Drainage pipe network space topology dynamic reconstruction method based on RTK and field survey fusion

Abstract

The invention provides a drainage pipe network spatial topology dynamic reconstruction method based on RTK and on-site survey fusion. The method comprises the following steps: firstly, dividing troubleshooting blocks based on a pump station service range and a water catchment unit; establishing communication connection between RTK equipment and a mobile acquisition terminal, acquiring three-dimensional coordinate data of key nodes of the pipe network, synchronously acquiring field images, and generating a pipe network space sketch in real time; inputting attribute information on the basis, and constructing an initial vector pipe network topology model; the core is to obtain on-site on-site survey feature data, construct a multi-source evidence confidence correction model, perform logic conflict detection and dynamic correction on an initial topology model by using the model, and intelligently identify and repair anomalies such as reverse slopes, disconnection and dark connection. And finally, mapping the corrected model into a standardized GIS pipe network database and outputting the standardized GIS pipe network database. By fusing high-precision physical actual measurement and on-site microscopic trace evidence, the problem of topology distortion caused by old pipe network drawing missing and settlement is effectively solved, and the accuracy of pipe network reconstruction is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of drainage pipe networks, and in particular to a method for dynamic reconstruction of the spatial topology of drainage pipe networks based on the fusion of RTK and field survey. Background Technology

[0002] Urban drainage networks are the "lifeline" ensuring the normal operation of cities and the safety of the ecological environment. However, in many older urban areas, the construction of drainage networks spans a wide period, often resulting in severely distorted or even missing drawings and data. In these areas, traditional network management and analysis heavily rely on historical CAD drawings or archival data. However, due to early construction errors, failure to update drawings in a timely manner during later renovations and expansions, and geological subsidence, the network models recorded in the archives often deviate significantly from the actual operating conditions. Especially in areas with high groundwater levels (such as phreatic layers with a stable water level of around 3 meters), soil erosion and subsidence can easily cause physical displacement of pipelines, creating "pseudo-reverse slopes" or misconnections due to subsidence, leading to a severe disconnect between the design model and the actual site conditions.

[0003] In the existing technology, various technical means have emerged in the industry to acquire and organize drainage pipe network information. For example, the existing technology CN120125762B discloses a method for generating urban drainage pipe networks based on remote sensing images. It uses high-resolution satellite remote sensing images combined with visual models to identify the location of manhole covers and generate topology. Although this method has advantages in acquiring macro layout, it is difficult to explore the physical state inside the underground pipes (such as siltation and water level marks), and it is not sensitive enough to centimeter-level elevation changes and hidden underground misconnections.

[0004] The prior art CN115795122B discloses a method for sorting out the topological relationships of urban drainage pipe networks. It uses a directed graph to search for loop structures and logical errors by constructing a directed graph and calculating the node degree, in-degree, and out-degree, and performs DAG testing. This method mainly focuses on checking the rationality of mathematical logic. However, in the absence of on-site physical evidence, it is easy to misjudge the real physical reverse slope caused by settlement as a logical error, or fail to distinguish between "logical reverse slope" and "physical settlement".

[0005] In addition, the existing technology CN106382471B proposes a diagnostic and assessment method for urban drainage pipe networks that considers key nodes. It constructs a hydraulic model by adding water level and flow monitors and combining them with CCTV detection. Although it can accurately assess the operating status, it relies on the deployment of high-cost monitoring equipment and is difficult to support rapid surveys of the entire area and reconstruction of the basic topology from scratch.

[0006] Existing technology CN117745473A discloses a GIS-based method for assessing the health status of urban drainage pipe networks. This method focuses on calculating the density of misconnected and reverse-slope pipes to evaluate health. Essentially, it's a post-processing assessment of known pipe network data, failing to address the challenge of distinguishing genuine from fake data using on-site traces at the source of missing data. In summary, existing technologies primarily focus on logical correction based on existing data or macroscopic image recognition, lacking a dynamic reconstruction method that deeply integrates high-precision geographic coordinate measurements with on-site microscopic physical traces (such as watermarks and sediments) to address topological distortion caused by high groundwater levels in real-time during the investigation phase. Summary of the Invention

[0007] The main objective of this invention is to provide a method for dynamic reconstruction of the spatial topology of drainage pipe networks based on RTK and field survey. This application solves the technical problem that in old urban areas and areas with high groundwater levels, the spatial location deviation of drainage pipe networks is large due to missing map data and geological subsidence, and the physical operating status (such as subsidence slope) is difficult to distinguish from the logical topology, thus making it impossible to construct an accurate GIS base.

[0008] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for dynamic reconstruction of the spatial topology of drainage pipe networks based on the fusion of RTK and field survey, characterized in that: the method includes: S1: The investigation area is divided based on the service range of the pumping station and the hierarchical level of the water catchment unit; S2: Establish communication connection between RTK equipment and mobile acquisition terminal, collect three-dimensional coordinate data of key nodes in pipeline network, synchronously collect on-site image data, and generate pipeline network spatial sketch in real time. S3: Input pipe diameter, material and flow direction attribute information into the mobile acquisition terminal, associate the three-dimensional coordinate data with the attribute information, and construct an initial vector pipe network topology model; S4: Obtain on-site survey feature data, construct a multi-source evidence confidence correction model, and use this model to perform logical conflict detection and dynamic correction on the initial vector pipeline topology model; S5: Map the corrected topology model to a standardized GIS pipeline database and output it.

[0009] In the preferred embodiment, step S2, which involves establishing a communication connection and data processing between the RTK device and the mobile acquisition terminal, includes: RTK devices transmit NMEA format positioning data streams to mobile acquisition terminals via Bluetooth or serial communication protocols; The solution module in the mobile acquisition terminal parses the NMEA data stream, extracts longitude, latitude and ellipsoidal elevation information, and uses built-in local coordinate transformation parameters to convert the WGS84 coordinate system to the local engineering coordinate system in real time. The mobile acquisition terminal maps the converted coordinate points onto the preloaded electronic map base map. When the acquisition point is a manhole or pipe opening, it triggers an image acquisition command, calls the terminal camera to take a picture of the scene, and writes the high-precision location data acquired by the current RTK into the Exif information of the picture.

[0010] In the preferred scheme, step S4, the construction steps of the multi-source evidence confidence correction model include: Define the physical slope calculated from the measured pipe bottom elevation using RTK as the first evidence vector; Define the natural slope trend of the surrounding terrain as the second evidence vector; The flow direction indicated by the highest liquid level trace line on the well wall of the inspection well is defined as the third evidence vector; Assign weight coefficients to the first evidence vector, the second evidence vector, and the third evidence vector respectively. , , The overall flow confidence index is obtained through weighted calculation.

