the specification and claims of the present application do not denote any order, quantity, or importance, but are used to distinguish different components. Similarly, the terms "one" or "a" or similar terms do not denote a quantity limitation, but mean that at least one exists. "Multiple" or "several" means at least two. "Include" or "including" and similar terms mean that the elements or objects before the "include" or "including" are encompassed by the elements or objects listed after the "include" or "including", and their equivalents, and do not exclude other elements or objects. "Connected" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0041] As shown in Figure 1 As an implementation manner, the present application provides a water supply network topology verification method. The water supply network includes a plurality of topology nodes and topology edges connecting at least two topology nodes. The plurality of topology nodes include, but are not limited to, a water source node (such as a water plant, a reservoir, etc.), a user node (such as a residential area, an industrial area, etc.), and a regulating node (such as a water tower, a booster pump station, etc.). The topology edges include, but are not limited to, pipelines connecting the topology nodes. The water supply network topology verification method includes the following steps:

[0042] S101: Obtain the pipe network topology data of the water supply network of different data sources and the operation data of the topology nodes in the pipe network topology data.

[0043] In some embodiments, the different data sources include a GIS geographic information data source, a CAD computer-aided design data source, and a device list data source.

[0044] In some embodiments, the pipe network topology data includes the connection relationship of the topology nodes and the topology edges in the water supply network, the distribution of the topology nodes and the topology edges, and the branch level of the water supply network. The branch level of the water supply network refers to a hierarchical structure in which the topology nodes and the topology edges are divided into different levels according to different standards (such as service range, pressure range, etc.).

[0045] In some embodiments, the operation data includes the flow, pressure, and flow rate of the topology nodes at each time period. The operation data can represent the dynamic hydraulic behavior of the water supply network, which is the process of the flow, pressure, and flow rate changing over time due to various factors when the fluid flows in the water supply network.

[0046] For example, the various factors include external factors, which can be the rapid closing of a user-side valve (such as the sudden closing of a total water inlet valve in a certain area). If the user-side valve is rapidly closed, the pressure of the water supply pipeline in this section will increase sharply.

[0047] The plurality of factors also include internal factors, which can be corrosion of the inner wall of the pipeline. If the inner wall of the pipeline is corroded, the pressure upstream of the corrosion section rises, and the pressure downstream of the corrosion section drops.

[0048] The pipe network topology data and the operation data have different data types and different data sources. For example, a GIS geographic information data source can provide the connection relationship of the topological nodes and the topological edges in the water supply pipe network, the distribution of the topological nodes and the topological edges, and the branch level of the water supply pipe network; a CAD computer-aided design data source can provide a design drawing of the water supply pipe network; and a device list data source can provide the flow, pressure, and flow rate of the topological nodes in each time period.

[0049] In some embodiments, the pipe network topology data of the water supply pipe network of each data source and the operation data of the topological nodes in the pipe network topology data are uniformly formatted, so that the pipe network topology data of the water supply pipe network of each data source and the operation data of the topological nodes in the pipe network topology data are converted into a data structure suitable for subsequent processing, facilitating subsequent analysis.

[0050] Exemplarily, the pipe network topology data of each data source is converted into a unified image form. The converted image includes the topological nodes, the topological edges, and the connection relationship of the topological nodes and the topological edges.

[0051] Exemplarily, the operation data of the topological nodes in each pipe network topology data is converted into a unified matrix form. The converted matrix can represent the flow, pressure, and other data of the topological nodes in each pipe network topology data in different time periods.

[0052] In the implementation manner of the present application, the pipe network topology data and the operation data of each data source are cleaned, converted, and aligned to realize uniform formatting. The cleaning can identify errors, incompleteness, or inconsistency in the pipe network topology data and the operation data of each data source. The conversion can convert the pipe network topology data and the operation data of each data source into a format suitable for analysis or modeling. The alignment can integrate the pipe network topology data and the operation data of each data source into a unified format.

[0053] Exemplarily, the errors, incompleteness, or inconsistency in the pipe network topology data of each data source can be inconsistency in the connection relationship of the topological nodes and the topological edges. For example, an early version of the pipe network topology data shows that a water valve is connected to two pipelines, and a later version of the pipe network topology data shows that the water valve is connected to only one pipeline.

