A sewer network mixed point identification method, electronic equipment and storage medium

Abstract

The present application relates to a kind of identification methods of sewer network mixed junction, electronic equipment and storage medium, comprising: step S1, obtain sewer network asset data, asset data is used to describe by node and pipe section Topology of sewer network structure;Step S2, according to starting node identification, end node identification and associated pipe section mark set, the connection between each pipe section and each node is parsed, to build by node and pipe section Topology connection chain alternately consisting of;Step S3, traverse Topology connection chain, and according to the pipe network type of each pipe section, the node of the component intersection of different pipe network types is identified, as candidate mixed junction;Step S4, from all candidate mixed junction, eliminate the node of pre-set design allowed intersection, obtain and output final mixed junction, so that human cost and time cost can be greatly reduced.

Description

Technical Field

[0001] This invention relates to the technical field of drainage pipe networks, and more particularly to a method for identifying mixed connection points in drainage pipe networks, an electronic device, and a storage medium. Background Technology

[0002] Drainage pipe networks are core infrastructure for urban water environment management and flood control, and their operational status directly affects urban water environment improvement, water ecological security, public safety, and residents' quality of life. With the acceleration of urbanization, problems such as lagging renovation of aging pipe networks, non-standard construction, and improper maintenance are becoming increasingly prominent, leading to widespread "mixed connections" in drainage pipe networks. Mixed connections in drainage pipe networks mainly refer to incorrect connections between rainwater and sewage systems, specifically manifested as sewage pipes connected to rainwater inspection wells / inlets, rainwater pipes connected to sewage inspection wells, and combined sewer pipes illegally connected to single-function pipe networks.

[0003] Furthermore, the problem of mixed connections in drainage pipe networks can cause a series of serious hazards: On the one hand, when sewage is mixed with rainwater and enters the stormwater system, the mixed sewage will be directly discharged into natural water bodies through the stormwater network, causing water pollution; on the other hand, when rainwater is mixed with sewage and enters the sewage system, the large amount of rainwater mixed into the sewage system will increase the treatment load of sewage treatment plants, leading to increased treatment costs, and even affecting treatment efficiency due to overload operation. In addition, mixed connections may also cause local siltation and blockage in the pipe network, reducing drainage capacity and exacerbating the risk of urban flooding. Therefore, accurately identifying mixed connection points (inspection wells, stormwater inlets) in the drainage pipe network is key to improving the quality and efficiency of the drainage system, improving the water environment, and preventing urban flooding.

[0004] Currently, existing solutions mainly rely on manual on-site inspections point by point. However, the number of inspection wells and storm drains in urban drainage networks typically reaches tens or even hundreds of thousands, and they are widely distributed. Some nodes are located in hard-to-reach areas such as busy traffic sections, green belts, and under buildings. This results in long inspection cycles and high labor costs, making it difficult to achieve full network coverage inspections, and especially unable to meet the needs of large-scale routine inspections of pipelines. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method, electronic device and storage medium for identifying mixed connection points in drainage pipe networks, which solves the technical problem of low investigation efficiency in the prior art.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0009] In a first aspect, embodiments of the present invention provide a method for identifying mixed connection points in a drainage network, comprising: step S1, acquiring drainage network asset data, wherein the asset data is used to describe the drainage network topology structure composed of nodes and pipe segments; wherein the pipe segment data includes at least: a starting node identifier and an ending node identifier used to indicate the network type of the drainage system to which the pipe segment belongs, and a starting node identifier and an ending node identifier used to determine the connection relationship of the pipe segment in the drainage network topology structure; the node data includes at least: a set of associated pipe segment markers used to determine the associated pipe segments of the node in the drainage network topology structure; step S2, parsing the connection relationship between each pipe segment and each node according to the starting node identifier, the ending node identifier, and the set of associated pipe segment markers, to construct a topological connection chain composed of alternating nodes and pipe segments; step S3, traversing the topological connection chain, and identifying nodes where components of different network types intersect according to the network type of each pipe segment, as candidate mixed connection points; step S4, removing preset design-allowed intersection nodes from all candidate mixed connection points, and obtaining and outputting the final mixed connection point.

[0010] Optionally, the node data also includes: node types for distinguishing between manholes and storm drains; the specific rules for identifying candidate mixed connection points in step S3 include: for a node of type manhole, if its associated pipe segment contains both storm drain pipe and sewage pipe, or if its associated pipe segment contains a combined drain pipe and contains both storm drain pipe and sewage pipe, then the manhole is regarded as a candidate mixed connection point; for a node of type storm drain, if its associated pipe segment contains a sewage pipe or combined drain pipe flowing into the storm drain, or contains a sewage pipe flowing out of the storm drain, then the storm drain is regarded as a candidate mixed connection point.

[0011] Optionally, the pipe segment data also includes the pipe diameter of the pipe segment; after step S4, the following steps are also included: for the final mixing point, determining the pipe diameter of the mixing pipe and its mixing type;

[0012] For nodes that serve as the final mixed connection point and whose node type is a manhole, the pipe segments associated with that node are grouped according to the pipe network type. For each group of pipe segments, if all pipe segments in the group flow into that node or all pipe segments flow out of that node, then the pipe segments in that group are considered mixed pipes, and the diameter of the mixed pipe is taken from the diameter of the pipe segments in that group. The pipe segments associated with that node that have both inflow and outflow pipe network types are considered normal pipe segments. The mixed connection type is that the pipe network type of each group of mixed pipes is connected to the pipe network type of the normal pipe segment.

[0013] Optionally, the pipe segment data also includes the pipe diameter of the pipe segment; after step S4, the following steps are also included: for the final mixing point, determining the pipe diameter of the mixing pipe and its mixing type;

[0014] For a node that is the final connection point and is a storm drain, all inflow pipes of the storm drain that belong to sewage pipes or combined sewers are designated as combined sewer pipes, with the pipe diameter of the combined sewer pipe being the same as that pipe segment, and the connection type being the combined sewer pipe network type connected to the storm drain pipe; and all outflow pipes of the storm drain that belong to sewage pipes are designated as combined sewer pipes, with the pipe diameter of the combined sewer pipe being the same as that pipe segment, and the connection type being the storm drain pipe network type connected to the combined sewer pipe network.

[0015] Optionally, the node data also includes a node identifier for uniquely identifying the node and the node's location coordinates; the identification method further includes: marking and displaying the final cross-connection point on an electronic map based on the location coordinates, and at least displaying the node identifier of the final cross-connection point; and, when available, displaying the cross-connection type of the final cross-connection point and / or the pipe segment identifier corresponding to the cross-connection pipe.

[0016] Optionally, the identification method further includes: after step S3 and before step S4, the identification method further includes: acquiring monitoring data of candidate mixing points during non-rainfall periods; the monitoring data includes at least one of flow data and water quality data; for candidate mixing points belonging to the stormwater pipe network, if there is a stable flow and the water quality pollution index exceeds the standard in the monitoring data, the candidate mixing point is marked as a high-confidence mixing point; for candidate mixing points belonging to the sewage pipe network, acquiring their monitoring data during rainfall periods, if it is determined through monitoring that there is an increase in flow or a decrease in water quality pollution index, the candidate mixing point is marked as a high-confidence mixing point; in step S4, pre-set design-allowed intersection nodes are removed from all candidate mixing points, wherein design-allowed intersection nodes are removed even if they are marked as high-confidence mixing points; the high-confidence mixing points among the remaining candidate mixing points are taken as the final mixing points.

[0017] Optionally, step S3 includes: extracting the topological features of each node from the topological connection chain; the topological features include at least one or more of the following: the number of pipe segments associated with the node, the distribution of pipe network types of each associated pipe segment, the flow direction relationship of each associated pipe segment, and the position level of the node in the topological connection chain; determining the confidence level of each node as a candidate hybrid connection point based on the topological features of each node; and selecting nodes with a confidence level greater than a preset threshold as candidate hybrid connection points.