[0011] In the preferred scheme, step S4, which involves logical conflict detection and dynamic correction of the initial vector pipeline topology model, includes a correction algorithm for reverse slope anomalies. Calculate the RTK physical slope between two adjacent critical nodes. If the physical slope is negative, it is judged as a suspected reverse slope. Retrieve on-site survey feature data of the two adjacent key nodes; If a third evidence vector exists and indicates that the flow direction is opposite to the physical slope direction, and the comprehensive flow direction confidence index exceeds a preset threshold, the RTK physical slope is determined to be a pseudo reverse slope caused by settlement. The system automatically locks the upstream node coordinates, forcibly modifies the downstream node's logical flow direction attribute to be consistent with the third evidence vector, and marks the pipe segment as a physical settlement reverse slope pipe segment in the database.

[0012] In the preferred embodiment, step S4, which involves logical conflict detection and dynamic correction of the initial vector pipeline topology model, includes correction algorithms for disconnections and hidden connections. Retrieve isolated nodes with a degree of 1 in the initial vector network topology model; Using the isolated node as the center, set a search radius R, and search for other pipeline nodes within the search radius R; Based on the distribution range of overflow sediments in the on-site reconnaissance data, if the overflow sediments show a banded distribution pointing towards a certain downstream node within the search radius R, it is determined that there is a hidden connection relationship. A virtual connection edge is automatically generated between the isolated node and the downstream node, and the attribute of the virtual connection edge is marked as speculative hidden connection.

[0013] In the preferred scheme, in step S3, the initial vector network topology model is constructed and stored using a hierarchical graph data structure: The first layer is the physical node layer, which stores the geometric point elements of inspection wells, discharge outlets, and pump station inlets and outlets; The second layer is the pipeline edge layer, which stores the pipeline segment elements that connect physical nodes. Each pipeline edge element contains the start node ID, the end node ID, and the corresponding pipe diameter weight. The third layer is the attribute association layer, which uses a unique coded ID to attach on-site image data and material text data to the physical node layer or pipeline edge layer; The mobile data acquisition terminal uses an embedded SQLite database to store the hierarchical graph data locally and encapsulates it in GeoJSON format.

[0014] In the preferred embodiment, this method relies on a system that includes a mobile app and a cloud server for deployment. Step S5, which outputs a standardized GIS pipeline database, involves the following data synchronization and conversion processes: The mobile data acquisition terminal uploads the GeoJSON incremental data packet corrected in step S4 to the cloud server via the HTTPS encryption protocol; The cloud server uses Nginx as a reverse proxy server to receive data requests and forward them to the application middleware; The application middleware calls the topology check service to verify the data connectivity again. After successful verification, the GeoJSON data is parsed and loaded into the PostGIS spatial extension module of the PostgreSQL database using an ETL tool; The PostGIS spatial extension module executes a stored procedure to convert the data into a Shapefile or GDB format file that conforms to the urban underground pipeline data standard, and completes the database output.

[0015] In the preferred embodiment, for areas with high groundwater levels, the correction algorithm in step S4 further includes a liquid level correction sub-step: Collect real-time groundwater level data at key nodes; When the pipe bottom elevation measured by RTK is lower than the real-time groundwater level data, calculate the influence factor of buoyancy on pipe displacement; If the impact factor exceeds the safety threshold, the weight coefficient of the first evidence vector should be reduced in the multi-source evidence confidence correction model. And increase the weight coefficient of the third evidence vector. .

[0016] In the preferred embodiment, after dividing the investigation blocks in step S1, the method further includes a preprocessing step for the historical drawing data within the blocks: Import historical CAD pipeline drawings into the GIS platform for georeferencing; Extract the node coordinates from historical drawings as a reference set; In step S4, if the coordinates collected by RTK deviate from the corresponding points in the reference set by more than a set threshold, but the topological connection relationship is consistent, the measured coordinates of RTK are retained in the database, and the coordinates of the historical drawings are recorded as the original drawing deviation reference value.

[0017] In the preferred embodiment, a Docker containerized knowledge base service is deployed on the cloud server to store the on-site image data collected in step S2; When outputting a standardized GIS pipeline database, the system automatically generates a URL link pointing to the knowledge base service and writes the URL link into the attribute table of the GIS data, realizing a cloud-based hyperlink index of pipeline spatial data and on-site survey photos.

[0018] This invention provides a method for dynamic reconstruction of the spatial topology of drainage pipe networks based on the fusion of RTK and field survey. This invention adopts real-time linkage between RTK equipment and mobile acquisition terminal, which can realize high-precision acquisition of centimeter-level three-dimensional coordinates of key nodes in the pipe network. By embedding location data through photo Exif information, the authenticity and traceability of source data are ensured, effectively solving the positioning problem caused by missing drawings.

[0019] This invention innovatively constructs a multi-source evidence confidence correction model that includes physical slope, topographic trend, and liquid level traces. In particular, it introduces a buoyancy influence factor correction mechanism for areas with high groundwater levels. By weighting different evidence vectors, it can intelligently distinguish between real reverse slopes caused by construction design and "pseudo-reverse slopes" caused by geological subsidence. It can also automatically infer the hidden connection relationship based on the distribution of overflow sediments in the inspection well, which greatly improves the logical accuracy and physical authenticity of the topology model.

[0020] Furthermore, this invention achieves seamless integration from on-site data collection to standardized GIS database output through automated data synchronization and conversion processes using cloud servers, Nginx reverse proxy, and PostGIS spatial databases, significantly improving the operational efficiency of the general survey and digital reconstruction of drainage pipe networks in old areas. Attached Figure Description

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the method for dynamic reconstruction of the spatial topology of drainage pipe network according to the present invention; Figure 2 This is a flowchart illustrating the principle of the dynamic reconstruction method for the spatial topology of drainage pipe networks according to the present invention. Figure 3This is a diagram of the data acquisition terminal interface of the drainage pipe network spatial topology dynamic reconstruction method of the present invention. Detailed Implementation

[0022] Example 1 like Figure 1-3 As shown, a method for dynamic reconstruction of the spatial topology of drainage pipe networks based on the fusion of RTK and field survey is characterized by the following: The method includes: S1: The investigation area is divided based on the service range of the pumping station and the hierarchical level of the water catchment unit; S2: Establish communication connection between RTK equipment and mobile acquisition terminal, collect three-dimensional coordinate data of key nodes in pipeline network, synchronously collect on-site image data, and generate pipeline network spatial sketch in real time. S3: Input pipe diameter, material and flow direction attribute information into the mobile acquisition terminal, associate the three-dimensional coordinate data with the attribute information, and construct an initial vector pipe network topology model; S4: Obtain on-site survey feature data, construct a multi-source evidence confidence correction model, and use this model to perform logical conflict detection and dynamic correction on the initial vector pipeline topology model; S5: Map the corrected topology model to a standardized GIS pipeline database and output it.

