Low-voltage power grid data automatic modeling method and system based on visual acquisition and recording

Abstract

The invention provides a low-voltage power grid data automatic modeling method and system based on visual acquisition and recording, and relates to the technical field of low-voltage power grids, and the method comprises the steps: dividing a plurality of sub-regions of a low-voltage power grid region, and constructing a plurality of local coordinate systems; analyzing the to-be-collected and recorded power grid equipment, and determining at least one matched local coordinate system; configuring a visual acquisition and recording task template, and performing data acquisition and recording to obtain at least one group of equipment modeling data; and constructing a multi-region coordinate system conversion model, performing coordinate system conversion, obtaining equipment modeling conversion data corresponding to the at least one group of equipment modeling data, and generating a visual modeling simulation result of the to-be-collected and recorded task. The technical problems that in the low-voltage power grid data modeling process in the prior art, in general, a large unified global modeling coordinate system is established to conduct data collection or modeling on all power grid equipment, although the unified effect can be achieved, the data collection or modeling is not accurate enough, and the real reflection and the application effect of power grid data are affected are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of low-voltage power grids, and in particular to a method and system for automatically modeling low-voltage power grid data based on visual data collection and recording. Background Art

[0002] Power grid data modeling is used to support various tasks such as power grid operation monitoring, fault diagnosis, planning and design. Accurate equipment modeling and data collection are particularly important in the construction of smart grids. However, in the current low-voltage power grid data modeling process, there are significant problems of "field-to-field separation" and "inconsistent operating standards among personnel." The current existing technology is to establish a large unified global modeling coordinate system to collect data or model all power grid equipment. However, due to the large scale of the power grid and the different responsibilities of each technician, the objects collected are often collected for a single object. Although the collection of a single object under a large unified coordinate system can achieve a unified effect, the accuracy of the modeling data collected in this case may not be accurate enough, affecting the true reflection and application effect of the power grid data. Summary of the Invention

[0003] The present application provides a method and system for automatic modeling of low-voltage power grid data based on visual collection and recording, aiming to solve the technical problem in the existing technology that in the process of low-voltage power grid data modeling, a large unified global modeling coordinate system is usually established to collect data or model all power grid equipment. Although the established unified global system can achieve a unified effect, it is not accurate enough, which affects the true reflection of power grid data and the application effect.

[0004] The first aspect disclosed in the present application provides a method for automatic modeling of low-voltage power grid data based on visual recording, the method comprising: dividing the low-voltage power grid area into multiple sub-areas, and constructing multiple local coordinate systems corresponding to the multiple sub-areas; parsing the power grid equipment to be recorded of the first user's task to be recorded, and determining at least one matching local coordinate system from the multiple local coordinate systems according to the type of the power grid equipment to be recorded; configuring a visual recording task template, the visual recording task template performs data recording on the power grid equipment to be recorded of the task to be recorded according to the at least one matching local coordinate system, and obtains at least one set of equipment modeling data; constructing a multi-area coordinate system conversion model based on the multiple local coordinate systems, using the multi-area coordinate system conversion model to perform coordinate system conversion on the at least one matching local coordinate system, and obtain equipment modeling conversion data corresponding to at least one set of equipment modeling data, and generating a visual modeling simulation result of the task to be recorded according to the equipment modeling conversion data.

[0005] The second aspect disclosed in the present application provides a low-voltage power grid data automatic modeling system based on visual recording, which is used for the above-mentioned low-voltage power grid data automatic modeling method based on visual recording. The system includes: a sub-area division module, which is used to divide the low-voltage power grid area into multiple sub-areas and construct multiple local coordinate systems corresponding to the multiple sub-areas; a coordinate system determination module, which is used to parse the power grid equipment to be recorded of the first user's task to be recorded, and determine at least one matching local coordinate system from the multiple local coordinate systems according to the type of the power grid equipment to be recorded; a data recording module, which is used to configure a visual recording task template, and the visual recording task template collects data from the power grid equipment to be recorded of the task to be recorded according to the at least one matching local coordinate system to obtain at least one set of equipment modeling data; a simulation result generation module, which is used to construct a multi-area coordinate system conversion model based on the multiple local coordinate systems, use the multi-area coordinate system conversion model to perform coordinate system conversion on the at least one matching local coordinate system, obtain equipment modeling conversion data corresponding to at least one set of equipment modeling data, and generate a visual modeling simulation result of the task to be recorded according to the equipment modeling conversion data.