[0018] In a second aspect, embodiments of the present invention provide an electronic device, including a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to implement a method for identifying mixed connection points of a drainage network as described in any one aspect by executing the executable instructions.

[0019] Thirdly, embodiments of the present invention provide a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for identifying mixed connection points of drainage pipe networks as described in any one of the first aspects.

[0020] (III) Beneficial Effects

[0021] The beneficial effects of this invention are:

[0022] This application provides a method, electronic device, and storage medium for identifying mixed connection points in drainage pipe networks. By acquiring drainage pipe network asset data, which describes the drainage pipe network topology composed of nodes and pipe segments, the pipe segment data includes at least: a pipe network type representing the drainage system to which the pipe segment belongs, and a start node identifier and an end node identifier for determining the connection relationship of the pipe segment in the drainage pipe network topology. The node data includes at least: a set of associated pipe segment markers for determining the associated pipe segments of the node in the drainage pipe network topology, and a set of markers based on the start node identifier, end node identifier, and associated pipe segment markers. The system uses a segment tag set to parse the connection relationships between each pipe segment and each node, constructing a topological connection chain composed of alternating nodes and pipe segments. It then traverses this topological connection chain and identifies nodes where components of different pipe network types intersect, based on the pipe network type of each pipe segment. These nodes are then used as candidate mixed connection points. Finally, it removes pre-defined design-allowed intersection nodes from all candidate mixed connection points, obtaining and outputting the final mixed connection points. This automated parsing of pipe network topology relationships enables mixed connection verification without the need for manual on-site inspection. It can complete the identification of mixed connections across the entire pipe network in a short time, significantly reducing labor and time costs and enabling full-network coverage inspection. Attached Figure Description

[0023] Figure 1 A flowchart of a method for identifying mixed connection points in a drainage pipe network provided in this application is shown;

[0024] Figure 2 This application provides a schematic diagram of a manhole connection for connecting a sewage pipe to a rainwater pipe.

[0025] Figure 3 This application provides a schematic diagram of a manhole connection for connecting a rainwater pipe to a sewage pipe.

[0026] Figure 4 This application provides a schematic diagram of a manhole connection for a combined sewer pipe to a rainwater pipe.

[0027] Figure 5 This application provides a schematic diagram of a rainwater inlet connection for connecting a sewage pipe to a rainwater pipe.

[0028] Figure 6 This application provides a schematic diagram of a rainwater inlet connection for connecting a rainwater pipe to a sewage pipe.

[0029] Figure 7 This application provides a schematic diagram of a rainwater inlet connection for a combined drain pipe connected to a rainwater pipe. Detailed Implementation

[0030] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] Besides manual point-to-point inspection, another approach combines manual on-site inspection with testing equipment. The core idea of ​​this method is for staff to carry specialized testing equipment (such as pipe endoscopes, water quality analyzers, and flow monitors) and conduct on-site inspections of key nodes in the drainage network, such as manholes, storm drains, and pipe joints, following a pre-set network inspection route. To identify these mixed connection issues, specialized source tracing methods are often required. Currently, methods such as QV testing, CCTV pipeline detection, dye tracing, and pollutant concentration detection are mainly used. These methods have demanding testing conditions, requiring preliminary work such as dewatering, dredging, and unblocking, resulting in high costs and low efficiency. Another approach involves sampling manholes in complex areas of the storm drain network and using three-dimensional fluorescence spectroscopy to characterize the organic matter in the drainage pipes through three-dimensional fluorescence spectroscopy and ultraviolet-visible absorption spectroscopy. However, this method is often limited in practical applications due to its low resolution in identifying different types of domestic sewage.

[0032] The implementation process of the existing technical solution is roughly as follows: Preliminary preparation: Collect basic information such as drainage pipe network planning drawings and as-built data for the target area, and delineate the scope of investigation and key areas (such as old residential areas, areas surrounding industrial parks, and areas along water bodies where cross-connections are likely to occur); On-site verification: After arriving at the site, staff first preliminarily determine the pipe network type (sewage / rainwater) by checking manhole cover markings and pipe network route markings. Then, they open the manholes and use pipe endoscopes to observe the pipe connections and check for any unplanned connection pipes. Simultaneously, they use a water quality analyzer to test the COD of the water in the manholes. Pollution indicators such as ammonia nitrogen and total phosphorus are detected. If the pollution indicators in the rainwater pipe network exceed the standards, or if the water in the sewage pipe network shows abnormalities such as a sudden increase in flow or clearing of water quality during non-rainfall periods, it is preliminarily determined that there is a suspicion of cross-connection. Result confirmation: For nodes suspected of cross-connection, the source of the connecting pipe is traced (such as checking the drainage outlets of surrounding residential buildings, shops, and factories) to finally confirm the specific location, type of cross-connection, and source of pollution. Recording and archiving: The location, type, and rectification suggestions of the cross-connection points are manually recorded and marked on a paper map or a simple electronic map to form an investigation report.

[0033] However, the above methods require manual on-site inspections and testing equipment such as QV / CCTV, flow rate, and water quality sensors, resulting in high costs and long implementation cycles. With the continuous expansion of urban pipeline networks (large urban pipeline networks often reach tens of thousands of kilometers in length), the above methods are no longer sufficient to meet the demands for efficient, accurate, and comprehensive testing. There is an urgent need to develop automated verification technology based on pipeline topology to achieve rapid identification and location of mixed-connection nodes.

[0034] In addition, the existing solutions also have the following problems:

[0035] The accuracy of investigations relies heavily on human experience, making it prone to missed or incorrect detections. Identifying mixed connections depends on staff's understanding of pipeline diagrams, identification of on-site nodes, and interpretation of test data. For older pipeline networks (with missing diagrams or blurred markings) and complex intersecting networks (multiple pipe segments converging), manual judgment is prone to errors. Furthermore, the interpretation of results from water quality testing and endoscopic observations is subjective, potentially leading to missed (e.g., concealed access pipes going undetected) or incorrect (e.g., stormwater networks with excessive water quality due to initial rainwater contamination being mistakenly identified as mixed connections).

[0036] The display of mixed connection points is not intuitive and makes it difficult to support accurate rectification and management: The existing solutions record mixed connection points mostly in text description or simple paper map markings, which cannot be linked to digital pipeline asset information, nor can they intuitively display the spatial location of mixed connection points and the surrounding pipeline topology on electronic maps. This results in a lack of accurate spatial data support for subsequent rectification plan formulation, rectification effect tracking, pipeline operation and maintenance management, etc., leading to low efficiency.

[0037] Unable to achieve dynamic verification and poor timeliness: The existing solution is a "one-time static inspection". Misconnection problems that may occur during pipeline renovation and maintenance are difficult to be identified in time through static inspection, resulting in repeated mixed connection problems and failing to achieve normalized and dynamic mixed connection monitoring.

[0038] Data cannot be reused and collaboration is poor: the mixed data recorded by manual investigation are mostly isolated reports that cannot be connected with the city's existing digital platforms such as drainage network asset information system and water environment monitoring system. The data is difficult to reuse and cannot provide continuous data support for subsequent network optimization and water environment governance decisions, forming "data silos".

[0039] Based on this, embodiments of this application provide a method, electronic device, and storage medium for identifying mixed connection points in drainage pipe networks. By acquiring drainage pipe network asset data, which describes the drainage pipe network topology composed of nodes and pipe segments, the pipe segment data includes at least: a pipe network type representing the drainage system to which the pipe segment belongs, and a starting node identifier and an ending node identifier for determining the connection relationship of the pipe segment in the drainage pipe network topology. The node data includes at least: a set of associated pipe segment markers for determining the associated pipe segments of the node in the drainage pipe network topology, and parsing each node's associated pipe segment markers based on the starting node identifier, ending node identifier, and associated pipe segment marker set. The system automatically analyzes the network topology to verify the cross-connection of pipe segments and nodes. This eliminates the need for manual on-site verification and allows for the identification of cross-connection points across the entire network in a short time (e.g., verification of a network with hundreds of thousands of nodes can be completed in tens of seconds). This significantly reduces labor and time costs and enables full network coverage investigation.