[0023] The technical solution first initiates the investigation by scientifically dividing the space. Specifically, it uses the service area of ​​the pumping station as the basic unit and combines it with the hierarchical relationship of the water catchment units to divide the specific investigation blocks. 1 This method of division ensures that subsequent data collection can be carried out in an orderly manner within a clearly defined and hydraulically significant geographical area, avoiding the waste of resources caused by blind surveys.

[0024] After the area was demarcated, technicians used high-precision measuring equipment to collect data. By establishing a communication connection between the real-time dynamic carrier phase differential (RTK) equipment and the mobile acquisition terminal, the system can acquire three-dimensional coordinates of key nodes in the drainage network with centimeter-level accuracy, such as manholes, pipe openings, and pump station inlets and outlets. This coordinate data includes longitude, latitude, and elevation information. Simultaneously, the mobile terminal collects on-site image data, using this location and image information to generate a spatial sketch of the network in real time within the terminal software, thus achieving instant data visualization.

[0025] Subsequently, workers input the pipeline network's attribute information onto a mobile data acquisition terminal. This information includes key parameters such as pipe diameter, pipe material, and water flow direction. The system then associates and binds the high-precision three-dimensional coordinate data acquired in this process with this attribute information, thereby constructing an initial vector pipeline network topology model containing geometric location and attribute data. This model provides a digital foundation for subsequent logical analysis.

[0026] To correct potential logical errors in the initial model caused by settlement or measurement mistakes, this method further acquires feature data from on-site reconnaissance and constructs a multi-source evidence confidence correction model. This model is used to detect and dynamically correct logical conflicts in the initial vector pipeline topology model, especially for complex situations such as reverse slopes, disconnections, and concealed connections.

[0027] Its core judgment logic is based on the comprehensive flow confidence index formula. Perform the calculation. In this formula, the variables... This represents the overall flow direction confidence index, used to quantitatively assess the reliability of the current topological flow direction. (Variable) This represents the physical slope vector calculated based on RTK-measured pipe bottom elevation, reflecting flow direction evidence at the geometric measurement level. Variables This represents the natural slope trend vector of the surrounding terrain, reflecting flow evidence at the geographical level. (Variable) The flow vector indicated by the highest liquid level trace line on the well wall reflects the flow evidence at the historical hydraulic operation level.

[0028] parameter , and These represent weighting coefficients for physical slope evidence, topographic trend evidence, and liquid level trace evidence, respectively. These weighting coefficients are allocated based on the credibility of the different evidence sources; for example, they are reduced in subsidence areas. The value and increase The numerical value. Through this mathematical model, the system can integrate physical measurement data with on-site evidence to intelligently identify and correct logical errors such as pseudo-inverted slopes.

[0029] Finally, the topology model, corrected by multi-source evidence, is mapped and converted into a standardized Geographic Information System (GIS) database for pipeline network management and then output. This step ensures that the final deliverables meet industry data standards and can be directly used as accurate base maps for subsequent monitoring point deployment, engineering design, and pipeline network operation and maintenance management.

[0030] In the preferred embodiment, step S2, which involves establishing a communication connection and data processing between the RTK device and the mobile acquisition terminal, includes: RTK devices transmit NMEA format positioning data streams to mobile acquisition terminals via Bluetooth or serial communication protocols; The solution module in the mobile acquisition terminal parses the NMEA data stream, extracts longitude, latitude and ellipsoidal elevation information, and uses built-in local coordinate transformation parameters to convert the WGS84 coordinate system to the local engineering coordinate system in real time. The mobile acquisition terminal maps the converted coordinate points onto the preloaded electronic map base map. When the acquisition point is a manhole or pipe opening, it triggers an image acquisition command, calls the terminal camera to take a picture of the scene, and writes the high-precision location data acquired by the current RTK into the Exif information of the picture.

[0031] In this step, establishing the communication connection and data processing between the RTK device and the mobile acquisition terminal first involves the transmission and parsing of the underlying data. The high-precision positioning device, i.e., the RTK device, uses Bluetooth wireless transmission technology or a serial communication interface protocol to establish a stable data transmission channel, continuously sending the real-time acquired positioning data stream to the mobile acquisition terminal. This data stream strictly follows the NMEA standard format defined by the National Marine Electronics Association (NMEA), which includes raw positioning information and satellite status data. The mobile acquisition terminal is equipped with a dedicated processing module responsible for real-time parsing of the received NMEA data stream, accurately extracting core three-dimensional spatial information such as longitude, latitude, and ellipsoidal elevation of the current location. Since the raw acquired data is usually based on the WGS84 world geodetic coordinate system, while engineering applications require specific local coordinate systems, the system uses built-in transformation parameters for real-time coordinate transformation.

[0032] Coordinate transformation typically involves the transformation of a spatial rectangular coordinate system. Its mathematical principle can be described using a seven-parameter transformation model, with the specific transformation formula expressed as follows: In this formula, the column vector This represents the three-dimensional coordinates obtained after transformation in the local engineering coordinate system. This is the final data required for subsequent mapping and engineering applications. (Column vector) Represents the three-dimensional coordinates in the raw WGS84 coordinate system parsed from the NMEA data stream. Column vector. These represent three translation parameters used to describe the positional deviation between the origins of two coordinate systems. (Variable) This represents the scale factor, used to correct for scale differences between two coordinate systems. (Matrix) This represents a rotation matrix consisting of three rotation angles, used to correct the deflection angles of the axes of the two coordinate systems. Through the calculation of this mathematical model, the system can accurately map globally accepted latitude and longitude coordinates to plane coordinates conforming to local engineering standards.

[0033] After coordinate transformation, the mobile acquisition terminal projects and displays the calculated precise coordinates in real time on a pre-loaded electronic map base, enabling visualized monitoring of the acquisition trajectory. When operators confirm that the currently acquired point belongs to a critical facility such as a manhole or pipe opening, the system immediately triggers an image acquisition command, automatically calling the terminal's camera hardware to capture on-site photos. Simultaneously, the system writes the centimeter-level high-precision location data acquired by the RTK device into the Exif metadata format of the photos. This operation ensures the permanent binding of on-site image evidence with high-precision spatial location information, achieving the integrated storage of geographic information data and multimedia data, and providing legally valid electronic evidence for subsequent pipeline data verification and traceability.

