Underground pipe network twin modeling system based on autonomous surveying and mapping data

Abstract

The invention discloses an underground pipe network twin modeling system based on autonomous surveying and mapping data. The system comprises a surveying and mapping data acquisition module, a data extraction module, a twin modeling module, a correction module and an association module. A surveying and mapping data acquisition module acquires point cloud and mileage data through an in-pipe crawler, and an original three-dimensional point cloud sequence with an accurate space-time label is generated through fusion processing. The data extraction module performs circle fitting and feature sequence analysis on the point cloud, and generates a feature vector and a preliminary pipe network topology network through template matching; the twinborn modeling module automatically constructs a parameterized twinborn model according to the network; the correction module realizes incremental updating of the model by comparing new and old feature vectors; and the association module allocates a unique identification code to each part to realize accurate association of the multi-source data and the three-dimensional model. According to the method, efficient incremental updating of the parameterized twin model is realized through the correction module, and closed-loop management from surveying and mapping, automatic modeling and updating to data association is realized.

Description

Technical Field

[0001] This invention relates to the field of underground pipeline twin modeling technology, specifically to an underground pipeline twin modeling system based on autonomous surveying data. Background Technology

[0002] The underground pipeline network system of large industrial plants (such as power plants, chemical plants, and steel mills) is the core artery for energy transmission, material transport, and emission treatment. It typically consists of multiple subsystems, including water supply, drainage, heating, gas, power cables, and industrial pipelines. However, underground pipeline networks are gradually formed over decades or even longer construction periods, and generally suffer from numerous problems such as complex spatial distribution, inconsistent pipeline materials, and incomplete construction standards and as-built documentation. Therefore, an underground pipeline twin modeling system is needed to facilitate a three-dimensional display of the spatial location, connection relationships, and operational status of underground pipeline networks.

[0003] For example, the Chinese patent announcement number CN 119989598 B discloses a method and system for underground pipe network modeling based on feature location calibration, which improves the accuracy and practicality of three-dimensional models of underground pipe networks and enables real-time updates of pipe network status.

[0004] However, most current underground pipeline twin modeling systems suffer from the following problems: the twin model is disconnected from the real-time status: existing visualization platforms are mostly simple overlays of static 3D scenes and 2D data panels, failing to establish a deep semantic relationship between pipeline entities and their real-time monitoring data in 3D space; at the same time, there is a lack of effective model update and version management mechanisms: after local pipeline modifications, existing systems lack the ability to quickly update local models based on new on-site survey data, often requiring overall reconstruction, resulting in high costs. Summary of the Invention

[0005] To address this, the present invention provides an underground pipeline twin modeling system based on autonomous surveying data, in order to solve the problems in the prior art of failing to establish a deep semantic association between the pipeline entity and its real-time monitoring data in three-dimensional spatial location and lacking local model update capabilities.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The underground pipeline twin modeling system based on autonomous surveying data includes a surveying data acquisition module, a data extraction module, a twin modeling module, a correction module, and a correlation module.

[0008] The mapping data acquisition module can collect and synchronize the pipeline cross-section contour point cloud sequence and mileage data according to the in-pipe crawler, and form the original three-dimensional point cloud sequence data inside the pipeline with spatiotemporal labels through real-time correction and fusion.

[0009] The data extraction module can perform least-squares circle fitting on each frame of the original 3D point cloud to extract the basic cross-sectional features. These features are then arranged according to the acquisition trajectory to form a continuous feature sequence. The similarity between the actual directional changes of the feature sequence and the standard pipe fitting template is then calculated. The pipe fitting types are determined and feature vectors are generated to form a preliminary pipe network topology.

[0010] Similarity The calculation formula is as follows:

[0011]

[0012] in, For the actual sequence number The vector and the standard pipe fitting template sequence The distance between vectors For all regularized paths in the distance matrix;

[0013] The twin modeling module constructs a parametric twin model based on the preliminary pipeline topology network and a pre-set parametric parts library;

[0014] The correction module can establish an incremental update mechanism based on data comparison, quickly locate the changed area using spatial index, determine the change type by combining the difference analysis algorithm, and dynamically correct the parameterized twin model.

[0015] The association module is used to assign a globally unique spatiotemporal entity identifier to each parametric part in the parametric twin model. The spatiotemporal entity identifier is permanently bound to the part's record in the global topology association table and its source feature vector.