[0040] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0041] Please see Figure 1 , Figure 1 A flowchart illustrating a method for identifying mixed connection points in a drainage pipe network according to this application is shown. It should be understood that this identification method can be executed by an electronic device, and the specific device of the electronic device can be configured according to actual needs; the embodiments of this application are not limited thereto. For example, the electronic device can be a computer or a server, etc. Specifically, the identification method includes the following steps:

[0042] Step S110: Obtain drainage network asset data, which describes the drainage network topology consisting of nodes and pipe segments. The pipe segment data includes at least: a pipe network type indicating the drainage system to which the pipe segment belongs, and a start node identifier and an end node identifier used to determine the connection relationship of the pipe segment in the drainage network topology. The node data includes at least: a set of associated pipe segment tags used to determine the associated pipe segments of the node in the drainage network topology.

[0043] It should be understood that the data contained in this pipe section can be set according to actual needs, and those skilled in the art are not limited thereto.

[0044] Optionally, the data for each drainage pipe segment shall include at least the following fields:

[0045] Pipe segment identifier: An ID number used to uniquely identify the pipe segment;

[0046] Pipe diameter: The diameter of this pipe section;

[0047] Pipeline type: This indicates the type of drainage system to which this pipe section belongs, including rainwater pipe, sewage pipe, or combined sewer.

[0048] Starting point node identifier: The ID number of the node (inspection well or rainwater inlet) connected to the upstream end of this pipe section;

[0049] End point identifier: The ID number of the node (inspection well or rainwater inlet) connected to the downstream end of this pipe section;

[0050] In addition, it may include attribute data such as pipe material, slope, length, and burial depth.

[0051] It should also be understood that the data contained in this node can be configured according to actual needs, and those skilled in the art are not limited thereto.

[0052] Optionally, the data for each node (including inspection wells and storm drains) shall include at least the following fields:

[0053] Node identifier: An ID number used to uniquely identify the node;

[0054] Node type: A category identifier used to distinguish between inspection wells and storm drains;

[0055] Location coordinates: The geographic coordinates (such as latitude and longitude) of the node;

[0056] Pipeline type: The type of drainage system to which this node belongs, including rainwater wells, sewage wells, or combined sewer wells;

[0057] Associated pipe segment tag set: a list of IDs of all pipe segments connected to this node (i.e., it includes the ID of at least one pipe segment), used to describe the association between the node and the pipe segments;

[0058] In addition, it may include attribute data such as well depth, manhole cover type, and ground elevation.

[0059] The data of the aforementioned pipe segments and nodes together constitute the topology of the drainage network. That is, by using the starting / ending node identifiers of the pipe segments and the set of associated pipe segment tags of the nodes, the connection relationship between "node-pipe segment-node" can be fully described, providing a data foundation for the subsequent construction of the topological connection chain.

[0060] To facilitate understanding of step S110, it will be described below through specific embodiments.

[0061] Specifically, this method can be implemented by a digital system combining software and hardware, relying on the existing drainage network asset information system and adding a mixed connection verification module to achieve its core function. Furthermore, the system architecture can adopt a layered design, consisting of a data acquisition layer, a data storage layer, a core verification layer, and a display application layer, with each layer interacting bidirectionally through data interfaces. The core verification layer is the core of this system, containing three sub-modules: a topology relationship parsing module, a mixed connection rule matching module, and a result verification module.

[0062] In addition, this data acquisition layer is used to acquire basic data from the drainage network asset information system, providing a data source for subsequent verification. The data it collects includes two core types: static topology data and dynamically updated data. The static topology data originates from the existing drainage network asset information system and is the fundamental data describing the inherent structure of the network, specifically including:

[0063] Drainage pipe segment data includes pipe segment ID, pipe diameter, type of pipe network to which it belongs (such as sewage pipe, rainwater pipe and combined pipe), starting node ID, ending node ID, etc.

[0064] Inspection well data includes inspection well ID, location coordinates (latitude and longitude), type of pipeline network to which it belongs, and a list of associated pipeline segment IDs.

[0065] Rainwater inlet data includes rainwater inlet ID, location coordinates, type of pipe network to which it belongs, and a list of associated pipe segment IDs.

[0066] In addition, dynamically updated data is used to reflect real-time changes in the pipeline network. This data is obtained through integration with the drainage pipeline network asset information system and includes:

[0067] Pipeline renovation construction records: including information on newly added, deleted, or modified pipe sections and node information;

[0068] Drainage user connection approval record: including information on the connection pipe section of the newly added drainage outlet, etc.

[0069] It should be noted that when this system identifies mixed connection points, it first executes steps S110 to S140 based on the currently acquired static topology data to complete the identification. When the data acquisition layer detects dynamically updated data access, the system automatically triggers dynamic update identification, that is, it re-executes steps S110 to S140 based on the updated drainage network asset data to identify mixed connection points in the updated network, thereby achieving normalized and dynamic monitoring of mixed connection issues.

[0070] Furthermore, this data storage layer is connected to the data acquisition layer, and it is used to store the acquired drainage network asset data, as well as intermediate data generated during the verification process and the final verification result data. This layer can adopt a hybrid storage method of relational database and spatial database, which supports both efficient querying of structured data and visualization applications of spatial data.

[0071] Step S120: Based on the starting node identifier, ending node identifier, and associated pipe segment marker set, parse the connection relationship between each pipe segment and each node to construct a topological connection chain composed of alternating nodes and pipe segments.

[0072] Specifically, firstly, all node data (including inspection wells and rainwater inlets) and pipe segment data obtained in step S110 are loaded into memory, and hash index tables are built using node identifiers and pipe segment identifiers as keys to quickly retrieve complete information about nodes and pipe segments. Based on this, the system traverses all nodes, selecting nodes with 1 associated pipe segment as the traversal starting point and storing them in the starting point queue. If a loop exists in the pipeline network or all nodes have more than 1 associated pipe segment, a node is randomly selected as the starting point, and visited nodes are recorded to prevent repeated traversal.

[0073] After retrieving a node from the starting queue as the current node, a depth-first traversal is initiated to construct the topology connection chain. First, the system reads the set of associated pipe segment markers from the node index table based on the current node's identifier. Then, it traverses this marker set to find pipe segments not yet added to the current chain. If multiple unvisited pipe segments exist, it indicates a branch exists. The system records the current state and creates a new independent chain for each branch for subsequent construction. Assuming an unvisited pipe segment is found, it is designated as the current segment, and its starting and ending node identifiers are read from the pipe segment index table. Based on the matching relationship between the current node and the pipe segment endpoints, the next node is determined. If the current node is the starting point of the pipe segment, the next node is the ending point; otherwise, it is the starting point. Next, the current node and the current pipe segment are sequentially recorded in the list of topology connection chains under construction, and the determined next node is set as the new current node. This process is repeated until all pipe segments in the current node's associated pipe segment marker set have been visited. At this point, a complete topology connection chain is constructed, and the system stores it and marks all nodes and pipe segments in the chain as belonging to the current node.

[0074] The above process is repeated until all starting points in the starting queue and all nodes and pipe segments not marked as belonging have been processed, ultimately resulting in one or more topological connection chains covering the entire drainage network. Finally, based on the raw data obtained in step S110, the system assigns a corresponding network type label (rainwater pipe, sewage pipe, or combined sewer) to each pipe segment in each chain, and marks the network type of the nodes for subsequent mixed connection rule judgment. Through the above steps, the original, mutually referencing node identifiers and pipe segment identifiers are transformed into intuitive, continuous "node-pipe segment-node" topological connection chains.

[0075] To facilitate understanding of step S120, it will be described below through specific embodiments.

[0076] Specifically, the core verification layer of the system is the core functional layer of this system, connected to the data storage layer, and is responsible for the automatic identification and verification of mixed connection points. Among them, the topology relationship parsing module reads the associated data of drainage pipe sections, inspection wells, and rainwater inlets from the data storage layer, and automatically parses and forms a "node-pipe section-node" topology connection chain based on the starting node identifier and ending node identifier of the pipe section and the associated pipe section label set of the node, and marks the pipe network type of each component in the topology connection chain.