[0034] In the preferred scheme, step S4, the construction steps of the multi-source evidence confidence correction model include: Define the physical slope calculated from the measured pipe bottom elevation using RTK as the first evidence vector; Define the natural slope trend of the surrounding terrain as the second evidence vector; The flow direction indicated by the highest liquid level trace line on the well wall of the inspection well is defined as the third evidence vector; Assign weight coefficients to the first evidence vector, the second evidence vector, and the third evidence vector respectively. , , The overall flow confidence index is obtained through weighted calculation.

[0035] In step S4, the construction process of the multi-source evidence confidence correction model aims to verify the logical correctness of the pipeline network topology by fusing physical data from different dimensions. The system first defines a first evidence vector, which is the physical slope direction calculated based on the pipeline bottom elevation data obtained from real-time dynamic carrier phase differential technology (RTK equipment), representing the theoretical flow direction based on geometric measurement data. Simultaneously, the system defines a second evidence vector, which characterizes the natural tilt trend of the surrounding terrain, reflecting the natural flow direction of surface water or groundwater under gravity. Furthermore, the system defines a third evidence vector, determined based on the height distribution of the highest liquid level traces remaining on the manhole walls, serving as direct physical evidence reflecting the historical hydraulic state of the pipeline network. These three vectors describe the pipeline network flow direction from three dimensions: precise measurement, natural geographical environment, and actual operational traces.

[0036] To quantify the contribution of evidence from different sources to the final determination of the flow direction, the model introduces a weighted calculation mechanism. The system assigns specific weight coefficients to the first, second, and third evidence vectors, each represented by a Greek letter. , and The comprehensive flow confidence index is derived by synthesizing these three weighted vector components. This calculation process can be expressed mathematically as follows: .

[0037] In this formula, the variables This represents the overall flow direction confidence index. This value is used to quantitatively assess the reliability of the current pipe segment's topological flow direction. If this value is lower than a preset threshold, it indicates that there may be a logical conflict in the system. (Variable) The normalized quantized value representing the first evidence vector is determined by the ratio of the difference in RTK measured pipe bottom elevation between adjacent nodes to the pipe segment length, reflecting the significance and directional consistency of the physical slope. Variable The normalized quantized value representing the second evidence vector is derived from the terrain slope component along the pipeline direction obtained from digital elevation model analysis. (Variable) The normalized quantized value representing the third evidence vector is quantified by the height difference of the liquid level traces in the upstream and downstream inspection wells, and is used to indicate the actual water flow direction. Parameters , and These are dimensionless weighting coefficients, and they typically satisfy the normalization condition. The specific values ​​of these three coefficients are dynamically adjusted according to the working conditions of the data acquisition scenario. For example, in areas with high groundwater levels or severe geological subsidence, the system will automatically reduce the values. The value and increase The value of is used to reduce the weight of misjudgments in physical slope measurements caused by geological subsidence.

[0038] In the preferred scheme, step S4, which involves logical conflict detection and dynamic correction of the initial vector pipeline topology model, includes a correction algorithm for reverse slope anomalies. Calculate the RTK physical slope between two adjacent critical nodes. If the physical slope is negative, it is judged as a suspected reverse slope. Retrieve on-site survey feature data of the two adjacent key nodes; If a third evidence vector exists and indicates that the flow direction is opposite to the physical slope direction, and the comprehensive flow direction confidence index exceeds a preset threshold, the RTK physical slope is determined to be a pseudo reverse slope caused by settlement. The system automatically locks the upstream node coordinates, forcibly modifies the downstream node's logical flow direction attribute to be consistent with the third evidence vector, and marks the pipe segment as a physical settlement reverse slope pipe segment in the database.

[0039] This step details the automated correction logic for abnormal reverse slopes in drainage pipe networks, aiming to distinguish between genuine construction reverse slopes and spurious reverse slopes caused by geological settlement. The algorithm first calculates the geometric relationship between two adjacent key nodes to derive the physical slope based on RTK measured data, denoted as [the physical slope is not specified in the original text]. Its calculation formula is expressed as follows: In this formula, the variables The measured elevation value representing the endpoint of the pipe section to be determined is a variable. The measured elevation value representing the starting point of the pipe section to be determined is a variable. This represents the horizontal projected distance between two nodes. When the calculated... When the value is negative, it indicates that the pipeline is in a physical state where the downstream is higher than the upstream, and the system will then mark the pipeline segment as a suspected reverse slope object.

[0040] To further verify the nature of the reverse slope, the system automatically retrieves on-site survey feature data associated with these two key nodes for multi-dimensional verification. The core of the verification lies in introducing a third evidence vector and a comprehensive flow direction confidence index. A joint judgment is performed. The judgment logic executed by the system must simultaneously meet the following three conditions: First, it is confirmed that there is a valid third evidence vector, namely the highest liquid level trace line data, in the field data; second, it is confirmed that the water flow direction indicated by the third evidence vector is completely opposite to the physical slope direction calculated by RTK; and finally, it is confirmed that the comprehensive flow direction confidence index calculated in the previous steps is... Greater than the preset judgment threshold .variable It is an empirical constant used to set the minimum confidence standard for accepting on-site trace evidence to overturn the conclusions of physical measurements.

[0041] Once all the above conditions are met, the system will determine that the current negative slope is a pseudo-reverse slope phenomenon caused by geological subsidence or manhole subsidence, rather than the original design reverse slope. Based on this determination, the system will perform automated correction operations, specifically including locking the spatial coordinates of the upstream node to maintain the authenticity of its measurement data, and forcibly modifying the logical flow direction attribute of the downstream node to align it with the direction indicated by the third evidence vector, thereby restoring the correct topological connection relationship. Finally, the system will specifically mark the attribute field of this pipe segment as a physically subsided reverse slope pipe segment in the database, so that the physical subsidence problem existing in this pipe segment can be identified in subsequent pipeline maintenance and hydraulic model analysis.

[0042] In the preferred embodiment, step S4, which involves logical conflict detection and dynamic correction of the initial vector pipeline topology model, includes correction algorithms for disconnections and hidden connections. Retrieve isolated nodes with a degree of 1 in the initial vector network topology model; Using the isolated node as the center, set a search radius R, and search for other pipeline nodes within the search radius R; Based on the distribution range of overflow sediments in the on-site reconnaissance data, if the overflow sediments show a banded distribution pointing towards a certain downstream node within the search radius R, it is determined that there is a hidden connection relationship. A virtual connection edge is automatically generated between the isolated node and the downstream node, and the attribute of the virtual connection edge is marked as speculative hidden connection.