[0006] One or more technical solutions provided in this application have at least the following beneficial effects: The traditional method regards the entire power grid as a single global coordinate system, which leads to reduced positioning accuracy of equipment in a large area. Especially in large-scale power grids, the spatial characteristics and position differences of different equipment may lead to coordinate deviation. By dividing into multiple sub-areas and establishing a local coordinate system for each sub-area, different equipment and areas can be modeled separately, which improves the positioning accuracy of equipment in each sub-area. Due to the differences in equipment type, function and location, it is difficult to accurately meet the specific needs of different equipment using a unified coordinate system for the collection task of each device. According to the equipment type and regional characteristics, the most matching local coordinate system is selected from multiple local coordinate systems for each device to be recorded, ensuring that the coordinate system is highly consistent with the characteristics of the equipment during the data collection process. Due to the inconsistent operating standards of personnel, different collection methods may lead to data quality Inconsistency affects the final modeling effect. By configuring the visual recording task template, the collection path, viewing angle, collection distance, etc. of each device are standardized. Templated recording can ensure that each device follows a unified standard when collecting, thereby improving the standardization and consistency of data collection; the conversion between local coordinate systems is prone to errors. By establishing a multi-region coordinate system conversion model, the conversion relationship between each local coordinate system and the global coordinate system is clearly defined. This ensures that the data in each local coordinate system can be seamlessly and accurately converted to the global coordinate system, eliminating the error between the local and global. This not only improves the uniformity and consistency between data in different regions, but also enables the device modeling between regions to be compared and analyzed under the same reference framework, thereby improving the accuracy and reliability of the entire modeling system.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A flow chart of a method for automatic modeling of low-voltage power grid data based on visual recording provided in an embodiment of the present application.

[0009] Figure 2 A schematic diagram of the structure of a low-voltage power grid data automatic modeling system based on visual recording provided in an embodiment of the present application.

[0010] Description of the accompanying drawings: sub-area division module 10, coordinate system determination module 20, data collection and recording module 30, simulation result generation module 40. DETAILED DESCRIPTION

[0011] The embodiments of the present application provide a method and system for automatic modeling of low-voltage power grid data based on visual collection and recording, which solves the technical problem in the prior art that in the process of low-voltage power grid data modeling, a large unified global modeling coordinate system is usually established to collect data or model all power grid equipment. Although the established unified global system can achieve a unified effect, it is not accurate enough, which affects the true reflection of power grid data and the application effect.

[0012] After introducing the basic principles of this application, various non-limiting embodiments of this application will be specifically described below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to explain this application and are not used to limit this application.

[0013] Example 1, as Figure 1 As shown, an embodiment of the present application provides a method for automatically modeling low-voltage power grid data based on visual acquisition and recording, the method comprising: The low-voltage power grid area is divided into multiple sub-areas, and multiple local coordinate systems corresponding to the multiple sub-areas are constructed.

[0014] Information on all devices within the low-voltage power grid area, including transformers, circuit breakers, and distribution boxes, is obtained. For each device, basic attributes such as type, volume, spatial location, and electrical connections are recorded. This basic attribute information forms the basis for subsequent region division and construction of a local coordinate system. Feature vectors are constructed based on each device's basic attribute information. After normalization, these feature vectors are used to calculate the similarity between devices using methods such as cosine similarity. Devices with high similarity are clustered together, and each device cluster corresponds to a sub-region. Based on the clustering results, the low-voltage power grid area is divided into multiple sub-regions, each with similar device types and distribution characteristics. This allows for more accurate construction of the local coordinate system. Principal component analysis is used to analyze the spatial coordinates of the devices in each sub-region, extracting the principal axis direction. Once the principal axis direction is determined, this direction is selected as the X-axis, and the Y and Z axes are determined using orthogonal rules to form a three-dimensional coordinate system. The center point of each sub-region serves as the origin of the coordinate system. Ultimately, each sub-region has a corresponding local coordinate system, enabling precise data collection for each device and improving modeling accuracy.

[0015] The power grid equipment to be recorded of the task to be recorded of the first user is parsed, and at least one matching local coordinate system is determined from the multiple local coordinate systems according to the type of the power grid equipment to be recorded.

[0016] When the first user submits a task to be recorded, the content of the task to be recorded is parsed, and the power grid equipment to be recorded that needs to be recorded in the task is determined. According to the type, installation location, adjacent equipment and other information of the equipment to be recorded, matching is performed from multiple established local coordinate systems. For example, if a device is located in a sub-area, it needs to use the local coordinate system of the sub-area. If the device spans multiple sub-areas or is adjacent to multiple sub-areas, the local coordinate system related to the device is selected for matching, and finally at least one matching local coordinate system is determined.

[0017] A visual data collection task template is configured, and the visual data collection task template collects data from the power grid equipment to be collected in the task to be collected according to the at least one matching local coordinate system to obtain at least one set of equipment modeling data.

[0018] The visual recording task template is configured based on the matched local coordinate system. It ensures that the device collects data in the correct coordinate system. The template includes a series of collection parameters that affect the collection method and quality of the device, including the collection path, viewing angle parameters, image clarity, collection distance range, point cloud density threshold, and coverage angle range. After the visual recording task template is configured, it is automatically distributed to the collection device. The collection device executes the task according to the visual recording task template, generating at least one set of device modeling data.

[0019] A multi-region coordinate system conversion model is constructed based on the multiple local coordinate systems, and the multi-region coordinate system conversion model is used to perform coordinate system conversion on at least one matching local coordinate system to obtain device modeling conversion data corresponding to at least one set of device modeling data, and a visual modeling simulation result of the task to be recorded is generated according to the device modeling conversion data.