[0077] Step S130: Traverse the topology connection chain and identify nodes where components of different pipe network types intersect, based on the pipe network type of each pipe segment, as candidate mixed connection points.

[0078] It should be understood that the specific rules for identifying candidate mixed connection points can be set according to actual needs, and the embodiments of this application are not limited thereto.

[0079] Optionally, the node data also includes: node types for distinguishing between manholes and storm drains; the specific rules for identifying candidate mixed connection points in step S130 include: for a node of type manhole, if its associated pipe segment contains both storm drain pipe and sewage pipe, or if its associated pipe segment contains a combined drain pipe and contains both storm drain pipe and sewage pipe, then the manhole is regarded as a candidate mixed connection point; for a node of type storm drain, if its associated pipe segment contains a sewage pipe or combined drain pipe flowing into the storm drain, or contains a sewage pipe flowing out of the storm drain, then the storm drain is regarded as a candidate mixed connection point.

[0080] For example, the mixing points of a drainage network include mixed manholes and mixed rainwater inlets.

[0081] Among them, inspection wells that meet one of the following conditions are defined as mixed connection points:

[0082] The inspection well is connected to both rainwater pipes and sewage pipes;

[0083] The inspection well has a combined sewer pipe, and rainwater or sewage pipes are connected and disconnected.

[0084] Furthermore, a rainwater inlet is defined as a mixed connection point if it falls under one of the following conditions:

[0085] A sewage pipe or combined sewer pipe is connected to the storm drain.

[0086] A sewage pipe is connected to the rainwater inlet.

[0087] Furthermore, it should be noted that the method described above for determining candidate hybrid connection points based on specific rules is suitable for scenarios with clear pipeline types and standardized topological relationships, and has the advantages of simple calculation and strong interpretability. However, for complex pipeline network scenarios, the following method can also be used:

[0088] Extract the topological features of each node from the topological connection chain; the topological features include at least one or more of the following: the number of pipe segments associated with the node, the distribution of pipe network types of each associated pipe segment, the flow direction relationship of each associated pipe segment, and the position level of the node in the topological connection chain.

[0089] Based on the topological features of each node, the confidence level of each node as a candidate hybrid connection point is determined. For example, the topological features of each node are input into a machine learning model, which then outputs the confidence level of each node as a candidate hybrid connection point. Alternatively, existing algorithms can also be used to determine the confidence level of each node as a candidate hybrid connection point.

[0090] Nodes with a confidence level greater than a preset threshold are selected as candidate hybrid connection points. The specific value of the preset threshold can be set according to actual needs, and this embodiment is not limited thereto.

[0091] Therefore, this method can adapt to various types of pipe segment intersections and irregular topological relationships by learning from historical data, and has stronger adaptability.

[0092] It should also be understood that the specific model and its structure of the machine learning model can be set according to actual needs, and the embodiments of this application are not limited thereto.

[0093] Optionally, step S130 can employ a deep learning-based physical sensing dual-channel heterogeneous network (PADCH-Net) to identify candidate mixed connection points. This model is interchangeable with the aforementioned rule-based judgment scheme based on node type and pipe segment type (see related content above) and is suitable for complex pipe network scenarios. Furthermore, this physical sensing dual-channel heterogeneous network model can consist of a feature extraction layer, a dual-channel input layer, an adjacency matrix construction layer, a flow direction sensing gating message passing layer, a multi-scale feature fusion layer, and a confidence output layer. These modules are sequentially connected to form an end-to-end trainable network. The final output of the model is a confidence vector for each node belonging to a candidate mixed connection point.

[0094] First, this feature extraction layer is responsible for transforming the topological connection chain constructed in step S120 into an initial feature vector for each node. The input to this feature extraction layer is the topological connection chain constructed in step S120 (i.e., an ordered sequence of alternating nodes and pipe segments), which fully describes the type of each pipe segment (rainwater pipe, sewage pipe, or combined sewer), the flow direction (from the starting node to the ending node), and the connection relationships between nodes.

[0095] Internally, the feature extraction layer traverses each topological connection chain, calculating an initial feature vector for each node in the chain, specifically including at least one of the following features:

[0096] Number of pipe segments associated with a node: This counts the total number of all pipe segments connected to the node, reflecting the density of the node's connections in the pipeline network.

[0097] Pipeline network type distribution for each associated pipe segment: The number of pipe segments associated with each node is counted according to the pipeline network type (rainwater pipe, sewage pipe, combined sewer), forming a type distribution vector. For manhole nodes, inflow and outflow are further distinguished, and the inflow and outflow numbers of each type of pipe segment are counted separately; for rainwater inlet nodes, the number of each type of pipe segment connecting to and disconnecting from the inlet are counted separately.

[0098] Flow direction relationship among related pipe segments: Combined with the distribution characteristics of pipe network types, the flow direction information of each pipe segment relative to the node is encoded by distinguishing inflow and outflow directions. Specifically, this is represented by a quantity vector statistically analyzed by type and direction;

[0099] The node's position level in the topology connection chain: Calculate the node's position index in the topology connection chain, such as the number of hops from the chain start point, or the normalized relative position value, reflecting the node's relative position in the global network structure.

[0100] In addition, if the data contains physical attributes of the nodes (such as the depth of inspection wells), these can be added as optional features to enrich the node representation.

[0101] Through the above processing, each topological connection chain is transformed into a series of node feature vectors, each feature vector fully containing at least one of the four types of topological features. The feature vectors of all nodes together constitute a node feature matrix, which serves as the input for subsequent modules.

[0102] Furthermore, considering that drainage networks contain two distinct types of nodes: manholes and storm drains, manholes are located at underground pipe junctions, typically connecting multiple pipe segments and forming the core nodes of the network. Storm drains, on the other hand, are located on the surface, primarily collecting rainwater from the road surface, and usually connect only a few pipe segments. Treating these two types of nodes the same and using the same feature transformation network would result in the loss of significant domain knowledge. Therefore, this dual-channel input layer provides separate feature transformation channels for each type of node.

[0103] The input to this dual-channel input layer is the node feature matrix output from the feature extraction layer, where each node's feature vector carries its type identifier (manhole or storm drain). Internally, the system first splits the feature matrix into two subsets based on node type: a manhole feature subset and a storm drain feature subset. Then, the manhole feature subset is fed into the manhole channel, which consists of a multi-layer fully connected network that performs a non-linear transformation on the manhole features to extract deep, abstract features. The storm drain feature subset is fed into the storm drain channel, where a separate, independent fully connected network performs the same operation. The network parameters of the two channels are not shared; each learns the inherent patterns of its corresponding node type.

[0104] After transformation through their respective channels, the features of the two types of nodes are mapped to the same dimensional space (denoted as d_hidden, which can be set to 128 dimensions in this embodiment), and then merged into a unified node feature matrix, denoted as H0 (initial unified feature matrix), with a dimension of n×d_hidden, where n is the total number of nodes. This unified feature space design enables the subsequent message passing layer to process all nodes in a unified manner, preserving the individual characteristics of the two types of nodes while laying the foundation for global information interaction.

[0105] Furthermore, before entering the message passing layer, the topological connection chain constructed in step S120 needs to be transformed into the adjacency matrix form required by the graph neural network. This transformation is performed by the adjacency matrix construction layer. The input to this adjacency matrix construction layer is the topological connection chain constructed in step S120. The topological connection chain is an ordered sequence of alternating nodes and pipe segments, containing information on the type (rainwater pipe, sewage pipe, or combined sewer) and flow direction (from the starting node to the ending node) of each pipe segment.