[0043] This technical solution addresses common issues of disconnected and concealed connections in drainage pipe network topology models by proposing an automated correction algorithm based on a combination of geometric spatial analysis and physical trace verification. The algorithm first traverses the entire initial vector pipe network topology model, retrieving all isolated nodes with a degree of 1, denoted as... In graph theory, a node with a degree of 1 usually represents the endpoint of a pipeline network. If the node is not a pre-defined discharge outlet or source, there is a high probability that there is a topological break or missing data. Therefore, it is considered a potential breakpoint object to be corrected.

[0044] The system then used this isolated node For the geometric center, a specific spatial search radius is set, denoted as . The system performs a buffer search operation in the spatial database to filter out all nodes that are isolated. The Euclidean distance is less than or equal to The neighboring pipeline nodes constitute a candidate target set. This step, from a purely geometric perspective, delineates a certain range of potential downstream nodes that may have connections with isolated nodes, providing a filtering range for subsequent physical logic judgments.

[0045] To accurately identify genuine hidden connections from the candidate set, the algorithm further incorporates overflow sediment distribution information from on-site reconnaissance data. The system transforms the recorded sediment banding pattern into a mathematical direction vector, defined as the sediment indicator vector. This vector objectively reflects the physical flow traces left on the ground or at the bottom of a well when sewage overflows or leaks, guided by gravity and underground pipes. The system determines the candidate node by calculating the cosine of the angle between this indicator vector and the geometric vector pointing from the isolated node to the candidate node. The determination formula is as follows: .

[0046] In this determination formula, the variable This represents the direction matching coefficient, used to quantify the consistency between the physical trace direction and the inferred connection direction; its value ranges from -1 to 1. (Vector) Represents the search radius from an isolated node. The geometric connection vector of a specific downstream node within the node. (Symbol) The dot product operation represents vectors, where the denominator represents the product of the magnitudes of the two vectors. When the calculated result... The value is greater than the preset confidence threshold. When the distribution trend of the ground sediments is strong and clearly points to the downstream node, it is determined that there is a hidden physical connection between the two points, i.e., a hidden connection relationship. The system will then automatically generate a new topological edge between these two points and mark the attribute status field of this virtual connection edge as a hypothetical hidden connection, thereby completing the closure repair of the topological network.

[0047] In the preferred scheme, in step S3, the initial vector network topology model is constructed and stored using a hierarchical graph data structure: The first layer is the physical node layer, which stores the geometric point elements of inspection wells, discharge outlets, and pump station inlets and outlets; The second layer is the pipeline edge layer, which stores the pipeline segment elements that connect physical nodes. Each pipeline edge element contains the start node ID, the end node ID, and the corresponding pipe diameter weight. The third layer is the attribute association layer, which uses a unique coded ID to attach on-site image data and material text data to the physical node layer or pipeline edge layer; The mobile data acquisition terminal uses an embedded SQLite database to store the hierarchical graph data locally and encapsulates it in GeoJSON format.

[0048] In step S3, the process of constructing the initial vector pipeline network topology model employs a layered graph data structure for storage. This design aims to logically decouple the complex pipeline network geographic information system data, thereby improving the efficiency of data retrieval and topology analysis. The first layer of this data structure is defined as the physical node layer, primarily responsible for storing all discrete geometric point features in the pipeline network system. These point features encompass crucial facilities in the urban drainage system, including inspection wells for maintenance and repair, final sewage discharge outlets, and the inlet and outlet locations of pumping stations. Mathematically, the data set of this layer can be represented as follows: In this set, each element It represents an independent physical node object, which contains the node's precise coordinates in three-dimensional space, namely longitude, latitude, and elevation data.

[0049] The second layer of the data structure is defined as the pipeline edge layer, built upon the physical node layer, and used to store the pipeline segment elements connecting the various physical nodes. Each pipeline segment is abstracted in the data structure as an edge in graph theory, used to describe the channel of water flow. To construct a complete topology network, each pipeline edge element not only contains geometric line data, but also must contain a unique identifier for the starting node (starting node ID) and a unique identifier for the ending node (ending node ID). These two IDs strictly point to a specific physical node in the first layer, thus determining the logical direction of water flow. Furthermore, the pipeline edge element also contains a corresponding pipe diameter weight, a parameter used for subsequent hydraulic analysis. The mathematical definition of this layer can be described as a set of edges. any one of the edges It can be represented as an ordered tuple In this tuple, A unique code representing the upstream node. A unique code representing the downstream node, while variables This represents the numerical weight of the pipe diameter for that pipe segment, used in topology analysis to calculate flow flux or as an impedance factor for path search.

[0050] The third layer of the data structure is defined as the attribute association layer. This layer's function is to mount non-geometric business attribute data onto the geometric objects of the first two layers. The system uses a unique coded ID as a foreign key to precisely associate the multimedia image data collected on-site and the textual description data of the pipe material with physical nodes or pipe edges. This design achieves separate storage and logical connection between spatial data and attribute data, avoiding data redundancy. The association relationships can be established using mapping functions. To describe. In this mapping relationship, variables A unique identifier representing any node or pipe segment, a set This represents the on-site photo data stream and material text description information bound to the object, ensuring that each pipeline facility has a complete electronic file.

[0051] For local storage on the mobile data acquisition terminal, the system uses an embedded SQLite database to manage the aforementioned layered map data. SQLite is lightweight, requires zero configuration, and supports transaction processing, making it ideal for the resource environment of mobile devices. To facilitate subsequent data exchange and network transmission, the mobile terminal serializes and encapsulates the structured data stored in the database, converting it into GeoJSON format. GeoJSON is a JSON-based geospatial data exchange standard that clearly describes geometric objects such as points, lines, and polygons, along with their attribute information, in text form, ensuring compatibility and parsing efficiency between the mobile terminal and the cloud server.

[0052] In the preferred embodiment, this method relies on a system that includes a mobile app and a cloud server for deployment. Step S5, which outputs a standardized GIS pipeline database, involves the following data synchronization and conversion processes: The mobile data acquisition terminal uploads the GeoJSON incremental data packet corrected in step S4 to the cloud server via the HTTPS encryption protocol; The cloud server uses Nginx as a reverse proxy server to receive data requests and forward them to the application middleware; The application middleware calls the topology check service to verify the data connectivity again. After successful verification, the GeoJSON data is parsed and loaded into the PostGIS spatial extension module of the PostgreSQL database using an ETL tool; The PostGIS spatial extension module executes a stored procedure to convert the data into a Shapefile or GDB format file that conforms to the urban underground pipeline data standard, and completes the database output.