[0016] Furthermore, during the real-time correction and fusion, the line laser scanner, inertial measurement unit, and odometer are synchronized using the same clock source;

[0017] Inside the pipeline, the pose is updated recursively using data collected by the inertial measurement unit as the main source of information, through the inertial navigation equations; at the same time, the displacement data from the odometer is used as a local observation for real-time correction.

[0018] Furthermore, the specific content of the data extraction module is as follows:

[0019] 1) Extraction of basic cross-sectional features;

[0020] 2) Arrange the basic features in trajectory order;

[0021] 3) Generate pipeline feature vectors.

[0022] Furthermore, the calculation formulas for the center coordinates, diameter, and roundness deviation values ​​of the fitted cross-sectional basic features and pipe deformation are as follows:

[0023]

[0024]

[0025]

[0026] in,( , ( ) represents the coordinates of the center of the circle on the two-dimensional fitting plane. To fit the radius and diameter of the circle , This represents the roundness deviation value due to pipe deformation. C and D are both parameter values.

[0027] Furthermore, the twin modeling module includes a matching submodule and a connection submodule;

[0028] The matching submodule can retrieve the corresponding geometric generation function from the parameterized parts library based on the feature vectors in the preliminary pipeline topology network, and call the modeling engine to generate a single part representation model.

[0029] The connection submodule can determine whether the interface coordinates of two parts coincide within the tolerance range based on the coordinates of the two ends in the representation model of adjacent individual parts; if they coincide, the interfaces of adjacent parts are automatically aligned and matched according to the preliminary pipeline topology network.

[0030] Furthermore, the correction module includes a retrieval submodule, a calling submodule, and an update submodule;

[0031] The retrieval submodule is constructed based on the octree spatial partitioning algorithm, which can build a spatial index of the preliminary pipeline topology network and all the feature vectors it contains.

[0032] The calling submodule can call the surveying data acquisition module and the data extraction module to perform a new round of operations on the pipeline area, generating a new feature vector set describing the changed area and a new preliminary pipeline topology network.

[0033] Furthermore, the update submodule uses spatial indexing to quickly locate the set of pipeline components that overlap with the spatial range of the new feature vector set. It then compares each new feature vector with the corresponding feature parameters of the old model within the overlapping area, calculating the degree of deviation between the new and old feature vectors on key parameters. The calculation formula is as follows:

[0034]

[0035] in, The diameter of the pipe in the new feature vector. The pipe diameter in the original parametric twin model, The coordinates of the center of the circle in the new feature vector. The coordinates of the center of the circle in the original parametric twin model. and All are weighting coefficients.

[0036] Furthermore, if the degree of deviation If the value exceeds the set threshold and the spatial coordinate ranges corresponding to the old and new feature vectors overlap, it is determined that the pipeline parameters have been modified; if an old feature vector exists within a specific spatial coordinate range, but after spatial index retrieval, no new vector with matching spatial coordinates can be found in the new feature vector set, it is determined that the pipeline parameters have been deleted.

[0037] This invention offers the following advantages: It achieves efficient and traceable incremental updates of the parametric twin model through a correction module. After pipeline network modifications, the system re-engineers the surveying data acquisition and extraction modules, generating a new set of feature vectors describing the changed areas. Utilizing an octree-based spatial index, the system quickly locates the set of components to be updated. By calculating the deviation of key parameters such as pipe diameter and center coordinates between the old and new feature vectors, the twin modeling module is then invoked to reconstruct the local models of the changed components and refresh the global topology association table, avoiding overall reconstruction and reducing update costs.

[0038] Simultaneously, by assigning a globally unique spatiotemporal entity identifier to each parametric part generated by the twin modeling module, it is permanently bound to the feature vector and the global topology association table, forming a unique index for cross-system association. The spatial index of the correction module is automatically attached, and all data is categorized into the spatiotemporal event book of the corresponding part according to a unified time scale. This makes the parametric twin model no longer a static geometric display, but can dynamically integrate and intuitively present multi-dimensional information of the pipeline network, improving the overall modeling effect.

[0039] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. Attached Figure Description

[0040] To more intuitively illustrate the prior art and this application, exemplary drawings are provided below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing this application; for example, based on the technical concept disclosed in this application and the exemplary drawings, those skilled in the art are able to easily make conventional adjustments or further optimizations to the addition / reduction / classification, specific shapes, positional relationships, connection methods, size ratios, etc. of certain units (components).