[0054] S102: Perform consistency checking on the pipe network topology data of different data sources to generate a preliminary checking result.

[0055] It should be noted that the preliminary verification result is computer-readable data, and the preliminary verification result can represent the consistency of the pipe network topology data of different data sources.

[0056] In some embodiments, the consistency verification can compare the pipe network topology data of different data sources after unified formatting, and detect structural contradictions caused by factors such as differences between different data sources, recording errors, or inconsistent versions. Among them, the structural contradiction refers to the conflict of the same topology node in the structural attribute, the location attribute, and the connection relationship.

[0057] For example, when comparing whether the structural attributes of the same topology node conflict, the connection relationship between nodes can be encoded into a matrix form, and the consistency can be calculated through the difference of the matrix elements, and the corresponding result can be obtained. When comparing whether the location attributes of the same topology node conflict, the node coordinate values (such as latitude and longitude, Cartesian coordinates) can be directly compared.

[0058] In some embodiments, the consistency verification is to determine the same topology node in different data sources according to the unique identification code of the topology node, and after determining the same topology node in different data sources, the connection relationship, location attribute, structural attribute, and / or classification attribute of the same topology node in different data sources are compared, wherein the structural attribute includes pipe diameter, length, and elevation, and the classification attribute includes material and current state.

[0059] For example, if the unique identification code of a topology node is α, it is determined whether the same topology node with the unique identification code α exists in the GIS geographic information data source and the CAD computer-aided design data source, and whether the topology edges connected to the topology node with the unique identification code α in the GIS geographic information data source and the CAD computer-aided design data source are consistent, and whether the structural attributes of the topology node with the unique identification code α and the topology edges connected thereto are consistent.

[0060] In some embodiments, the preliminary verification result is interpreted by a large language model to generate a preliminary structural difference report representing the structural differences of the pipe network topology data of different data sources.

[0061] S103: If the preliminary verification result is within the preset verification range, feature extraction is performed on the pipe network topology data to obtain structured features representing the connection relationship between the topology nodes and the topology edges, and / or the location of the topology nodes, and determine the first embedding vector of the topology node according to the structured features.

[0062] It should be noted that the preset verification range is a limitation on the consistency of the pipe network topology data of different data sources. If the consistency is high, the obtained preliminary verification result is within the preset verification range; if the consistency is low, the obtained preliminary verification result is outside the preset verification range.

[0063] In some embodiments, if the preliminary verification result is outside the preset verification range, the pipe network topology data corresponding to the preliminary verification result is cleared to avoid reducing the accuracy of subsequent calculations.

[0064] In some embodiments, the pipe network topology data of different data sources is converted into a unified image form. After consistency verification of all images, if the preliminary verification result is within the preset verification range, feature extraction is performed based on the image information and the specific application scenario, and the structured features extracted are normalized and preprocessed to obtain a first embedding vector.

[0065] S104: Obtain a time series feature representing the behavior pattern of the topology node according to the operation data, determine a second embedding vector of the topology node according to the time series feature, and combine the first embedding vector and the second embedding vector to obtain a fusion feature vector of the topology node.

[0066] The fusion feature vector can represent the correlation between the structured feature and the behavior pattern of the topology node, i.e., the fusion feature vector is a feature vector that can reflect the state of the topology node itself.

[0067] In some embodiments, in the process of obtaining the time series feature representing the behavior pattern of the topology node according to the operation data, first, the operation data of the topology node at multiple time nodes in a time period is obtained, and the behavior pattern of the topology node in the time period is obtained based on the operation data. The behavior pattern is the pressure change and / or flow change of the topology node in the time period. Then, the basic features and time series features of the pressure change and flow change are obtained. The basic features include the pressure mean, pressure variance, pressure maximum and pressure minimum of the topology node at each time node, and the flow mean, flow variance, flow maximum and flow minimum. The time series features include Fourier transform coefficients and autocorrelation coefficients. Finally, the basic features and time series features are normalized to obtain the time series feature.

[0068] The Fourier transform coefficients can represent the regularity of the time series feature from the perspective of frequency, and the autocorrelation coefficients can quantify the linear correlation degree of the time series feature at different time intervals.