[0106] Internally, the system traverses each topological connection chain and constructs six different adjacency matrices based on the pipe segment type and flow direction. For each pipe segment, the system first identifies its type and then determines whether it is an inflow or outflow relationship based on its direction: if the pipe segment flows from node A to node B, then for node B it is an inflow pipe segment; for node A it is an outflow pipe segment. Based on this, the system records the pipe segments in the corresponding adjacency matrices. The definitions of the six adjacency matrices are as follows:

[0107] Rainwater inflow adjacency matrix: records the connection relationships of rainwater flowing into a node through the rainwater pipe;

[0108] Rainwater outflow adjacency matrix: records the connection relationships of rainwater flowing out of a node through a rainwater pipe;

[0109] Sewage inflow adjacency matrix: records the connection relationships of sewage flowing into a node through the sewage pipe;

[0110] Sewage outflow adjacency matrix: records the connection relationships of sewage flowing out of a node through a sewage pipe;

[0111] Merging Inflow Adjacency Matrix: Records the connection relationships of a node flowing into a node through a merging pipe;

[0112] Merging-outflow adjacency matrix: records the connection relationships of flows out of a certain node through the merging pipe.

[0113] Each adjacency matrix is ​​of size n×n. If there is a pipe segment connection of corresponding type and direction between node i and node j, the element at the corresponding position in the matrix is ​​1; otherwise, it is 0. Through this processing, the original topological connection chain is transformed into six heterogeneous adjacency matrices, providing a precise topological information foundation for the subsequent flow-aware gating message passing layer, enabling the model to learn the feature patterns of different types of connection relationships.

[0114] Furthermore, this flow-aware gating message passing layer is used to transmit information between nodes and gradually aggregate neighbor features. This flow-aware gating message passing layer fully considers the physical meaning of water flow direction in the drainage network and the impact of different pipe segment types on mixed connection judgment, and designs a message passing method based on a gating mechanism, enabling the model to automatically focus on abnormal connection relationships.

[0115] The input to this layer consists of two parts: first, the unified node feature matrix H0 output from the dual-channel input layer; and second, the six heterogeneous adjacency matrices output from the adjacency matrix construction layer. Internally, this layer employs a stacked multi-layer structure; in this embodiment, three layers can be stacked, referred to as the first message passing layer, the second message passing layer, and the third message passing layer. Each layer has the same structure but independent parameters. Through layer-by-layer transmission, each node can progressively aggregate neighbor information within its multi-hop range.

[0116] The first message passing layer takes H0 and six adjacency matrices as input. For each node i, it first finds its neighbor node sets for various pipe segment types and directions based on the six adjacency matrices (e.g., rainwater inflow neighbors, rainwater outflow neighbors, etc.). Then, for each group of neighbors, a gating coefficient is calculated through a dedicated gating unit. This gating unit takes the current features of the central node i and the aggregated features of the group of neighbors as input, and dynamically outputs a gating coefficient between 0 and 1 through a multilayer perceptron. The physical meaning of the gating coefficient is: when the pipe segment type and flow direction represented by the group of neighbors conform to normal drainage patterns (e.g., sewage pipe connected to sewage well and the flow direction is correct), the gating coefficient is small, restricting the transmission of this group of information to the central node; when the group of neighbors is suspected of cross-connection (e.g., sewage pipe connected to rainwater well), the gating coefficient is large, allowing the group of information to be amplified and transmitted to the central node. This mechanism enables the model to automatically learn and identify which connections are abnormal, thus focusing on the key to cross-connection identification.

[0117] Next, the features of each group of neighbors are weighted and aggregated according to the gating coefficient to obtain the aggregated neighbor information. This aggregated information is then fused with the features of the central node itself (after linear transformation) and processed by a nonlinear activation function (such as ReLU) to obtain the updated features of node i. All nodes perform the above operations in parallel. The feature matrix output by the first message passing layer is denoted as H1 (the first layer output feature matrix), with a dimension of n×d_hidden.

[0118] The second message passing layer takes the output H1 from the first layer and six adjacency matrices as input and repeats the above process. Since the first layer has already aggregated the information of direct neighbors, the second layer further aggregates the information of two-hop neighbors, and the output feature matrix is ​​denoted as H2 (the output feature matrix of the second layer), with a dimension of n×d_hidden.

[0119] The third message passing layer takes the output H2 from the second layer and six adjacency matrices as input, and repeats the above process again, extending the information passing scope to a wider topological context. The output feature matrix is ​​denoted as H3 (the output feature matrix of the third layer), with a dimension of n×d_hidden.

[0120] Through three layers of stacking, each node obtains feature representations at three different levels of abstraction: H1 mainly contains local information from direct neighbors, H2 incorporates structural information from two-hop neighbors, and H3 further extends to a broader topological context. The gating coefficients generated during the computation of each layer serve as part of the model's internal attention mechanism to guide feature aggregation, but are not output independently.

[0121] Furthermore, this multi-scale feature fusion layer integrates node features from different levels of abstraction to fully utilize both local details and global structural information. Although the H3 output of the third layer already contains information about multi-hop neighbors, the specific details of the earlier layers may be diluted after multiple nonlinear transformations. Explicitly concatenating and fusing the features from each layer allows the final features to simultaneously possess high-level semantic information and low-level detailed information, thereby improving the accuracy of hybrid recognition.

[0122] The input to this layer is the feature matrices H1, H2, and H3 output from the previous three message-passing layers. Each layer's feature matrix has n rows equal to the total number of nodes and d_hidden columns.

[0123] Internally, H1, H2, and H3 are first concatenated column-wise to form a fusion feature matrix H_fusion, with dimensions n×(3×d_hidden). The concatenated features contain multi-scale information about nodes, from direct neighbors and two-hop neighbors to multi-hop neighbors, including both local details and global structure. However, direct concatenation may result in excessively high feature dimensions and redundancy. Therefore, a fully connected layer is added to reduce the dimensionality and further abstract the fusion features, resulting in the final multi-scale feature matrix H_multi, with dimensions n×d_multi, where d_multi is the dimension of the final fusion feature (e.g., 256 dimensions in this embodiment). Each row in H_multi corresponds to a node and is the comprehensive feature representation used for the final confidence judgment of that node.

[0124] Furthermore, this confidence output layer is used to generate the final confidence score for each node as a candidate hybrid connection point, i.e., the confidence score for each node as a candidate hybrid connection point. Considering that hybrid connections in drainage pipe networks are mainly divided into two categories: one is sewage or combined sewer access to the stormwater system (leading to direct discharge of pollutants into rivers), and the other is stormwater access to the sewage system (leading to increased load on sewage treatment plants), this layer is designed with multiple parallel output heads, each specifically responsible for identifying one type of hybrid connection, thereby making the final confidence score judgment more accurate.

[0125] The input to this layer is H_multi (multi-scale fusion feature matrix) output from the multi-scale feature fusion layer. Internally, K parallel output heads are set up; in this embodiment, K can be 2, corresponding to the two hybrid connection types. Each output head consists of a fully connected layer and a sigmoid activation function, mapping the multi-scale fusion feature of each node to a confidence score between 0 and 1. The first output head outputs the confidence score of the node belonging to the "sewage / combined sewer access to stormwater" type; the second output head outputs the confidence score of the node belonging to the "stormwater access to sewage" type. The parameters of the two output heads are independent, and each learns the feature pattern corresponding to the hybrid connection type.

[0126] For each node i, the larger of the two output header values ​​is taken as the final confidence score for that node as a candidate hybrid point. This design allows the model to model different types of hybrids separately, improving the accuracy of the confidence score output.

[0127] Finally, this layer outputs a confidence vector P=[p1, p2, ..., pn] of length n, where each element pi represents the confidence level of the corresponding node as a candidate hybrid point. This vector is the final output of the model and can be directly used for subsequent processing: nodes with confidence levels greater than a preset threshold (e.g., 0.6) are selected to form a list of candidate hybrid points, and then proceed to step S140 for secondary verification.