[0041] Figure 1 This is a flowchart illustrating the implementation process of the underground pipeline twin modeling system based on autonomous surveying data, as described in this invention. Detailed Implementation

[0042] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these embodiments are merely for further explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Technical engineers in the field can make some non-essential improvements and adjustments to the present invention based on the above-described content. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Please see Figure 1 The underground pipeline twin modeling system based on autonomous surveying data includes a surveying data acquisition module, a data extraction module, a twin modeling module, a correction module, and a correlation module.

[0044] The mapping data acquisition module can collect and synchronize the pipeline cross-section contour point cloud sequence and mileage data based on the in-pipe crawler, and form the original three-dimensional point cloud sequence data of the pipeline interior with spatiotemporal labels through real-time correction and fusion, thereby facilitating the direct acquisition of geometric information inside the pipeline and overcoming the limitations of surface detection in perceiving the internal state of the pipeline.

[0045] The pipe crawler integrates a line laser scanner, an inertial measurement unit (IMU), and an odometer. As the crawler moves inside the pipe, the line laser scanner heads at both ends rotate and emit lasers at a fixed frequency, scanning the inner wall of the pipe around its entire length, thereby continuously acquiring cross-sectional profile point cloud data of the pipe. The cross-sectional profile point cloud data includes the actual inner diameter of the pipe at that location, its geometry (such as circular or elliptical deformations), and the outline of any attachments on the inner wall.

[0046] The inertial measurement unit is used to continuously measure the acceleration and angular velocity of the crawler inside the tube, and calculates the attitude and position changes through integration. The odometer is used to collect the relative movement distance of the crawler.

[0047] During real-time correction and fusion, the line laser scanner, inertial measurement unit (IMU), and odometer are first synchronized using the same clock source, and a global coordinate reference is provided by a GNSS (Global Navigation Satellite System) system enhanced by ground control points. Inside the pipeline, the pose is updated recursively using inertial navigation equations, primarily based on IMU data; simultaneously, the odometer displacement data is used as local observations for real-time correction to suppress rapid IMU drift. The calculation formula is as follows:

[0048]

[0049] in, To update the pose, K is a coefficient. Relative displacement provided for the odometer. Update pose. The calculation formula is as follows:

[0050] in, Acceleration measured by the IMU It is the acceleration due to gravity. The sampling time interval, No. The speed of the crawler at any given time, relative to the global coordinate system. For the first The position of the crawler at any given time relative to the global coordinate system.

[0051] When the crawler passes a well or inlet with known coordinates of the ground control point, it compares the calculated predicted position with the known absolute coordinates. The resulting difference is used to correct the accumulated positioning error, and then outputs the original 3D point cloud sequence data of the pipeline interior with accurate spatiotemporal labels, providing a reliable data foundation for subsequent automatic feature extraction and twin construction.

[0052] The data extraction module performs least-squares circle fitting on each frame of the original 3D point cloud, extracting basic cross-sectional features including the coordinates of the fitted circle's center, diameter, and roundness deviation. These features are then arranged sequentially according to the acquisition trajectory, forming a continuous feature sequence. The module automatically identifies pipe fitting feature locations by analyzing abrupt changes in the sequence. Finally, based on a dynamic time warping algorithm, the similarity between the actual direction change and a standard pipe fitting template is calculated. This determines the pipe fitting type and generates a structured feature vector containing geometric parameters and connection relationships, thus forming a preliminary pipe network topology. This facilitates the automated and structured transformation of the original point cloud data into an engineering semantic description capable of driving the automatic generation of 3D twins. A structured feature vector is generated for each identified pipe segment and fitting, forming a preliminary pipe network topology. Specific details are as follows:

[0053] 1) Extraction of basic cross-sectional features;

[0054] First, the point cloud is rotated from the global coordinate system to a local two-dimensional planar coordinate system. For each frame of the 3D point cloud, a least-squares circle fit is performed to extract the basic features of the cross-section, including the coordinates of the fitted circle center, the diameter, and the roundness deviation value of the pipe deformation. The calculation formula is as follows:

[0055]

[0056]

[0057]

[0058]

[0059] in,( , ( ) represents the coordinates of the center of the circle on the two-dimensional fitting plane. To fit the radius and diameter of the circle , This represents the roundness deviation value due to pipe deformation. C and D are both parameter values, and the calculation formula is as follows:

[0060] in,( ) is the first The two-dimensional coordinates of each point projected onto the fitting plane This represents the number of valid points in the current cross-sectional point cloud.