[0069] In some embodiments, in the process of combining the first embedding vector and the second embedding vector to obtain the fusion feature vector of the topology node, first, the first embedding vector and the second embedding vector are spliced to obtain a preprocessed vector. Then, the preprocessed vector is calculated based on a linear layer to adjust the dimension of the preprocessed vector to obtain the fusion feature vector.

[0070] The fusion feature vector can comprehensively describe the comprehensive features of the state of the topology node itself. It should be noted that the vector is a computer language, and the above expression methods of the first embedding vector, the second embedding vector and the fusion feature vector are all for the convenience of computer recognition and calculation.

[0071] The preliminary checking result is obtained through the consistency check, and the data within the preset checking range is extracted again to obtain the fusion feature vector through the double-layer checking, which can improve the accuracy of the water supply network topology checking.

[0072] S105: Determine the correlation index of the two topology nodes according to the fusion feature vectors of the two topology nodes respectively, and determine the inconsistency score of the topology edge connecting the two topology nodes based on the correlation index of the two topology nodes.

[0073] The correlation index can represent the correlation between the fusion feature vectors of the two topology nodes respectively.

[0074] It should be noted that the inconsistency score can only represent consistency, that is, when the inconsistency score is higher than the preset score threshold, the connection relationship of the topology edge in the pipe network topology data and the connection relationship in the actual scene must have deviation.

[0075] In some embodiments, the higher the inconsistency score, the greater the deviation between the connection relationship of the topology edge in the pipe network topology data and the connection relationship in the actual scene.

[0076] The correlation index includes a Pearson correlation coefficient index of the running data of the two topology nodes respectively, and a difference index of the fusion feature vectors of the two topology nodes respectively.

[0077] The Pearson correlation coefficient index can quantify the linear relationship strength and direction of the running data of the two topology nodes respectively. If the Pearson correlation coefficient index is positively correlated, the running data of the two topology nodes changes in the same direction, and if the Pearson correlation coefficient index is negatively correlated, the running data of the two topology nodes changes in the opposite direction. The closer the absolute value of the Pearson correlation coefficient index is to 1, the stronger the correlation between the running data of the two topology nodes respectively; the closer the absolute value of the Pearson correlation coefficient index is to 0, the weaker the correlation between the running data of the two topology nodes respectively.

[0078] The difference index can quantify the difference between the fusion feature vectors of the two topology nodes respectively.

[0079] The weaker the correlation between the running data of the two topology nodes respectively reflected by the Pearson correlation coefficient index, the higher the inconsistency score based on the Pearson correlation coefficient index; the greater the difference between the fusion feature vectors of the two topology nodes respectively reflected by the difference index, the higher the inconsistency score based on the difference index.

[0080] S106: Determine the deviation between the connection relationship of topological edges in the pipeline topology data and the connection relationship in the actual scenario based on the inconsistency score.

[0081] In some embodiments, the higher the inconsistency score, the greater the deviation between the connection relationship of the topology edge in the pipeline topology data and the connection relationship in the actual scenario.

[0082] The above-mentioned water supply network topology verification method quantifies the consistency between two topology nodes by fusing their respective feature vectors. It does not rely on a complete hydraulic model, which improves the flexibility of water supply network topology verification, reduces the manual cost of water supply network topology verification, integrates network topology data and operational data from different data sources, avoids inconsistencies between network topology data and operational data, and improves the accuracy and reliability of water supply network topology verification.

[0083] like Figure 2 As shown, in some embodiments, inconsistency scores can be analyzed using a large language model to determine the deviation between the connectivity of topological edges in the pipeline topology data and the connectivity in the actual scenario.

[0084] Specifically, after acquiring pipeline topology data and operational data, both types of data are uniformly formatted to obtain a data structure that facilitates subsequent processing. Consistency checks enable comparative analysis of the uniformly formatted pipeline topology data from different data sources, detecting structural inconsistencies caused by differences in data sources, recording errors, or version discrepancies. Furthermore, feature extraction is performed on the structured features determined based on the pipeline topology data and the time-series features determined based on the operational data, followed by cross-validation to obtain a quantified inconsistency score. This inconsistency score is input into a large language model, which can then retrieve relevant information from a pre-stored database, along with the consistency check results, to generate the final verification result.