[0020] In the low-voltage power grid area, each sub-area has its own independent local coordinate system. Different local coordinate systems need to be converted into a unified global coordinate system. To achieve this goal, a multi-area coordinate system conversion model is first constructed based on multiple local coordinate systems. The multi-area coordinate system conversion model includes the conversion relationship between multiple local coordinate systems.

[0021] The key parameters of each local coordinate system include the three-dimensional coordinates of the origin in the global coordinate system, the direction vectors of the coordinate axes (i.e., the directions of the X, Y, and Z axes), the units of the local coordinate system, and the angle definition method. Based on the key parameters of the local coordinate system, a transformation formula is used to construct a transformation matrix from the local coordinate system to the global coordinate system. The transformation matrix transforms the coordinates of the local coordinate system into the coordinates of the global coordinate system. The transformation matrices of multiple local coordinate systems are indexed and stored according to the area to which the device belongs, so that they can be quickly retrieved and applied later.

[0022] When at least one set of collected device modeling data is in different local coordinate systems, coordinate transformation is performed on at least one set of device modeling data according to the coordinate system transformation model, and the data in different local coordinate systems is transformed into the global coordinate system. After the transformation is completed, device modeling transformation data corresponding to at least one set of device modeling data is obtained, and the device modeling transformation data includes information such as the position, posture, and size of the device in the global coordinate system.

[0023] Based on the resulting device modeling conversion data, a modeling engine generates three-dimensional visualization models of power grid equipment. These models can include electrical equipment, lines, transformers, and more, showcasing the actual layout of the power grid. All equipment is placed in a unified virtual space and arranged according to the actual layout, showcasing the electrical connections, positions, and structures between devices. This process generates a power grid simulation scene that reflects the physical relationships and operating conditions of the power grid equipment. The results of the visual modeling simulation are presented through an interactive simulation module, which allows users to interact with the simulation scene, including rotating, panning, zooming, and even highlighting specific devices or areas. Users can adjust the perspective and view as needed to obtain detailed information about the device or the power grid. This interactive process provides an immersive experience for users, helping them gain a deeper understanding of all aspects of the power grid.

[0024] Furthermore, the method for dividing the low-voltage power grid area into multiple sub-areas includes: Obtain the equipment set of the low-voltage power grid area and basic attribute information of each equipment, wherein the basic attribute information includes equipment type, equipment volume calibration interval, equipment spatial position and electrical connection relationship; extract the recording similarity feature vector according to the basic attribute information; perform cluster analysis on the equipment set of the low-voltage power grid area according to the recording similarity feature vector to obtain clustering results, wherein the clustering results include multiple equipment clustering clusters, and use the multiple equipment clustering clusters to obtain multiple sub-areas of the low-voltage power grid area.

[0025] Low-voltage power grids contain a wide variety of equipment, including transformers, circuit breakers, and distribution boxes. All equipment within these areas is collected and categorized to form a device collection. For each device, key basic attribute information includes: device type, which indicates the type of device; device volume calibration interval, which indicates the volume range of each device and is used for subsequent spatial positioning; device spatial location, which indicates the three-dimensional position coordinates (such as X, Y, and Z coordinates) of the device within the power grid area; and electrical connection relationships, which indicate the electrical connection information of each device and describe the electrical connection relationships between devices, such as the connection between a transformer and a switch, or between a distribution box and a cable.

[0026] When modeling on-site, different types of equipment have very different collection methods and requirements. In order to effectively adapt to the on-site modeling and recording habits of different types of equipment, the recording similarity feature vector is extracted based on the basic attribute information. This helps with subsequent equipment classification and cluster analysis, ensuring the effectiveness of the data collection task.

[0027] The devices in the low-voltage power grid area are clustered according to their recording similarity feature vectors. For example, K-means clustering is used to calculate the similarity between devices and divide the devices into a predetermined number of clusters. The devices in each cluster have similar attribute characteristics and collection requirements. After the cluster analysis is completed, each cluster represents a sub-area. The devices in each sub-area will be modeled and data collected accordingly based on their characteristics. The devices in each sub-area have similar collection requirements, which can improve the accuracy and efficiency of data collection.

[0028] Furthermore, the method for extracting the recording similarity feature vector according to the basic attribute information includes: An initialized feature vector dimensional space is defined, wherein the initialized feature vector dimensional space includes feature items and feature values ​​corresponding to each feature item, wherein the feature items include device type code, device volume level, average distance to adjacent devices, electrical connection topology depth, and installation direction classification; the feature value corresponding to each feature item is normalized to obtain a normalized feature vector dimensional space, and the feature vectors in the normalized feature vector dimensional space are similarly calculated using cosine similarity to extract the recording similarity feature vector.