[0061] 2) Arrange the basic features in trajectory order;

[0062] The extracted basic features are arranged in the order of the trajectory to form a continuous feature sequence, and it is analyzed whether there are pipe fitting feature positions, such as the start and end points of elbows or the branch points of tees.

[0063] 3) Generate pipeline feature vectors

[0064] Based on the pattern matching algorithm, the feature sequence of the center direction vector of the continuous cross section is compared with the standard pipe fitting templates of standard elbows, tees, and other pipe fittings. Calculation. Based on the similarity calculation results, the pipe fitting type is determined, a feature vector is created for the identified pipe fitting type, and its connection relationship with adjacent components is recorded according to its order in the trajectory, forming a preliminary pipe network topology. Similarity The calculation formula is as follows:

[0065]

[0066] in, For the actual sequence number The vector and the standard pipe fitting template sequence The distance between vectors , This is the distance matrix formed by the Euclidean distance between each pair of vectors in the feature sequence and the standard pipe fitting sequence in the standard pipe fitting template. Let the distance matrix contain all regular paths such that the sum of the local distances of all points on the path is minimized. For the first A feature sequence, For the first A standard pipe fitting sequence in a standard pipe fitting template.

[0067] The twin modeling module constructs a parametric twin model based on the preliminary pipeline topology network and a pre-set parametric parts library. The parametric parts library includes the geometric generation function and assembly interface rules corresponding to each part.

[0068] The twin modeling module includes a matching submodule and a connection submodule. The matching submodule retrieves the corresponding geometric generation function from the parametric parts library based on the feature vectors in the initial pipeline topology. It then passes the specific forming parameters from the feature vectors as input parameters to this function. The function then executes, calling the kernel modeling engine to generate a single part representation model with precise dimensions and located in the correct global coordinate system.

[0069] The connection submodule can represent the coordinates of both ends in the model based on adjacent individual parts. It determines whether the interface coordinates of two parts coincide within the tolerance range; if they coincide, a rigid connection relationship is established between them. At the same time, it automatically aligns and matches the interfaces of adjacent parts according to the preliminary pipeline network topology to ensure the geometric continuity of the model, and records this adjacency-connection relationship in the global topology association table, thereby facilitating the integration of discrete parts into a complete pipeline twin model.

[0070] The correction module enables dynamic correction of the parametric twin model by establishing an incremental update mechanism based on data comparison, quickly locating changed areas using spatial indexing, and determining the change type using a difference analysis algorithm. The correction module includes a retrieval submodule, a calling submodule, and an update submodule.

[0071] The retrieval submodule is built based on the octree spatial partitioning algorithm. It can construct a spatial index from the preliminary pipeline topology network and all the feature vectors it contains. This index is used to register each part into the index according to the global coordinate range recorded in its feature vector, thereby achieving fast retrieval based on spatial location.

[0072] The submodule can call the surveying data acquisition module and the data extraction module to perform a new round of operations on the pipeline area, generating a new feature vector set describing the changed area and a new preliminary pipeline topology network.

[0073] The update submodule uses spatial indexing to quickly locate the set of pipeline components that overlap with the spatial range of the new feature vector set. It then compares each new feature vector with the corresponding feature parameters of the old model within the overlapping region. Finally, it calculates the degree of deviation between the new and old feature vectors on key parameters. The calculation formula is as follows:

[0074]

[0075] in, The diameter of the pipe in the new feature vector. The pipe diameter in the original parametric twin model, The coordinates of the center of the circle in the new feature vector. These are the coordinates of the center of the circle in the original parametric twin model. and All are weighting coefficients.

[0076] Degree of deviation Compare with a preset judgment threshold: If the spatial coordinate ranges (based on the center coordinate C) of the new and old feature vectors coincide, and the degree of deviation is... If the value exceeds the set threshold, it is determined that the parameter of the component has been modified; if an old feature vector exists within a specific spatial coordinate range, but after spatial index retrieval, no new vector with matching spatial coordinates can be found in the new feature vector set, it is determined that the component has been deleted; conversely, if a feature vector exists in the new feature vector set, and its spatial coordinate range cannot be matched with any old model component through indexing, it is determined that a new component has been added to that region.

[0077] Based on the judgment results, the correction module calls the matching and connection sub-modules of the twin modeling module to perform incremental updates. Only for modified and newly added components, the parameterized parts library is driven to regenerate the model, and the global topology association table is refreshed according to the new preliminary pipeline topology.