[0085] like Figure 3 As shown, in some possible implementations, generating the final verification result specifically includes the following steps:

[0086] S301: Obtain topological edges whose inconsistent scores are not less than a preset score threshold.

[0087] This involves identifying topological edges in the pipeline topology data that deviate significantly from the actual network topology.

[0088] S302: Transmit the association information of the topological edges to the terminal device.

[0089] The association information includes an inconsistency score, attribute information, and a preliminary verification result. The attribute information includes a connection relationship of the topological edge, a position attribute, a structure attribute, and / or a classification attribute. The structure attribute includes a pipe diameter, a length, and an elevation. The classification attribute includes a material and a current state.

[0090] S303: Based on a comparison between the database pre-stored in the terminal device and the association information of the topological edge, a final verification result is generated.

[0091] In some embodiments, the association information of the topological edge is input into a large language model. A retrieval enhancement generation technology is used to enable the large language model to retrieve relevant information (such as pipe network design specifications, common failure modes, historical maintenance records, etc.) from the database pre-stored in the terminal device before generating an answer. The large language model analyzes potential inconsistency reasons based on the association information and the relevant information, and generates an understandable and specific final verification result.

[0092] In some embodiments, the final verification result includes reasons for inconsistency between the connection relationship in the pipe network topology data and the connection relationship in the actual scene, and / or a repair suggestion list, and / or a report.

[0093] For example, two topological nodes are defined as node A and node B. If there is a pipe between node A and node B, but the pipe flow between node A and node B tends to be zero, and the pressure of node A and node B has no correlation, the final verification result can be “the pipe between node A and node B exists on the graph, but the flow pressure data shows that there is almost no hydraulic connection between them. It is suggested to check whether the pipe actually exists or the intermediate valve is permanently closed”.

[0094] Through the above setting, the large language model automatically generates guidance meaningful verification suggestions to guide the operation and maintenance personnel to carry out targeted troubleshooting, thereby effectively solving the repair problem of the pipe network topology structure.

[0095] As shown in FIG. 1, Figure 4 The water supply pipe network 100 includes a plurality of topological nodes 11 and topological edges 12 connecting at least two topological nodes 11. The present application also provides a water supply pipe network topology verification device 200. The water supply pipe network topology verification device 200 includes an acquisition module 21, a first verification module 22, a second verification module 23, and a calculation module 24.

[0096] The acquisition module 21 is configured to acquire pipe network topology data of the water supply pipe network 100 from different data sources, and operation data of the topological nodes 11 in the pipe network topology data.

[0097] The first verification module 22 is configured to perform consistency verification on the pipe network topology data from different data sources, and output a preliminary verification result.

[0098] The second verification module 23 is configured to, in a case where the preliminary verification result is within a preset verification range, perform feature extraction on the pipe network topology data to obtain a structured feature representing a connection relationship between the topology nodes 11 and the topology edges 12 and / or a position of the topology nodes 11, and determine a first embedding vector of the topology nodes 11 according to the structured feature. The second verification module 23 is further configured to obtain a time series feature representing a behavior pattern of the topology nodes 11 according to the operation data, and determine a second embedding vector of the topology nodes 11 according to the time series feature. The second verification module 23 can combine the first embedding vector and the second embedding vector to obtain a feature vector of the topology nodes 11.

[0099] The calculation module 24 is configured to determine a correlation index of two topology nodes 11 according to respective feature vectors of the two topology nodes 11, and determine an inconsistency score of a topology edge 12 connecting the two topology nodes 11 based on the correlation index of the two topology nodes 11. The inconsistency score is used to determine a deviation of the connection relationship of the topology edge 12 in the pipe network topology data from the connection relationship in the actual scenario.

[0100] In some embodiments, the topology nodes 11 include, but are not limited to, a water source node (such as a water plant, a reservoir, etc.), a user node (such as a residential area, an industrial area, etc.), and a regulating node (such as a water tower, a booster pump station, etc.), and the topology edges 12 include, but are not limited to, a pipe connecting the topology nodes 11.