[0128] Furthermore, during the training phase, the model utilizes labeled samples from historical mixed-signal data for supervised learning. Since the model has two output heads, the loss function calculates the cross-entropy loss for each output head's prediction and the true label separately, then sums them. Through backpropagation, all model parameters are jointly optimized, including the gating units of the dual-channel input layer, the message passing layer, and the weights of the fusion and output layers. After multiple rounds of iterative training, the model converges to a stable state capable of accurately outputting the confidence score of each node. During the inference phase, the model calculates the confidence score for each node according to the above process, serving as the final output.

[0129] Therefore, this model processes the features of two types of nodes—inspection wells and storm drains—separately through a dual-channel input layer. This allows the model to learn independent feature transformation patterns for different node types, avoiding information loss caused by conflating two different types of nodes and improving the targeting and accuracy of feature extraction. The flow-aware gating message passing layer introduces gating units based on pipe segment type and flow direction, enabling the model to automatically learn and focus on abnormal connections. When the connection conforms to normal drainage patterns, the gating coefficient is small, restricting information transmission; when there is suspicion of cross-connection, the gating coefficient is large, amplifying abnormal information, thus accurately capturing the topological features of cross-connection points. The multi-scale feature fusion layer stitches and fuses node features at different levels of abstraction, ensuring that the final features simultaneously possess local details and global structural information, effectively improving the robustness of cross-connection identification. The confidence output layer employs a dual-output head structure, corresponding to two different types of mixed connections. This allows the model to model different types of mixed connections separately, outputting a confidence score for each node as a candidate mixed connection point. The final confidence score is determined by taking the larger value, which not only improves the accuracy of the confidence output but also implicitly includes information about the mixed connection type, providing a more precise decision-making basis for subsequent rectification. After training with historical data, the model can automatically output the confidence score of each node in an end-to-end manner without manual intervention, efficiently and accurately completing the task of identifying mixed connection points in large-scale drainage pipe networks.

[0130] It should be noted that, in addition to the aforementioned physical perception dual-channel heterogeneous network model, those skilled in the art can replace this machine learning model with other models suitable for drainage network analysis, as long as they can output hybrid confidence scores based on the topological characteristics of the nodes. For example, it could be a dynamic neural network model based on NARX; or, for another example, a fusion model based on SWMM and BP neural networks.

[0131] It should be noted that, between steps S130 and S140, the identification method may further include: acquiring monitoring data of candidate mixing points; the monitoring data includes at least one of flow data and water quality data;

[0132] For candidate mixing points belonging to the stormwater pipe network, obtain their monitoring data during non-rainfall periods. If it is determined that there is a stable flow and the water quality pollution index exceeds the standard, then the candidate mixing point is marked as a high-confidence mixing point. Among them, stable flow and water quality pollution index exceeding the standard means that the variance of the flow value in the monitoring data within a preset time period is less than a preset threshold and the concentration of the water quality index is greater than a preset threshold.

[0133] For candidate cross-connection points belonging to the sewage pipe network, monitoring data during rainfall periods are obtained. If an increase in flow or a decrease in water quality pollution indicators is confirmed, the candidate cross-connection point is marked as a high-confidence cross-connection point. Here, an increase in flow means that the increase in flow value in the monitoring data compared to the average flow during the non-rainfall period before rainfall is greater than a preset flow increment threshold; a decrease in water quality pollution indicators means that the decrease in the concentration of water quality indicators compared to the average concentration during the non-rainfall period before rainfall is greater than a preset concentration decrease threshold.

[0134] In step S140, preset design-allowed cross nodes are removed from all candidate cross nodes, even if they are marked as high-confidence cross nodes; the high-confidence cross nodes among the remaining candidate cross nodes are taken as the final cross nodes.

[0135] For example, firstly, the system acquires monitoring data from each candidate mixing point. To avoid interference from rainfall runoff on the water quality of the stormwater pipe network and to accurately capture the process of rainwater mixing into the sewage pipe network, the system needs to correlate with meteorological data and set different monitoring periods for different pipe network types: for candidate mixing points belonging to the stormwater pipe network, monitoring data is collected during non-rainfall periods (e.g., a specific time window after rainfall ends, such as 24 hours after rain); for candidate mixing points belonging to the sewage pipe network, monitoring data is collected during rainfall periods (e.g., the duration after rainfall begins). This monitoring data may include at least one of flow data (e.g., instantaneous flow rate, flow velocity) and water quality data (e.g., pollution indicators such as COD, ammonia nitrogen, etc.).

[0136] Secondly, the system employs different auxiliary verification rules based on the type of pipeline network to which the candidate cross-connection points belong:

[0137] For candidate mixing points (including manholes and storm drains) belonging to the stormwater network: if a stable flow is detected at this point during non-rainfall periods, and the water quality pollution index exceeds the standard, it indicates that sewage is continuously mixing into this stormwater node. "Stable flow" means that the variance of the flow value within a preset time period is less than a preset variance threshold, indicating continuous water flow rather than a post-rainfall drainage process; "Water quality pollution index exceeding the standard" means that the concentration of the water quality index is greater than a preset concentration threshold, reaching or approaching the level of domestic sewage. In this case, the system marks the candidate mixing point as a "high-confidence mixing point."

[0138] For candidate mixing points belonging to the sewage pipe network: if the flow rate at this point during a rainfall period shows a significant increase compared to the average flow rate during the non-rainfall period before the rainfall, or if the concentration of water quality pollutants shows a significant decrease compared to the average concentration during the non-rainfall period before the rainfall, it indicates that rainwater has mixed into the sewage node. A flag is triggered if either of the above two conditions is met: the increase in flow rate is greater than a preset flow rate increment threshold, or the decrease in concentration is greater than a preset concentration decrease threshold. In this case, the system also marks the candidate mixing point as a "high-confidence mixing point".

[0139] Finally, when proceeding to step S140 to eliminate design-allowed cross-connection nodes, the system first removes all preset design-allowed cross-connection nodes (such as intercepting wells and regulating reservoir inlets / outlets) from the candidate cross-connection point list, regardless of whether these nodes are marked as "high-confidence cross-connection points". After completing the above elimination operation, for the remaining candidate cross-connection points marked as "high-confidence cross-connection points", the system prioritizes their retention and directly includes them in the final cross-connection point list; other unmarked candidate points are further filtered according to normal rules.

[0140] Step S140: Remove the preset design-allowed cross nodes from all candidate mixed connection points to obtain and output the final mixed connection points.

[0141] It should be understood that the preset design allows the specific points of the intersection nodes to be set according to actual needs, and the embodiments of this application are not limited thereto.

[0142] Optionally, in the drainage network asset information system, nodes with certain special functions are pre-marked. Although these nodes connect pipe segments of different network types in the topology, their existence is permitted by design and does not constitute a mixed connection issue. In this embodiment, the pre-defined design allows cross nodes to include at least the following two types:

[0143] Interception well node: Used to intercept initial rainwater or combined sewer overflow pollution. Its structural design allows sewage pipes and rainwater pipes (or combined pipes) to meet inside the well.

[0144] Storage tank inlet and outlet nodes: used for the inlet and outlet of rainwater storage facilities. Their design allows rainwater pipes to be connected to the storage tank or to the sewage system.

[0145] The aforementioned nodes have pre-set marker fields in the drainage network asset data (such as "Node Type = Interception Well" or "Design Allows Crossing Mark = Yes"), which can be queried and identified by the system.

[0146] To facilitate understanding of step S140, a specific embodiment will be described below.

[0147] Specifically, the result verification module can perform a secondary verification on the candidate mixed connection point list initially identified by the mixed connection rule matching module. By querying the "design allowed cross nodes" (such as intercepting wells and regulating tank inlet and outlet nodes) pre-marked in the drainage pipe network asset information system, nodes that meet the conditions are removed from the candidate list, and the mixed connection point list is finally confirmed. The module also records the node identifier, mixed connection type (such as "sewage pipe connected to rainwater pipe", "combined pipe connected to rainwater pipe", etc.) and mixed pipe diameter of each mixed connection point.