[0029] Define the initialized feature vector dimensional space, in which the feature items can effectively distinguish the differences between devices and reflect the collection requirements of the devices. The device type code is a unique code representation of each device type, and integers or strings are usually used to identify different types of devices; the device volume level is the volume range or size level of the device. According to the actual volume of the device, the device can be divided into multiple levels, such as small devices, large devices, etc.; the average distance to adjacent devices is the average value of the spatial distance between the device and its adjacent devices. This feature item reflects the layout density of the device in the power grid and affects the complexity of data collection; the electrical connection topology depth is the depth of the electrical connection relationship between the device and other devices. For example, device A is connected to device B, and device B is connected to device C. Such connection depth can be expressed in numbers. The greater the depth, the more complex the connection relationship between the devices; the installation direction is classified as the installation method of the device, such as vertical, horizontal, inclined, etc., which affects the viewing angle requirements of the collection task.

[0030] Since the numerical ranges and dimensions of each feature item are different, for example, the device type is a discrete integer, while the device volume is a continuous large-range value. The different scales of these feature items will affect the results of the similarity calculation. In order to avoid certain feature items from having too much influence on the similarity calculation, the feature items need to be normalized. For example, the minimum and maximum normalization method is used to convert all feature values ​​into a unified interval range, usually the [0,1] interval. For discrete feature items, such as device type code, installation direction, etc., one-hot encoding can be used or they can be directly mapped to discrete integer values ​​to avoid their influence on the similarity calculation.

[0031] Through normalization, the feature vector of each device is converted into a standardized vector, resulting in a normalized feature vector dimensional space. This makes the feature vectors of all devices consistent and comparable on the same scale. Cosine similarity is used as a measure of the similarity between two devices. Cosine similarity ranges from -1 to 1, with values ​​closer to 1 indicating more similarity and -1 indicating greater dissimilarity. By calculating cosine similarity for the feature vectors of all devices, we can obtain the similarity feature vector between each device and every other device.

[0032] Furthermore, the method of constructing a plurality of local coordinate systems corresponding to the plurality of sub-regions includes: Calculate the boundary envelopes of the multiple sub-areas to obtain a set of spatial boundary coordinates and spatial center coordinates; use principal component analysis to fit the device distribution coordinates of each area in the multiple sub-areas to extract the first principal axis direction, use the first principal axis direction as the X-axis and use the three-dimensional orthogonal rule to define the Y-axis and Z-axis, and output the three-dimensional coordinate axis; use the spatial center coordinates as the origin of the local coordinate system and the three-dimensional coordinate axes as the coordinate axes of the local coordinate system to construct multiple local coordinate systems corresponding to the multiple sub-areas.

[0033] For each sub-region, the spatial positions of all devices within the sub-region are first obtained. Spatial positions are typically 3D coordinates. Then, the bounding box of the sub-region is calculated using these device coordinates. The bounding box is a minimal rectangular box that completely encloses the spatial distribution of all devices. In three-dimensional space, the bounding box contains the minimum and maximum coordinates of the devices in the x, y, and z directions, forming a rectangular parallelepiped. After obtaining the bounding box, the spatial boundary coordinate set and spatial center coordinates of the sub-region are calculated. The spatial boundary coordinate set is the set of coordinate points on the bounding box. The spatial center coordinate is the center point of the bounding box and is determined by calculating the median value between the minimum and maximum coordinates in the x, y, and z directions.

[0034] Principal component analysis is a technique that maps high-dimensional data to low-dimensional space. It can help find the most important direction in the data. In the spatial distribution of power grid equipment, principal component analysis is used to find the most representative direction in the equipment layout, that is, the first principal axis direction. Specifically, the distribution coordinates of the equipment in space are analyzed by the principal component analysis algorithm, the variance of the equipment coordinates is maximized, and the principal components in the direction are extracted. That is, for all coordinate points of the equipment distribution, a covariance matrix is ​​constructed and eigenvalue decomposition is performed. The eigenvector represents the main direction of the equipment distribution, and the size of the eigenvalue represents the variance in this direction. The eigenvector corresponding to the largest eigenvalue is the first principal axis direction. The first principal axis direction is a three-dimensional vector, which means that the distribution of equipment in this direction is the largest. This direction is used as the X-axis direction of the local coordinate system. Next, we define the Y and Z axes by orthogonalizing them. We can use the cross product to calculate the other two directions perpendicular to the first principal axis. For example, if the first principal axis is the X axis, we can choose an arbitrary direction, such as a random point on the device coordinate axis, and then calculate the cross product with the X axis to get the Z axis direction. We can then use the cross product to calculate the Y axis direction. By orthogonalizing in this way, we ensure that the X, Y, and Z axes are perpendicular to each other.

[0035] The spatial center coordinate is set as the origin of the local coordinate system. This coordinate is the center point calculated from the envelope of the device distribution. This center point serves as the zero point of the local coordinate system. The obtained three-dimensional coordinate axes are used as the coordinate axes of the local coordinate system, corresponding to the X, Y, and Z directions respectively. The spatial position of the device (i.e., the original device coordinates) are converted to coordinates relative to the local coordinate system. This conversion is completed by setting the origin (space center) and the rotation matrix (the coordinate axis direction calculated by principal component analysis). For each sub-area, a corresponding local coordinate system is created and the device data in the area is converted to the local coordinate system. This ensures that the device data of different sub-areas are modeled under a unified standard and avoids coordinate deviations between different areas.