[0053] In step S5, the process of outputting the standardized geographic information system, i.e., the GIS pipeline database, involves a rigorous data synchronization and format conversion process. First, the mobile data acquisition terminal, acting as the data source, establishes a secure connection with the cloud using the Hypertext Transfer Protocol Secure (HTTPS) encrypted channel. The terminal encapsulates the pipeline data, which has undergone logical correction in step S4, into incremental data packets in the lightweight geographic data exchange format, GeoJSON, and uploads them to the cloud server. This incremental upload mechanism only transmits the changed data portions, effectively reducing network bandwidth consumption and improving transmission efficiency.

[0054] Once the data reaches the cloud, the cloud server uses the high-performance reverse proxy server Nginx as the traffic entry point to receive these data requests. After load balancing and security filtering of the requests, Nginx accurately forwards them to the backend application middleware. The application middleware, as the core processing unit of business logic, automatically calls the internally integrated topology check service module. The main function of this module is to perform secondary connectivity verification on the received incremental data, ensuring that the newly submitted data and the existing data in the database can be seamlessly connected spatially and logically, preventing network topology breaks due to data updates.

[0055] After rigorous verification by the application middleware, the system initiates an ETL (Extract, Transform, Load) tool to further process the data. The ETL tool parses the semi-structured GeoJSON data into database-recognizable structured records and loads them into the open-source object-relational database PostgreSQL. To support complex spatial geometric operations, the database comes pre-installed with the PostGIS spatial information extension module. Subsequently, the PostGIS module executes a pre-written stored procedure to automatically convert the geometric records within the database into a common file format conforming to urban underground pipeline data standards, specifically either Shapefile vector file format or GDB (Geographic Database) file format. This conversion process signifies the transformation of the data from raw sketches at the acquisition end into a standard output usable by industry software, completing the final database output.

[0056] In the preferred embodiment, for areas with high groundwater levels, the correction algorithm in step S4 further includes a liquid level correction sub-step: Collect real-time groundwater level data at key nodes; When the pipe bottom elevation measured by RTK is lower than the real-time groundwater level data, calculate the influence factor of buoyancy on pipe displacement; If the impact factor exceeds the safety threshold, the weight coefficient of the first evidence vector should be reduced in the multi-source evidence confidence correction model. And increase the weight coefficient of the third evidence vector. .

[0057] During the reconstruction of drainage pipe networks in areas with high groundwater levels, soil liquefaction or loose geology can easily cause uneven settlement or drift of pipelines due to the buoyancy of the groundwater layer, thus interfering with the judgment of the original design slope. To eliminate the error caused by this environmental factor, a liquid level correction sub-step is introduced in this process. The system first obtains the real-time groundwater level data of the area by using water level sensors deployed at key locations or by consulting real-time hydrological monitoring data, denoted as [data missing]. This data represents the absolute elevation of the free water surface and is a benchmark parameter for determining whether a pipeline is submerged.

[0058] Subsequently, the system will use the pipe bottom elevation data accurately measured by the RTK equipment. Compare and analyze the data with the groundwater level data. When detected... The value is less than When the value is [value missing], it indicates that the pipe section is below the submersion line of the groundwater level, and the pipe body is subjected to the vertical Archimedes buoyancy generated by the surrounding groundwater. Under this mechanical environment, the system initiates a buoyancy influence factor calculation program to quantify the potential destructive effect of buoyancy on the vertical displacement of the pipeline.

[0059] The influence factor of buoyancy on pipe displacement is denoted as Its calculation formula is expressed as In this formula, the variables This represents the density of groundwater, typically taken as approximately 1000 kg per cubic meter. (Variable) Represents gravitational acceleration. Variable This represents the cross-sectional area of ​​the pipe submerged in the water table, calculated based on the geometric relationship between the water level difference and the pipe diameter. The variable in the denominator... The variable represents the gravity load per unit length of the pipe itself. This represents the effective compressive resistance exerted on the pipeline by the soil covering it. The formula calculates... The numerical value is a dimensionless ratio that intuitively reflects the antagonistic relationship between the upward floating force and the downward pressure resistance experienced by the pipeline.

[0060] The system presets an empirical safety threshold. If the calculated impact factor If the gradient exceeds this safety threshold, it indicates a significant buoyancy effect, and the pipeline has most likely undergone non-structural physical displacement. In this case, the geometric slope measured by RTK can no longer accurately reflect the original design flow direction logic. Therefore, the system will automatically adjust the weight allocation strategy in the multi-source evidence confidence correction model. Specifically, it will reduce the weight coefficient of the first evidence vector, i.e., the physical slope evidence. The numerical value is adjusted because the physical measurement at this time contains significant environmental interference errors. Simultaneously, the system will correspondingly increase the weighting coefficient of the third evidence vector, namely the liquid level trace evidence. The numerical values ​​are more valuable because the historical high liquid level traces inside the inspection well record the true flow direction left by long-term hydraulic operation, and are less affected by short-term physical displacement. Through this dynamic redistribution of weights, the algorithm can effectively "distinguish between true and false" in areas with high water levels and geological hazards, restoring the true topological relationship of the pipeline network.

[0061] In the preferred embodiment, after dividing the investigation blocks in step S1, the method further includes a preprocessing step for the historical drawing data within the blocks: Import historical CAD pipeline drawings into the GIS platform for georeferencing; Extract the node coordinates from historical drawings as a reference set; In step S4, if the coordinates collected by RTK deviate from the corresponding points in the reference set by more than a set threshold, but the topological connection relationship is consistent, the measured coordinates of RTK are retained in the database, and the coordinates of the historical drawings are recorded as the original drawing deviation reference value.

[0062] The specific operation begins by importing historical drainage network drawings stored in computer-aided design software (CAD format) into a geographic information system (GIS) platform. Since historical drawings often use independent local coordinate systems or non-standard drawing scales, the system must perform georeferencing. This operation selects control points on the drawing that match actual geographical features, and uses affine or polynomial transformation algorithms to precisely distort and align the spatial positions of the CAD drawings to the unified geographic coordinate system used in the current project, thereby achieving spatial overlay of historical data with the real geographical environment.

[0063] After registration, the system uses vector data extraction algorithms to extract coordinate data of key nodes such as manholes and discharge outlets in batches from historical drawings. This coordinate data obtained from old drawings is compiled into a separate dataset, defined as the reference set. Logically, the reference set represents the theoretical spatial location of the pipeline network at the initial construction stage or during the last survey. Although these data may contain accuracy errors due to their age, the recorded pipeline connection relationships often have significant reference value, providing prior knowledge for subsequent topology reconstruction.