[0078] The association module is used to assign a globally unique spatiotemporal entity identifier to each parametric part in the parametric twin model. This identifier is permanently bound to the part's record in the global topology association table and its source feature vector.

[0079] The association logic of the association module is divided into two categories based on the characteristics of the data itself: For data with its own spatiotemporal entity identification code, the system directly maps it to the corresponding 3D model; for data that only carries spatial coordinates, the system calls the spatial index established and maintained by the correction module to quickly retrieve the model component where the coordinate is located or the nearest one, and completes the connection through its bound identification code.

[0080] After the association is completed, all data is assigned a unified timestamp and archived in time series into the spatiotemporal event book of the corresponding model component. This event book integrates all multi-source data related to the component along a timeline.

[0081] For example, a pipeline's record may include corrosion profiles parsed from raw 3D point cloud sequence data, real-time sensor readings, and historical work orders.

[0082] This invention achieves efficient and traceable incremental updates of the parametric twin model through a correction module. When the pipeline network is modified, the driving survey data acquisition and data extraction modules are reworked to generate a new set of feature vectors describing the changed areas. Utilizing an octree-based spatial index, the system quickly locates the set of components to be updated. By calculating the deviation of key parameters such as pipe diameter and center coordinates between the old and new feature vectors, the twin modeling module is then invoked to perform local model reconstruction and global topology table refresh for the changed components, avoiding overall reconstruction and reducing update costs.

[0083] Simultaneously, by assigning a globally unique spatiotemporal entity identifier to each parametric part generated by the twin modeling module, it is permanently bound to the feature vector and the global topology association table, forming a unique index for cross-system association. The spatial index of the correction module is automatically attached, and all data is categorized into the spatiotemporal event book of the corresponding part according to a unified time scale. This makes the parametric twin model no longer a static geometric display, but can dynamically integrate and intuitively present multi-dimensional information of the pipeline network, improving the overall modeling effect.

[0084] The following is an application example of replacing a section of corroded and old pipeline in a thermal power plant. The specific implementation steps are as follows:

[0085] 1. Data Acquisition and Modeling

[0086] Before replacement, the surveying data acquisition module drives the pipe crawler to work on the old pipe section and generate the original three-dimensional point cloud sequence data of the section.

[0087] The data extraction module processes this data, generates a feature vector describing the old pipeline (e.g., {type: "straight pipe", starting coordinates: (X1, Y1), ending coordinates: (X2, Y2), diameter: 0.5m}), and integrates it into the plant's initial pipeline topology network.

[0088] The twin modeling module generates the corresponding 3D model part based on this vector and updates the global topology association table. The association module binds a spatiotemporal entity identifier code to the part and archives its historical corrosion detection data to its spatiotemporal event book.

[0089] After the physical pipeline replacement is completed, the update process is initiated for the newly built pipeline section. The correction module calls the sub-module to drive the surveying data acquisition module and the data extraction module to survey the new pipeline section and generate a new feature vector set (e.g., {type: "straight pipe", starting coordinates: (X1, Y1), ending coordinates: (X2, Y2), diameter: 0.6m}).

[0090] The retrieval submodule locates the region that overlaps with the original pipeline model space using spatial indexing. The update submodule calculates the deviation between the new and old feature vectors. The comparison revealed that while the spatial coordinates coincided, the diameter changed from 0.5m to 0.6m, indicating a deviation. If the threshold is exceeded, the system determines that a parameter modification has occurred.

[0091] Subsequently, the twin modeling module is invoked only at this location to regenerate the pipe model based on the new diameter of 0.6m, and the connection relationships in the global topology association table are refreshed. The old model state is automatically saved as a historical version.

[0092] The new model inherits the original spatiotemporal entity identification code. The association module automatically collects weld flaw detection reports after the installation of new pipelines, new pressure sensor data, etc., into the spatiotemporal event book under this identification code, forming a continuous and complete asset file.