[0036] Furthermore, a visual recording task template is configured, and the visual recording task template includes configuring the recording path, viewing angle parameters, image clarity, recording distance interval, point cloud density threshold and coverage angle range.

[0037] Configuring the recording path sets the path or order of collection during the device collection process. The recording path ensures that the device collects along the optimal path to cover all necessary devices or areas, which can help avoid repeated collection or missed areas, and reduce unnecessary interference during the collection process; the viewing angle parameter sets the appropriate viewing angle for each device collection. By adjusting the viewing angle parameters of the camera or sensor, such as pitch angle, horizontal angle, lens focal length, etc., it can ensure that comprehensive device information is obtained from different angles and directions, improving the accuracy and completeness of data collection; image clarity is set according to the type of device and environmental requirements, including image pixel quality, contrast, brightness, etc., to ensure that the collected image information is clear and unambiguous, which is helpful for subsequent modeling and analysis; the recording distance range is based on the physical size and spatial layout of the device, and the collection device is configured. The recording distance and recording distance interval ensure that the device can effectively obtain clear and accurate data from a predetermined distance range. Collection distances that are too close or too far may affect data quality. The point cloud density threshold is the point cloud density threshold set according to the needs of the device during the point cloud data collection process. The point cloud density affects the details and accuracy of the collection. Too low a density will result in sparse data and affect the modeling effect, while too high a density may lead to an excessive processing burden. The point cloud density can be dynamically adjusted according to the specific device or collection environment to achieve a balance between data accuracy and processing efficiency. The coverage angle range is the coverage angle of the camera or sensor set according to the layout of the device and the required field of view. This ensures that each device can cover all required areas during the collection process, especially in complex spatial structures, to avoid collection blind spots. These configurations make the collection process more standardized, avoid errors caused by human differences, improve the accuracy of data modeling, and ensure that the final visualization results are clear, comprehensive, and precise.

[0038] Furthermore, the visual data collection task template collects data from the power grid devices to be collected in the task to be collected according to the at least one matching local coordinate system, and the method includes: The visual recording task template obtains at least one recording task instruction set corresponding to the at least one matching local coordinate system; binds the at least one matching local coordinate system with the corresponding at least one recording task instruction set and sends it to the recording terminal device of the first user, and the recording terminal device records data on the power grid equipment to be recorded for the task to be recorded, and obtains at least one set of equipment modeling data.

[0039] For each matching local coordinate system, the matching recording task instruction set is extracted from the visual recording task template. The recording task instruction set includes the rules and parameters that need to be followed in the visual recording task template, including configuring the recording path, viewing angle parameters, image clarity, recording distance interval, point cloud density threshold, and coverage angle range. For example, if a local coordinate system corresponds to the equipment distribution of a distribution transformer, then its task instruction set contains specific collection requirements for that type of equipment.

[0040] Each device to be recorded and its corresponding matching local coordinate system are bound to the recording task instruction set. This binding process ensures that the collection process is carried out according to the specific coordinate system of the area where the device is located, avoiding coordinate system confusion or data deviation. After binding the matching local coordinate system and the recording task instruction set, this information is sent to the first user's recording terminal device. The recording terminal device can be a laser scanner, 3D camera, drone, or other hardware capable of performing data collection tasks. When performing data collection tasks, the recording terminal device strictly follows the recording task instruction set. For example, it scans the device at a specified viewing angle to ensure that the image resolution, recording distance, and point cloud density requirements are met during the scanning process. The recorded data will be transmitted back to the system in real time to form device modeling data. After post-processing, it will be converted into model data for further analysis and use.

[0041] Furthermore, a multi-region coordinate system conversion model is constructed based on the multiple local coordinate systems, and the method includes: Obtain the key parameters of each local coordinate system, which include the three-dimensional coordinates of the origin in the global coordinate system, the coordinate axis direction vector, the local coordinate unit and the angle definition method; construct a local-global coordinate conversion matrix based on the key parameters of each local coordinate system and the transformation formula of the pre-constructed global coordinate system, and use the local-global coordinate conversion matrix to store it according to the corresponding sub-region index to establish a multi-region coordinate system conversion model.

[0042] The three-dimensional coordinates of the origin in the global coordinate system represent the position of the local coordinate system in the global coordinate system. By determining the origin's location, the local coordinate system can be aligned with the global coordinate system. The three-dimensional coordinates of the origin are typically obtained through geolocation or measurement and represent the reference position of the local coordinate system. The coordinate axis direction vectors define the directions of the three axes (X, Y, and Z) of the local coordinate system. These vectors reflect the spatial orientation of the local coordinate system and are typically determined by the physical location, installation angle, or other geometric characteristics of the device. These direction vectors can be used to determine the rotation and spatial orientation of the local coordinate system relative to the global coordinate system. The local coordinate unit refers to the length of the unit in the coordinate system. For example, if the local coordinate system is in meters, each coordinate value represents the actual spatial position in the local coordinate system, which may differ from the global coordinate system and therefore requires conversion. The angle definition method describes how to express the angular relationship between coordinate axes. In different coordinate systems, angles may use different units (degrees or radians) and standards (for example, clockwise or counterclockwise). A unified angle definition method is required to ensure consistent coordinate conversion.