[0064] In the subsequent step S4, the system uses this reference set to perform a differentiated processing logic based on "topology preservation but geometric correction". The system automatically calculates the Euclidean distance deviation between the latest measured coordinates acquired by the RTK device and the corresponding point coordinates in the reference set, denoted as . Simultaneously, the system checks whether the connectivity between the two is consistent at the graph theory level. When the calculated positional deviation... Exceeding the preset tolerance threshold However, when the topological connections between the two nodes (i.e., the upstream and downstream relationships between nodes) perfectly match, the system determines that this is a simple spatial location drift phenomenon, rather than a logical error. In this case, the database update strategy follows the principle of "physical priority, archival retention": the system forcibly retains the high-precision coordinates measured by RTK as the final valid geometric attribute of the node to ensure the timeliness of the map; at the same time, the system does not directly discard historical coordinates with huge deviations, but records them as an attribute field as a reference value for the original map deviation. This approach corrects spatial location errors while preserving the traceability of historical changes, providing a quantitative basis for subsequent analysis of geological subsidence or historical mapping errors.

[0065] In the preferred embodiment, a Docker containerized knowledge base service is deployed on the cloud server to store the on-site image data collected in step S2; When outputting a standardized GIS pipeline database, the system automatically generates a URL link pointing to the knowledge base service and writes the URL link into the attribute table of the GIS data, realizing a cloud-based hyperlink index of pipeline spatial data and on-site survey photos.

[0066] This technical solution deploys a knowledge base service based on Docker containerization technology within a cloud server architecture. This deployment method ensures the consistency and isolation of service operation across different computing environments by packaging the application and its dependent environments into lightweight containers. This knowledge base service is specifically used for the centralized storage and management of the massive amounts of on-site image data synchronously collected via mobile terminals in step S2, including unstructured data such as photos of the interior of inspection wells, close-ups of pipe openings, and panoramic views of the surrounding environment, enabling efficient archiving and retrieval of image data.

[0067] When the system executes the operation of outputting a standardized geographic information system (GIS) network database, the background program automatically triggers the Uniform Resource Locator (URL) generation mechanism. Based on the storage path and access protocol of each image data in the knowledge base, the system automatically constructs a unique network link address pointing to that specific image resource. These links are not merely simple text strings, but effective indexes with network redirection capabilities, enabling direct access to cloud resources via Hypertext Transfer Protocol (HTTP) in browsers or specialized software.

[0068] Next, the system batch-writes the generated URLs into the attribute table of the GIS vector data, typically stored in the reserved "Image Attachments" or "Site Photos" fields. This step completes the deep integration of spatial geographic information and multimedia imagery, constructing a cloud-based hyperlink index between pipeline spatial data and site survey photos. This means that in subsequent operation and maintenance management or engineering design, operators only need to click on specific pipeline facility elements on the electronic map to retrieve and view the corresponding high-definition site photos in real time through the hyperlinks in the attribute table, realizing instant association and visual verification from two-dimensional vector graphics to three-dimensional real-scene images.

[0069] Example 2 Further explanation in conjunction with Example 1, such as Figure 1-3 As shown in the attached diagram, this embodiment takes the renovation project of the xxx old drainage pipe network as an example, and elaborates in detail the specific implementation process of the drainage pipe network spatial topology dynamic reconstruction method based on RTK and field survey. Figure 1 As shown, the overall process of this method includes five core steps: dividing the investigation area into blocks, high-precision data acquisition, initial model construction, multi-source evidence correction, and final cloud output. The implementation work first divides the specific investigation blocks according to the service range of the main pumping station and the hierarchical relationship of the catchment units in the area. Historical CAD pipeline network drawings are then imported into the GIS platform for georeferencing, and the coordinates of inspection wells are extracted as a reference set for subsequent comparisons.

[0070] During the on-site operation phase, personnel bring RTK equipment and tablets into the block. For example... Figure 3As shown, the mobile data acquisition terminal software interface adopts a dark engineering color scheme adapted to strong outdoor light. The status indicator bar at the top of the screen displays the satellite positioning status in real time as an RTK fixed solution, and scrolls to display the horizontal and vertical positioning accuracy down to the centimeter level, ensuring the basic quality of data acquisition. After the operator establishes a Bluetooth connection between the RTK and the terminal, the RTK device transmits the NMEA data stream to the terminal. For example... Figure 2 In the system principle flow shown, the data processing module located at the mobile edge computing layer parses the NMEA data stream, uses built-in parameters to convert WGS84 coordinates into local engineering coordinates in real time, and synchronously writes them into the Exif information of the site photos. At this time... Figure 3 As shown in the software interface map work area, every time the operator confirms a manhole location, a blue solid node and a gray connecting pipe section are generated in real time on the interface base map, realizing real-time data visualization.

[0071] The workers then Figure 3 The lower half of the interface, in its collapsible panel, displays pipe diameter information such as 300 mm and material (reinforced concrete). The system backend then processes this information accordingly. Figure 2 The initial vector pipeline topology model construction logic shown utilizes a layered graph data structure to dynamically bind the physical node layer, pipeline edge layer, and attribute association layer. When the system detects that RTK measured data for a certain pipeline segment shows an upstream pipe bottom elevation of 3.5 meters and a downstream elevation of 3.6 meters, resulting in a negative calculated physical slope, a logic conflict detection is immediately triggered. At this time, Figure 3 The software interface map will highlight the pipe section in a striking red in the center and pop up a warning icon marked with an exclamation mark indicating a possible reverse slope, visually alerting the workers that there is a logical conflict.

[0072] In response to this anomaly, the system was started. Figure 2 The multi-source evidence confidence correction model shown in the core layer is used for calculation. Operators use... Figure 3 In the multi-source evidence correction area at the bottom right of the interface, the physical slope value was observed to be -1.5%. The "Take Photo" button was then clicked to capture an image of the well. The on-site photo showed a clear highest liquid level mark on the well wall, indicating a downstream flow. The operator selected downstream flow for the liquid level mark in the compass area of ​​the interface. The backend system, combined with the real-time acquired groundwater level data of 3.4 meters, used a formula... The buoyancy influence factor was calculated, and the results show that the pipeline is significantly affected by buoyancy. The system is based on... Figure 2 The logical decision path shown automatically adjusts the weight coefficients, reducing the weight of the physical slope. And increase the weight of liquid level traces .

[0073] Through the core formula The calculated overall flow confidence index exceeds a preset threshold. For example... Figure 3 As shown on the interface, the system immediately pops up a dynamic prompt box, suggesting confirmation of a pseudo-inverse slope and correction. After the operator clicks the execute button, the system automatically locks the upstream node coordinates, forcibly modifies the logical flow direction of the downstream node, and marks the pipe segment as a physical settlement inverse slope segment in the database. After the correction is completed, the mobile terminal uploads the encrypted GeoJSON incremental data packet. Figure 2 As shown at the bottom, the data enters the cloud via Nginx reverse proxy, undergoes a secondary topology check by the application middleware, and is then loaded into the PostGIS spatial database using an ETL tool. The final output is a standardized Shapefile format GIS network database containing a complete chain of on-site evidence.