[0101] In some embodiments, the pipe network topology data includes a connection relationship of the topology nodes 11 and the topology edges 12 in the water supply pipe network 100, a distribution of the topology nodes 11 and the topology edges 12, and a branch level of the water supply pipe network 100. The branch level of the water supply pipe network 100 refers to a hierarchical structure in which the topology nodes 11 and the topology edges 12 are divided into different levels according to different standards (such as a service range, a pressure range, etc.).

[0102] In some embodiments, the operation data includes a flow, a pressure, and a flow rate of the topology nodes 11 at each time period. The operation data can represent a dynamic hydraulic behavior of the water supply pipe network, which refers to a process in which the operation data changes over time due to various factors when a fluid flows in the water supply pipe network.

[0103] Through the above arrangement, the labor cost for verifying the topology of the water supply pipe network is reduced, the pipe network topology data and the operation data from different data sources are integrated, inconsistency of the pipe network topology data and the operation data is avoided, and the accuracy and reliability of the topology verification of the water supply pipe network are improved.

[0104] In some embodiments, the water supply network topology verification device 200 is in communication connection with the terminal device 300 through a communication network, the terminal device 300 includes a pre-stored database 31, the terminal device 300 can receive the association information of the topology edge 12 output by the water supply network topology verification device 200, and compare the association information of the topology edge 12 with the related information in the database 31 to generate a final verification result.

[0105] The association information includes an inconsistency score, attribute information, and a preliminary verification result, wherein the attribute information includes a connection relationship, a position attribute, a structure attribute, and / or a classification attribute of the topology edge 12, the structure attribute includes a pipe diameter, a length, and an elevation, and the classification attribute includes a material and a current state.

[0106] The related information includes but is not limited to a pipe network design specification, a common failure mode, and a historical maintenance record.

[0107] In some embodiments, the final verification result includes a reason and / or a repair suggestion list and / or a report for the inconsistency between the connection relationship in the pipe network topology data and the connection relationship in the actual scene.

[0108] Through the above setting, the association information of the topology edge 12 is analyzed and compared with the existing related information in the database 31, and the repair problem of the pipe network topology structure is effectively solved.

[0109] As shown in Figure 5 The present application also provides a computer device 400, which includes a memory 41 and a processor 42, the memory 41 stores a computer program, and the computer program is executed by the processor 42 to make the processor 42 execute the above-mentioned water supply network topology verification method.

[0110] In some embodiments, the computer device 400 further includes a communication unit 43, the communication unit 43 is connected with the memory 41 and the processor 42, and the communication unit 43 can realize the transmission of instructions and data between the memory 41 and the processor 42.

[0111] The present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed to realize the above-mentioned water supply network topology verification method.

[0112] The computer readable storage medium includes, but is not limited to, an electronic, magnetic, optical, infrared, or other physical storage device or apparatus, can contain or store information such as executable instructions, data, and the like. More specific examples of the computer readable storage medium include one or more wires, RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drives (such as hard drives), SSDs (Solid State Drives), any type of storage disks (such as optical disks), or similar storage, or any suitable combination thereof.

[0113] It should be understood that, for those of ordinary skill in the art, improvements or changes can be made according to the above description, and all these improvements and changes shall belong to the protection scope of the claims attached to this application.

Claims

1. A water supply network topology verification method, characterized in that, The method comprises: obtaining pipe network topology data of respective water supply networks of different data sources, and operation data of respective topology nodes in the pipe network topology data; performing consistency checking on the pipe network topology data of different data sources to generate a preliminary checking result; if the preliminary checking result is within a preset checking range, performing feature extraction on the pipe network topology data to obtain structured features representing connection relationships between the topology nodes and topology edges, and / or positions of the topology nodes, and determining a first embedding vector of the topology nodes according to the structured features; obtaining operation data of the topology nodes at multiple time nodes in a time period to obtain behavior patterns of the topology nodes in the time period, the behavior patterns being pressure changes and / or flow changes of the topology nodes in the time period; obtaining respective basic features and time sequence features of the pressure changes and the flow changes, and performing normalization processing on the basic features and the time sequence features to obtain time sequence features; determining a second embedding vector of the topology nodes according to the time sequence features, and combining the first embedding vector and the second embedding vector to obtain a fusion feature vector of the topology nodes; determining a correlation index of two topology nodes according to respective fusion feature vectors of the two topology nodes, and determining an inconsistency score of a topology edge connecting the two topology nodes based on the correlation index of the two topology nodes; determining a deviation of a connection relationship of the topology edge in the pipe network topology data from an actual connection relationship in a real scene based on the inconsistency score.