[0043] Based on the acquired key parameters of the local coordinate system, the first step is to determine how to transform the coordinates of the local coordinate system into the global coordinate system. By analyzing the origin coordinates, direction vectors, and angle definitions, the translation, rotation, and scaling of the local coordinate system relative to the global coordinate system can be inferred. The global coordinate system transformation formula has been pre-established and contains the mapping rules from the local coordinate system to the global coordinate system. This formula contains the relationship between all sub-region coordinate systems, including rotation matrices and translation matrices, ensuring seamless integration of each local coordinate system with the global coordinate system. Using the key parameters of the local coordinate system and the transformation formula of the global coordinate system, the local-to-global coordinate transformation matrix is ​​calculated. This matrix contains all the information required for coordinate transformation, ultimately achieving accurate mapping from the local coordinate system to the global coordinate system. The calculated local-to-global coordinate transformation matrix is ​​stored according to the sub-region index, forming a multi-region coordinate system transformation model. This model ensures that all local coordinate systems can be effectively transformed when needed based on the sub-region index, thereby achieving accurate cross-region data integration in large-scale power grid modeling.

[0044] Furthermore, the method of performing coordinate system transformation on the at least one matching local coordinate system using the multi-region coordinate system transformation model includes: The multi-region coordinate system conversion model is used to obtain the conversion key parameters of the at least one matching local coordinate system; the initial device modeling conversion data corresponding to at least one set of device modeling data is obtained according to the conversion key parameters; the cross-region anchor device is identified, and the cross-region anchor device is a public device located at the intersection of two or more sub-regions; the spatial deviation vector of the initial device modeling conversion data corresponding to the at least one set of device modeling data is calculated respectively through the cross-region anchor device, including a rotation deviation vector and a translation deviation vector; the initial device modeling conversion data is realigned according to the spatial deviation vector to obtain updated device modeling conversion data.

[0045] According to the identifier or index of at least one matching local coordinate system, the corresponding transformation key parameters are extracted from the multi-region coordinate system transformation model, including the origin, coordinate axis direction, rotation matrix, translation vector, etc. of the local coordinate system. These transformation key parameters are the basis for subsequent coordinate transformation.

[0046] The obtained key conversion parameters are applied to the device modeling data to be converted, and coordinate conversion is performed based on the relationship between the local coordinate system and the global coordinate system. The converted data represents the position and characteristics of the device in the global coordinate system. In this way, device data can be processed and integrated across regions to ensure that the modeling data of the entire low-voltage power grid is in the same reference coordinate system. Ultimately, the converted data becomes the initial device modeling conversion data. This data is the device information after coordinate system conversion and can be used for subsequent visual modeling or further analysis.

[0047] Cross-regional anchor devices are common devices that span two or more sub-regions. The location information of these common devices can serve as a reference between multiple local coordinate systems. These are typically key devices in the power grid, such as transformers, switchgear, or electrical connection devices at junctions. By analyzing the relationship between local coordinate systems and local devices, cross-regional anchor devices located at the junction of different sub-regions can be identified. Cross-regional anchor devices often have unique physical locations or electrical connections and can be identified by their spatial location or electrical connection relationships. These cross-regional anchor devices provide a common reference point between coordinate systems in different regions, enabling calibration during coordinate conversion.

[0048] The spatial deviation vector refers to the difference between the position of the device in the global coordinate system and the expected position during the coordinate system conversion process. The rotation deviation vector represents the rotation error of the device coordinate system, that is, the angular difference between the device's local coordinate system and the global coordinate system; the translation deviation vector represents the translation error of the device coordinate system, that is, the position difference of the device in the coordinate axis direction. Cross-region anchor devices are common devices that connect different regional coordinate systems. Therefore, they can be used as positioning references. Specifically, the rotation error is calculated by comparing the direction difference of the cross-region anchor device in the local coordinate system and the global coordinate system. This can be calculated by comparing the coordinate axis direction vectors, or using the rotation matrix to calculate the angular difference between the two coordinate systems; the translation error is calculated by comparing the position difference of the cross-region anchor device in the local coordinate system and the global coordinate system. Generally speaking, this involves calculating the difference in the position coordinates (x, y, z) of the device in space.