[0074] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be defined as the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A method for dynamic reconstruction of the spatial topology of drainage pipe networks based on the fusion of RTK and field survey, characterized by: The method includes: S1: The investigation area is divided based on the service range of the pumping station and the hierarchical level of the water catchment unit; S2: Establish communication connection between RTK equipment and mobile acquisition terminal, collect three-dimensional coordinate data of key nodes in pipeline network, synchronously collect on-site image data, and generate pipeline network spatial sketch in real time. S3: Input pipe diameter, material and flow direction attribute information into the mobile acquisition terminal, associate the three-dimensional coordinate data with the attribute information, and construct an initial vector pipe network topology model; S4: Obtain on-site survey feature data, construct a multi-source evidence confidence correction model, and use this model to perform logical conflict detection and dynamic correction on the initial vector pipeline topology model; S5: Map the corrected topology model to a standardized GIS pipeline database and output it.

2. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 1, characterized in that: Step S2 In this process, the steps for establishing a communication connection and data processing between the RTK device and the mobile acquisition terminal include: RTK devices transmit NMEA format positioning data streams to mobile acquisition terminals via Bluetooth or serial communication protocols; The solution module in the mobile acquisition terminal parses the NMEA data stream, extracts longitude, latitude and ellipsoidal elevation information, and uses built-in local coordinate transformation parameters to convert the WGS84 coordinate system to the local engineering coordinate system in real time. The mobile acquisition terminal maps the converted coordinate points onto the preloaded electronic map base map. When the acquisition point is a manhole or pipe opening, it triggers an image acquisition command, calls the terminal camera to take a picture of the scene, and writes the high-precision location data acquired by the current RTK into the Exif information of the picture.

3. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 1, characterized in that: Step S4 In this context, the construction steps of the multi-source evidence confidence correction model include: Define the physical slope calculated from the measured pipe bottom elevation using RTK as the first evidence vector; Define the natural slope trend of the surrounding terrain as the second evidence vector; Define the flow direction indicated by the highest liquid level trace line on the well wall of the inspection well as the third evidence vector; Assign weight coefficients to the first evidence vector, the second evidence vector, and the third evidence vector respectively. , , The overall flow confidence index is obtained through weighted calculation.

4. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 3, characterized in that: In step S4, the logical conflict detection and dynamic correction of the initial vector pipeline topology model includes a correction algorithm for reverse slope anomalies: Calculate the RTK physical slope between two adjacent critical nodes. If the physical slope is negative, it is judged as a suspected reverse slope. Retrieve on-site survey feature data of the two adjacent key nodes; If a third evidence vector exists and indicates that the flow direction is opposite to the physical slope direction, and the comprehensive flow direction confidence index exceeds a preset threshold, the RTK physical slope is determined to be a pseudo reverse slope caused by settlement. The system automatically locks the upstream node coordinates, forcibly modifies the downstream node's logical flow direction attribute to be consistent with the third evidence vector, and marks the pipe segment as a physical settlement reverse slope pipe segment in the database.

5. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 1, characterized in that: In step S4, the logical conflict detection and dynamic correction of the initial vector pipeline topology model includes correction algorithms for disconnections and hidden connections: Retrieve isolated nodes with a degree of 1 in the initial vector network topology model; Using the isolated node as the center, set a search radius R, and search for other pipeline nodes within the search radius R; Based on the distribution range of overflow sediments in the on-site reconnaissance data, if the overflow sediments show a banded distribution pointing towards a certain downstream node within the search radius R, it is determined that there is a hidden connection relationship. A virtual connection edge is automatically generated between the isolated node and the downstream node, and the attribute of the virtual connection edge is marked as speculative hidden connection.

6. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 1, characterized in that: In step S3, the initial vector network topology model is constructed and stored using a hierarchical graph data structure: The first layer is the physical node layer, which stores the geometric point elements of inspection wells, discharge outlets, and pump station inlets and outlets; The second layer is the pipeline edge layer, which stores the pipeline segment elements that connect physical nodes. Each pipeline edge element contains the start node ID, the end node ID, and the corresponding pipe diameter weight. The third layer is the attribute association layer, which uses a unique coded ID to attach on-site image data and material text data to the physical node layer or pipeline edge layer; The mobile data acquisition terminal uses an embedded SQLite database to store the hierarchical graph data locally and encapsulates it in GeoJSON format.

7. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 1, characterized in that: This method relies on a system that includes a mobile app and a cloud server for deployment. Step S5, which outputs a standardized GIS pipeline database, involves the following data synchronization and conversion processes: The mobile data acquisition terminal uploads the GeoJSON incremental data packet corrected in step S4 to the cloud server via the HTTPS encryption protocol; The cloud server uses Nginx as a reverse proxy server to receive data requests and forward them to the application middleware; The application middleware calls the topology check service to verify the data connectivity again. After successful verification, the GeoJSON data is parsed and loaded into the PostGIS spatial extension module of the PostgreSQL database using an ETL tool; The PostGIS spatial extension module executes a stored procedure to convert the data into a Shapefile or GDB format file that conforms to the urban underground pipeline data standard, and completes the database output.

8. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 4, characterized in that: For areas with high groundwater levels, the correction algorithm in step S4 also includes a liquid level correction sub-step: Collect real-time groundwater level data at key nodes; When the pipe bottom elevation measured by RTK is lower than the real-time groundwater level data, calculate the influence factor of buoyancy on pipe displacement; If the impact factor exceeds the safety threshold, the weight coefficient of the first evidence vector should be reduced in the multi-source evidence confidence correction model. And increase the weight coefficient of the third evidence vector. .

9. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 1, characterized in that: After dividing the investigation areas into blocks in step S1, the process also includes a preprocessing step for historical drawing data within each block: Import historical CAD pipeline drawings into the GIS platform for georeferencing; Extract the node coordinates from historical drawings as a reference set; In step S4, if the coordinates collected by RTK deviate from the corresponding points in the reference set by more than a set threshold, but the topological connection relationship is consistent, the measured coordinates of RTK are retained in the database, and the coordinates of the historical drawings are recorded as the original drawing deviation reference value.

10. The method for dynamic reconstruction of drainage network spatial topology based on RTK and field survey as described in claim 7, characterized in that: A Docker containerized knowledge base service is deployed on the cloud server to store the on-site image data collected in step S2; When outputting a standardized GIS pipeline network database, the system automatically generates a URL link pointing to the knowledge base service and writes the URL link into the attribute table of the GIS data, realizing a cloud-based hyperlink index of pipeline network spatial data and on-site survey photos.

Images