[0148] Therefore, through the above-mentioned auxiliary verification mechanism, this invention combines topology rule judgment with real-time monitoring data verification, effectively making up for the misjudgment that may be caused by incorrect asset information (such as incorrect pipeline type marking) due to the reliance on topology data alone, and further improving the robustness and on-site adaptability of mixed connection point identification.

[0149] In addition, the pipe segment data also includes the pipe diameter of the pipe segment; after step S140, it also includes: for the final mixing point, determining the pipe diameter of the mixing pipe and its mixing type;

[0150] For nodes that serve as the final mixed connection point and whose node type is a manhole, the pipe segments associated with that node are grouped according to the pipe network type. For each group of pipe segments, if all pipe segments in the group flow into that node or all pipe segments flow out of that node, then the pipe segments in that group are considered mixed pipes, and the diameter of the mixed pipe is taken from the diameter of the pipe segments in that group. The pipe segments associated with that node that have both inflow and outflow pipe network types are considered normal pipe segments. The mixed connection type is that the pipe network type of each group of mixed pipes is connected to the pipe network type of the normal pipe segment.

[0151] For example, for a manhole, if a certain type of pipeline connected to it has only an inlet pipe or only an outlet pipe, then that type of pipeline is a mixed-connection pipeline, and its diameter is the mixed-connection pipeline diameter. Pipelines with both inlets and outlets are normal pipelines. Mixed connection type: The type of pipeline in a mixed-connection pipeline is connected to the type of pipeline in a normal pipeline.

[0152] To facilitate understanding of the process of determining the mixed pipe, pipe diameter, and mixed connection type related to inspection wells, specific embodiments are described below.

[0153] For example, such as Figure 2 As shown in the diagram, the manhole is connected to both rainwater and sewage pipes, making it a mixed-connection manhole. The rainwater pipes connected to this manhole have both inlet and outlet connections; these are normal rainwater pipes. The sewage pipes connected to this manhole only have inlet connections and no outlet connections; these are mixed-connection pipes, and their diameter is the same as the mixed-connection pipe diameter. The mixed-connection type is: sewage pipe (the type of mixed-connection pipe) connected to rainwater pipe (the type of normal pipe).

[0154] For example, such as Figure 3 As shown in the diagram, the manhole is connected to both rainwater and sewage pipes, making it a mixed-connection manhole. The sewage pipe connected to this manhole has both inlet and outlet pipes, which is a normal pipeline. The rainwater pipe connected to this manhole only has inlet pipes and no outlet pipes, making it a mixed-connection pipe, and its diameter is the same as that of a mixed-connection pipe.

[0155] Mixed connection type: Rainwater pipe (the type of pipe in the mixed connection) is connected to sewage pipe (the type of pipe in the normal connection).

[0156] Verification Notes: If the inspection well in the diagram is a diversion well or the inlet of a regulating reservoir, this is permitted by design and is considered a permitted node, not a mixed connection point.

[0157] For example, such as Figure 4 As shown in the diagram, the manhole is connected to both rainwater pipes and combined drain pipes, making it a mixed-connection manhole. The rainwater pipes connected to this manhole have both inlet and outlet pipes, which are normal pipe configurations. The combined drain pipe connected to this manhole only has inlet pipes and no outlet pipes; therefore, it is a mixed-connection pipe with the same diameter as the mixed-connection pipe.

[0158] Mixed connection type: Combined sewer (the type of pipe in the mixed connection) is connected to rainwater pipe (the type of pipe in the normal connection).

[0159] Optionally, the pipe segment data also includes the pipe diameter of the pipe segment; after step S140, the method further includes: for the final mixing point, determining the pipe diameter of the mixing pipe and its mixing type; for the node that is the final mixing point and whose node type is a rainwater inlet, the pipe segments of all inflow pipes of the rainwater inlet that belong to sewage pipes or combined sewers are designated as mixing pipes, the pipe diameter of the mixing pipe is the pipe diameter of the pipe segment, and the mixing type is the pipe network type of the mixing pipe connected to the rainwater pipe; and, the pipe segments of all outflow pipes of the rainwater inlet that belong to sewage pipes are designated as mixing pipes, the pipe diameter of the mixing pipe is the pipe diameter of the pipe segment, and the mixing type is the pipe network type of the rainwater pipe connected to the mixing pipe.

[0160] For example, in a mixed-connection rainwater inlet, all non-rainwater pipes flowing into the rainwater inlet are mixed-connection pipes, and all sewage pipes flowing out of the rainwater inlet are mixed-connection pipes. If the mixed-connection pipe flows into the rainwater inlet, the mixed-connection type is: the pipe type of the mixed-connection pipe is connected to the rainwater pipe; if the mixed-connection pipe flows out of the rainwater inlet, the mixed-connection type is: the rainwater pipe is connected to the pipe type of the mixed-connection pipe.

[0161] To facilitate understanding of the process of determining the mixed pipe, pipe diameter, and mixed connection type related to rainwater inlets, specific embodiments are described below.

[0162] For example, such as Figure 5 As shown in the figure, the pipe connected to the rainwater inlet is a non-rainwater pipe (sewage pipe), and the rainwater inlet is a mixed rainwater inlet;

[0163] Mixed connection type: Sewage pipe (mixed pipe type) is connected to rainwater pipe.

[0164] For example, such as Figure 6 As shown in the diagram, the pipe connected to the rainwater inlet is a non-rainwater pipe (sewage pipe), and the rainwater inlet is a mixed-connection rainwater inlet.

[0165] Mixed connection type: Rainwater pipes (mixed pipe types) are connected to sewage pipes.

[0166] Verification Notes: If the inspection well in the diagram is a diversion well or the inlet of a regulating reservoir, this is permitted by design and is considered a permitted node, not a mixed connection point.

[0167] For example, such as Figure 7 As shown in the figure, the pipe connected to the rainwater inlet is a non-rainwater pipe (combined pipe), and the rainwater inlet is a mixed-connection rainwater inlet.

[0168] Mixed connection type: Combined pipe (mixed pipe type) connected to rainwater pipe.

[0169] In addition, the system's application layer is connected to the core verification layer, and is responsible for visualizing and applying the finally confirmed crossover point information. This layer mainly provides the following three core functions:

[0170] Map visualization: Based on the location coordinates of the nodes, the finally confirmed mixed connection points (including inspection wells and rainwater inlets) are marked and highlighted on the electronic map of the drainage pipe network asset information system. Users can click on the mark to view the detailed information of the mixed connection point (such as node identification, mixed connection type, mixed pipe diameter, etc.).

[0171] Report generation: Automatically generates cross-connection verification reports, and statistically analyzes information such as the number of cross-connection points, their distribution areas, and the distribution of cross-connection types, providing data support for operation and maintenance decisions;

[0172] Data Interface: Provides standardized data interfaces that can be connected to existing drainage network asset information systems, operation and maintenance management systems, etc., to achieve the reuse and collaborative management of mixed data and break down information silos.

[0173] In other words, the identification method also includes: marking and displaying the final cross-connection point on an electronic map based on the location coordinates, and at least displaying the node identifier of the final cross-connection point; and, when available, displaying the cross-connection type and / or the pipe segment identifier corresponding to the cross-connection pipe of the final cross-connection point, which can be displayed according to actual needs.

[0174] In summary, by utilizing the above technical solutions, this application has the following significant advantages:

[0175] (1) Significantly improves the efficiency of investigation and reduces the workload: This application realizes the mixed connection verification by automatically parsing the pipeline network topology relationship. There is no need for manual on-site verification point by point. The mixed connection identification of the entire pipeline network can be completed in a short time (such as the verification of a pipeline network with hundreds of thousands of nodes can be completed in tens of seconds), which greatly reduces the labor cost and time cost and can realize the investigation of the entire pipeline network.

[0176] (2) Improve the accuracy of mixed connection identification and reduce missed and false judgments: This application makes judgments based on objective data of pipeline topology connection relationship, avoiding reliance on manual experience; at the same time, the cross nodes allowed by the design are eliminated through "secondary verification", which further improves the identification accuracy and effectively reduces the phenomenon of missed judgment (such as hidden mixed connection can be identified by topology chain) and false judgment (such as excluding design cross nodes to avoid false judgment).