[0049] The initial device modeling transformation data is corrected using spatial deviation vectors (including rotation deviation vectors and translation deviation vectors) to more accurately align it with the global coordinate system. Specifically, the rotation deviation vector is used to adjust the device's rotation angle. Based on the calculated rotation error, the device's coordinates are rotated through the rotation matrix to align with the standard of the global coordinate system. For example, if a rotation deviation angle is calculated, the rotation deviation angle can be applied to adjust the device's coordinates. The translation deviation vector is used to adjust the device's position. By applying the translation error to the device's coordinates, the device can be accurately positioned in the correct position in the global coordinate system. For example, the coordinate values ​​of all devices are increased or decreased by the translation deviation vector to ensure that their coordinates are aligned with the global coordinate system. After completing the rotation and translation alignment, the initial device modeling transformation data is updated to the corrected new coordinates, namely the device modeling transformation data, which reflects the precise position of the device in the global coordinate system. After the deviation correction, they will accurately correspond to the correct grid position.

[0050] Furthermore, a visual modeling simulation result of the task to be recorded is generated according to the device modeling conversion data; wherein, the visual modeling simulation result is output through an interactive simulation module, and the interactive simulation module performs a visual display response by receiving the interactive operation of the first user.

[0051] The visual modeling simulation results are generated as a three-dimensional power grid model based on the acquired device modeling conversion data. They display the precise location of devices, their connections, and the overall layout of the power grid. Once generated, the visual modeling simulation results are provided to the first user for display and manipulation via the interactive simulation module. The core of the interactive simulation module is to enable the first user to interact with the power grid model, view detailed information about different areas, devices, or functions, and simulate and adjust them as needed. For example, the first user can use a mouse, touchpad, or other input device to rotate, translate, and zoom the power grid model to view individual devices from different angles. During the interaction, the first user's actions trigger real-time system responses. For example, if the first user drags a device or adjusts its parameters, the system updates the model in real time based on these changes, displaying the changes to the power grid as a result of these actions. Through this interactive simulation, the first user can perform power grid modeling and analysis more flexibly and accurately, improving operational efficiency and reducing human error.

[0052] In summary, the method for automatic modeling of low-voltage power grid data based on visual acquisition and recording provided by the embodiments of the present application has the following technical effects: The traditional method regards the entire power grid as a single global coordinate system, which leads to reduced positioning accuracy of equipment in a large area. Especially in large-scale power grids, the spatial characteristics and position differences of different equipment may lead to coordinate deviation. By dividing into multiple sub-areas and establishing a local coordinate system for each sub-area, different equipment and areas can be modeled separately, which improves the positioning accuracy of equipment in each sub-area. Due to the differences in equipment type, function and location, it is difficult to accurately meet the specific needs of different equipment using a unified coordinate system for the collection task of each device. According to the equipment type and regional characteristics, the most matching local coordinate system is selected from multiple local coordinate systems for each device to be recorded, ensuring that the coordinate system is highly consistent with the characteristics of the equipment during the data collection process. Due to the inconsistent operating standards of personnel, different collection methods may lead to data quality Inconsistency affects the final modeling effect. By configuring the visual recording task template, the collection path, viewing angle, collection distance, etc. of each device are standardized. Templated recording can ensure that each device follows a unified standard when collecting, thereby improving the standardization and consistency of data collection; the conversion between local coordinate systems is prone to errors. By establishing a multi-region coordinate system conversion model, the conversion relationship between each local coordinate system and the global coordinate system is clearly defined. This ensures that the data in each local coordinate system can be seamlessly and accurately converted to the global coordinate system, eliminating the error between the local and global. This not only improves the uniformity and consistency between data in different regions, but also enables the device modeling between regions to be compared and analyzed under the same reference framework, thereby improving the accuracy and reliability of the entire modeling system.

[0053] The second embodiment is based on the same inventive concept as the method for automatically modeling low-voltage power grid data based on visual acquisition and recording in the above embodiment. Figure 2 As shown, an embodiment of the present application provides a low-voltage power grid data automatic modeling system based on visual collection and recording, the system comprising: The sub-area division module 10 is used to divide the low-voltage power grid area into multiple sub-areas and construct multiple local coordinate systems corresponding to the multiple sub-areas; the coordinate system determination module 20 is used to parse the power grid equipment to be recorded of the first user's task to be recorded, and determine at least one matching local coordinate system from the multiple local coordinate systems according to the type of the power grid equipment to be recorded; the data collection module 30 is used to configure a visual collection task template, and the visual collection task template collects data from the power grid equipment to be recorded of the task to be recorded according to the at least one matching local coordinate system to obtain at least one set of device modeling data; the simulation result generation module 40 is used to construct a multi-area coordinate system conversion model based on the multiple local coordinate systems, use the multi-area coordinate system conversion model to perform coordinate system conversion on the at least one matching local coordinate system, obtain device modeling conversion data corresponding to at least one set of device modeling data, and generate a visual modeling simulation result of the task to be recorded according to the device modeling conversion data.

[0054] Furthermore, the sub-region division module 10 is configured to perform the following steps: Obtain the equipment set of the low-voltage power grid area and basic attribute information of each equipment, wherein the basic attribute information includes equipment type, equipment volume calibration interval, equipment spatial position and electrical connection relationship; extract the recording similarity feature vector according to the basic attribute information; perform cluster analysis on the equipment set of the low-voltage power grid area according to the recording similarity feature vector to obtain clustering results, wherein the clustering results include multiple equipment clustering clusters, and use the multiple equipment clustering clusters to obtain multiple sub-areas of the low-voltage power grid area.