[0093] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A twin modeling system for underground pipe networks based on autonomous surveying data, characterized in that, It includes a surveying and mapping data acquisition module, a data extraction module, a twin modeling module, a correction module, and a correlation module; The mapping data acquisition module can collect and synchronize the pipeline cross-section contour point cloud sequence and mileage data according to the in-pipe crawler, and form the original three-dimensional point cloud sequence data inside the pipeline with spatiotemporal labels through real-time correction and fusion. The data extraction module can perform least-squares circle fitting on each frame of the original 3D point cloud to extract the basic cross-sectional features. These features are then arranged according to the acquisition trajectory to form a continuous feature sequence. The similarity between the actual directional changes of the feature sequence and the standard pipe fitting template is then calculated. The pipe fitting types are determined and feature vectors are generated to form a preliminary pipe network topology. Similarity The calculation formula is as follows: in, For the actual sequence number The vector and the standard pipe fitting template sequence The distance between vectors For all regular paths in the distance matrix; The twin modeling module constructs a parametric twin model based on the preliminary pipeline topology network and a pre-set parametric parts library; The correction module can establish an incremental update mechanism based on data comparison, quickly locate the changed area using spatial index, determine the change type by combining the difference analysis algorithm, and dynamically correct the parameterized twin model. The association module is used to assign a globally unique spatiotemporal entity identifier to each parametric part in the parametric twin model. The spatiotemporal entity identifier is permanently bound to the part's record in the global topology association table and its source feature vector.

2. The underground pipeline network twin modeling system based on autonomous surveying data according to claim 1, characterized in that, The real-time correction and fusion process synchronizes the line laser scanner, inertial measurement unit, and odometer using the same clock source. Inside the pipeline, the pose is updated recursively using data collected by the inertial measurement unit as the main source of information, through the inertial navigation equations; at the same time, the displacement data from the odometer is used as a local observation for real-time correction.

3. The underground pipeline network twin modeling system based on autonomous surveying data according to claim 1, characterized in that, The specific contents of the data extraction module are as follows: 1) Extraction of basic cross-sectional features; 2) Arrange the basic features in trajectory order; 3) Generate pipeline feature vectors.

4. The underground pipeline network twin modeling system based on autonomous surveying data according to claim 3, characterized in that, The calculation formulas for the center coordinates, diameter, and roundness deviation values ​​of the fitted circle based on the cross-sectional basic features are as follows: in,( , ( ) represents the coordinates of the center of the circle on the two-dimensional fitting plane. To fit the radius and diameter of the circle , This represents the roundness deviation value due to pipe deformation. C and D are both parameter values.

5. The underground pipeline network twin modeling system based on autonomous surveying data according to claim 1, characterized in that, The twin modeling module includes a matching submodule and a connection submodule; The matching submodule can retrieve the corresponding geometric generation function from the parameterized parts library based on the feature vectors in the preliminary pipeline topology network, and call the modeling engine to generate a single part representation model. The connection submodule can determine whether the interface coordinates of two parts coincide within the tolerance range based on the coordinates of the two ends in the representation model of adjacent individual parts; if they coincide, the interfaces of adjacent parts are automatically aligned and matched according to the preliminary pipeline topology network.

6. The underground pipeline network twin modeling system based on autonomous surveying data according to claim 1, characterized in that, The correction module includes a retrieval submodule, a calling submodule, and an update submodule; The retrieval submodule is constructed based on the octree spatial partitioning algorithm, which can build a spatial index of the preliminary pipeline topology network and all the feature vectors it contains. The calling submodule can call the surveying data acquisition module and the data extraction module to perform a new round of operations on the pipeline area, generating a new feature vector set describing the changed area and a new preliminary pipeline topology network.

7. The underground pipeline network twin modeling system based on autonomous surveying data according to claim 6, characterized in that, The update submodule uses spatial indexing to quickly locate the set of pipeline components that overlap with the spatial range of the new feature vector set. It then compares each new feature vector with the corresponding feature parameters of the old model within the overlapping area, calculating the degree of deviation between the new and old feature vectors on key parameters. The calculation formula is as follows: in, The diameter of the pipe in the new feature vector. The pipe diameter in the original parametric twin model, The coordinates of the center of the circle in the new feature vector. The coordinates of the center of the circle in the original parametric twin model. and All are weighting coefficients.

8. The underground pipeline network twin modeling system based on autonomous surveying data according to claim 7, characterized in that, If the degree of deviation If the value exceeds the set threshold and the spatial coordinate ranges corresponding to the old and new feature vectors overlap, it is determined that the pipeline network parameters have been modified; if an old feature vector exists within a specific spatial coordinate range, but after spatial index retrieval, no new vector with matching spatial coordinates can be found in the new feature vector set, it is determined that the pipeline network parameters have been deleted.

Images