[0177] (3) The mixed connection point is displayed intuitively, supporting precise management: This application links the mixed connection point with the electronic map, which can intuitively display the spatial location of the mixed connection point and the surrounding pipeline topology. Staff can quickly locate the mixed connection point through the map, providing precise spatial support for the formulation of rectification plan (such as determining the rectification construction route and required materials) and the tracking of rectification effect (such as re-verification and comparison after rectification), thereby improving the efficiency of operation and maintenance management.

[0178] (4) Realize dynamic and normalized monitoring with strong timeliness: This application sets up a dynamic update trigger mechanism, which can re-verify in real time based on changes in pipeline asset information (such as new pipeline sections and renovation nodes), promptly discover new mixed connection problems, solve the defects of traditional static investigation that cannot track pipeline changes in a timely manner, and realize normalized and dynamic monitoring of mixed connection problems.

[0179] (5) Data is reusable and collaboration is enhanced: This application can be seamlessly connected with existing drainage network asset information systems, drainage network operation and maintenance management systems, etc. The mixed data generated by the verification can be directly reused in other systems, providing data support for network optimization and water environment governance decisions, breaking the "data silo" problem of traditional investigation, and enhancing the collaboration of municipal information management.

[0180] (6) No additional dedicated hardware required, low implementation cost: This application relies on the basic data of the existing drainage network asset information system. There is no need to purchase additional portable testing equipment. Only a new software verification module is needed. The implementation cost is low and it is easy to promote and apply in the existing municipal information system.

[0181] It should be understood that the above-mentioned method for identifying mixed connection points of drainage pipe networks is merely exemplary. Those skilled in the art can make various modifications based on the above system, and the modified solutions also fall within the protection scope of this application.

[0182] This application also provides an electronic device, which includes a processor and a memory. The memory stores a computer program, and when the processor executes the program, it implements the above-described method for identifying mixed connection points in drainage pipe networks.

[0183] This application embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the method for identifying mixed connection points in drainage pipe networks as described above.

[0184] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0185] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions.

[0186] It should be noted that any reference numerals placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In claims that enumerate several means, several of these means may be embodied by the same hardware. The use of the terms first, second, third, etc., is merely for convenience of expression and does not indicate any order. These terms can be understood as part of the component names.

[0187] Furthermore, it should be noted that in the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0188] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the claims should be interpreted to include both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0189] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

Claims

1. A method for identifying mixed connection points in a drainage pipe network, characterized in that, include: Step S1: Obtain drainage network asset data, which describes the drainage network topology consisting of nodes and pipe segments. The pipe segment data includes at least: a starting node identifier and an ending node identifier to indicate the drainage system network type to which the pipe segment belongs, and to determine the connection relationship of the pipe segment within the drainage network topology. The node data includes at least: a set of associated pipe segment markers to determine the associated pipe segments of the node within the drainage network topology. Step S2: Based on the starting node identifier, the ending node identifier, and the associated pipe segment marker set, parse the connection relationship between each pipe segment and each node to construct a topological connection chain composed of alternating nodes and pipe segments; Step S3: Traverse the topology connection chain and, based on the network type of each pipe segment, identify the nodes where components of different network types intersect, as candidate hybrid connection points; Step S4: Remove the preset design-allowed cross nodes from all the candidate hybrid points to obtain and output the final hybrid points.

2. The identification method according to claim 1, characterized in that, The node data also includes: node types for distinguishing between inspection wells and rainwater inlets; the specific rules for identifying the candidate mixed connection points in step S3 include: For a node whose node type is the manhole, if its associated pipe segment contains both rainwater pipe and sewage pipe, or if its associated pipe segment contains a combined pipe and contains both the rainwater pipe and the sewage pipe, then the manhole is considered as the candidate mixed connection point. For a node whose node type is the rainwater inlet, if its associated pipe segment contains a sewage pipe or a combined pipe flowing into the rainwater inlet, or contains a sewage pipe flowing out of the rainwater inlet, then the rainwater inlet is regarded as the candidate mixed connection point.

3. The identification method according to claim 2, characterized in that, The data for the pipe segment also includes the pipe diameter of the pipe segment; after step S4, the method further includes: determining the pipe diameter and the type of mixing pipe for the final mixing point; For a node that serves as the final cross-connection point and is of the inspection well type, the pipe segments associated with that node are grouped according to the pipe network type. For each group of pipe segments, if all pipe segments in the group flow into that node or all pipe segments flow out of that node, then that group of pipe segments is considered a cross-connection pipe, and the diameter of the cross-connection pipe is taken from the diameter of the pipe segments in that group. The pipe segments associated with that node that have both inflow and outflow pipe network types are considered normal pipe segments. The cross-connection type is that the pipe network type of each group of cross-connection pipes is connected to the pipe network type of the normal pipe segment.

4. The identification method according to claim 2, characterized in that, The data for the pipe segment also includes the pipe diameter of the pipe segment; after step S4, the method further includes: determining the pipe diameter and the type of mixing pipe for the final mixing point; For a node that serves as the final connection point and is of the type of storm drain, the pipe segments of all inflow pipes of the storm drain that belong to the sewage pipe or the combined sewer are designated as the combined pipe, the diameter of the combined pipe is the diameter of the pipe segment, and the connection type is the pipe network type of the combined pipe connected to the storm drain; and, the pipe segments of all outflow pipes of the storm drain that belong to the sewage pipe are designated as the combined pipe, the diameter of the combined pipe is the diameter of the pipe segment, and the connection type is the pipe network type of the storm drain connected to the combined pipe.

5. The identification method according to claim 3 or 4, characterized in that, The node data also includes a node identifier for uniquely identifying the node and the node's location coordinates; the identification method further includes: The final cross-connection point is marked and displayed on the electronic map according to the location coordinates, and at least the node identifier of the final cross-connection point is displayed; and, when available, the cross-connection type of the final cross-connection point and / or the pipe segment identifier corresponding to the cross-connection pipe is displayed.

6. The identification method according to claim 1, characterized in that, The identification method further includes: When updated drainage network asset data is obtained, steps S1 to S4 are re-executed based on the updated drainage network asset data to identify the updated final connection point.

7. The identification method according to claim 1, characterized in that, After step S3 and before step S4, the identification method further includes: Acquire monitoring data of the candidate mixing points; the monitoring data includes at least one of flow data and water quality data; For candidate mixing points belonging to the stormwater pipe network, obtain their monitoring data during non-rainfall periods. If it is determined that there is a stable flow and the water quality pollution index exceeds the standard, then mark the candidate mixing point as a high-confidence mixing point. For candidate cross-connection points belonging to the sewage pipe network, obtain their monitoring data during rainfall periods. If it is determined that there is an increase in flow or a decrease in water quality pollution indicators, then mark the candidate cross-connection point as a high-confidence cross-connection point. In step S4, preset design-allowed cross nodes are removed from all the candidate cross nodes, even if they are marked as high-confidence cross nodes; the high-confidence cross nodes among the remaining candidate cross nodes are taken as the final cross nodes.

8. The identification method according to claim 1, characterized in that, Step S3 includes: Extract the topological features of each node from the topological connection chain; the topological features include at least one or more of the following: the number of pipe segments associated with the node, the distribution of pipe network types of each associated pipe segment, the flow direction relationship of each associated pipe segment, and the position level of the node in the topological connection chain; Based on the topological characteristics of each node, the confidence level of each node as a candidate hybrid connection point is determined; Nodes with a confidence level greater than a preset threshold are selected as candidate hybrid connection points.

9. An electronic device, characterized in that, include: processor; The processor also includes a memory for storing executable instructions of the processor, wherein the processor is configured to implement the drainage network mixed connection point identification method according to any one of claims 1 to 8 by executing the executable instructions.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the drainage pipe network mixed connection point identification method according to any one of claims 1 to 8.

Images