2. The method according to claim 1, wherein the consistency checking on the pipe network topology data of different data sources comprises: determining a same topology node in different data sources according to a unique identification code of the topology node; performing consistency comparison on connection relationships, position attributes, structure attributes and / or classification attributes of the same topology node in different data sources; the structure attributes include pipe diameter, length and elevation, and the classification attributes include material and current state.

3. The method according to claim 1, wherein the basic features include pressure mean value, pressure variance, pressure maximum value and pressure minimum value, and flow mean value, flow variance, flow maximum value and flow minimum value of the topology node at each time node, and the time sequence features include Fourier transform coefficients and autocorrelation coefficients.

4. The method according to claim 1, wherein the combining of the first embedding vector and the second embedding vector to obtain the fusion feature vector of the topology node comprises: splicing the first embedding vector and the second embedding vector to obtain a preprocessed vector; performing calculation on the preprocessed vector based on a linear layer to adjust dimensions of the preprocessed vector to obtain the fusion feature vector.

5. The method according to claim 1, wherein the method further comprises: ​ ​ ​ ​ obtain the topological edge with the inconsistency score not less than a preset score threshold; transmit associated information of the topological edge to a terminal device, the associated information including the inconsistency score, attribute information, and the preliminary verification result; wherein the attribute information includes connection relationship, location attribute, structure attribute, and / or classification attribute of the topological edge, the structure attribute including pipe diameter, length, and elevation, and the classification attribute including material and current state; compare the associated information of the topological edge with a database pre-stored in the terminal device to generate a final verification result.

6. The water supply network topology verification method according to claim 1, wherein the correlation index includes: Pearson correlation coefficient indicators of the operation data of the two topological nodes respectively; and difference indicators of the fusion feature vectors of the two topological nodes respectively.

7. The water supply network topology verification method according to claim 1, wherein the data sources include GIS geographic information data source, CAD computer-aided design data source, and equipment list data source; and the operation data includes pressure and flow. The water supply network topology verification device includes: an acquisition module configured to acquire pipe network topology data of a water supply network from different data sources, and operation data of the topological nodes in the pipe network topology data; 8. A water supply network topology verification device, the water supply network comprising a plurality of topology nodes and topology edges connecting at least two of the topology nodes, characterized in that, a first verification module configured to perform consistency verification on the pipe network topology data from different data sources, and output a preliminary verification result; a second verification module configured to, in the case that the preliminary verification result is within a preset verification range, perform feature extraction on the pipe network topology data to obtain structured features representing connection relationship of the topological nodes and / or location of the topological nodes, and determine a first embedding vector of the topological nodes according to the structured features; acquire operation data of the topological nodes at multiple time nodes in a time period to obtain behavior patterns of the topological nodes in the time period, the behavior patterns being pressure change and / or flow change of the topological nodes in the time period; acquire basic features and time sequence features of the pressure change and the flow change respectively, and perform normalization processing on the basic features and the time sequence features to obtain time sequence features; determine a second embedding vector of the topological nodes according to the time sequence features, and combine the first embedding vector and the second embedding vector to obtain a feature vector of the topological nodes; a calculation module configured to determine correlation indexes of the two topological nodes according to the feature vectors of the two topological nodes respectively, determine an inconsistency score of the topological edge connecting the two topological nodes based on the correlation indexes of the two topological nodes, and determine deviation of the connection relationship of the topological edge in the pipe network topology data from the connection relationship in an actual scene based on the inconsistency score.

9. A computer device, comprising: ​ ​ A computer program product, comprising a memory and a processor, wherein the memory stores a computer program, and the computer program is executed by the processor to enable the processor to perform the water supply network topology verification method according to any one of claims 1-7.

10. A computer readable storage medium, having stored thereon a computer program, wherein the computer program is executed to implement the water supply network topology verification method according to any one of claims 1-7. ​

Images