[0055] Furthermore, the sub-region division module 10 is configured to perform the following steps: An initialized feature vector dimensional space is defined, wherein the initialized feature vector dimensional space includes feature items and feature values ​​corresponding to each feature item, wherein the feature items include device type code, device volume level, average distance to adjacent devices, electrical connection topology depth, and installation direction classification; the feature value corresponding to each feature item is normalized to obtain a normalized feature vector dimensional space, and the feature vectors in the normalized feature vector dimensional space are similarly calculated using cosine similarity to extract the recording similarity feature vector.

[0056] Furthermore, the sub-region division module 10 is configured to perform the following steps: Calculate the boundary envelopes of the multiple sub-areas to obtain a set of spatial boundary coordinates and spatial center coordinates; use principal component analysis to fit the device distribution coordinates of each area in the multiple sub-areas to extract the first principal axis direction, use the first principal axis direction as the X-axis and use the three-dimensional orthogonal rule to define the Y-axis and Z-axis, and output the three-dimensional coordinate axis; use the spatial center coordinates as the origin of the local coordinate system and the three-dimensional coordinate axes as the coordinate axes of the local coordinate system to construct multiple local coordinate systems corresponding to the multiple sub-areas.

[0057] Furthermore, a visual recording task template is configured, and the visual recording task template includes configuring the recording path, viewing angle parameters, image clarity, recording distance interval, point cloud density threshold and coverage angle range.

[0058] Furthermore, the data collection and recording module 30 is configured to perform the following steps: The visual recording task template obtains at least one recording task instruction set corresponding to the at least one matching local coordinate system; binds the at least one matching local coordinate system with the corresponding at least one recording task instruction set and sends it to the recording terminal device of the first user, and the recording terminal device records data on the power grid equipment to be recorded for the task to be recorded, and obtains at least one set of equipment modeling data.

[0059] Furthermore, the simulation result generating module 40 is configured to perform the following steps: Obtain the key parameters of each local coordinate system, which include the three-dimensional coordinates of the origin in the global coordinate system, the coordinate axis direction vector, the local coordinate unit and the angle definition method; construct a local-global coordinate conversion matrix based on the key parameters of each local coordinate system and the transformation formula of the pre-constructed global coordinate system, and use the local-global coordinate conversion matrix to store it according to the corresponding sub-region index to establish a multi-region coordinate system conversion model.

[0060] Furthermore, the simulation result generating module 40 is configured to perform the following steps: The multi-region coordinate system conversion model is used to obtain the conversion key parameters of the at least one matching local coordinate system; the initial device modeling conversion data corresponding to at least one set of device modeling data is obtained according to the conversion key parameters; the cross-region anchor device is identified, and the cross-region anchor device is a public device located at the intersection of two or more sub-regions; the spatial deviation vector of the initial device modeling conversion data corresponding to the at least one set of device modeling data is calculated respectively through the cross-region anchor device, including a rotation deviation vector and a translation deviation vector; the initial device modeling conversion data is realigned according to the spatial deviation vector to obtain updated device modeling conversion data.

[0061] Furthermore, a visual modeling simulation result of the task to be recorded is generated according to the device modeling conversion data; wherein, the visual modeling simulation result is output through an interactive simulation module, and the interactive simulation module performs a visual display response by receiving the interactive operation of the first user.

[0062] Through the above detailed description of the method for automatic modeling of low-voltage power grid data based on visual recording in this specification, those skilled in the art can clearly understand the automatic modeling system for low-voltage power grid data based on visual recording in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0063] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for automatically modeling low-voltage power grid data based on visual data collection and recording, characterized in that: The method comprises: Dividing the low-voltage power grid area into multiple sub-areas and constructing multiple local coordinate systems corresponding to the multiple sub-areas; Analyzing the power grid equipment to be recorded of the task to be recorded of the first user, and determining at least one matching local coordinate system from the multiple local coordinate systems according to the type of the power grid equipment to be recorded; Configuring a visual data collection task template, wherein the visual data collection task template collects data from the power grid devices to be collected in the data collection task according to the at least one matching local coordinate system, and obtains at least one set of device modeling data; A multi-region coordinate system conversion model is constructed based on the multiple local coordinate systems, and the multi-region coordinate system conversion model is used to perform coordinate system conversion on at least one matching local coordinate system to obtain device modeling conversion data corresponding to at least one set of device modeling data, and a visual modeling simulation result of the task to be recorded is generated according to the device modeling conversion data.

2. The method for automatic modeling of low-voltage power grid data based on visual acquisition and recording according to claim 1, characterized in that: Dividing the low-voltage power grid area into multiple sub-areas includes: Obtaining a device set in the low-voltage power grid area and basic attribute information of each device, wherein the basic attribute information includes device type, device volume calibration interval, device spatial location, and electrical connection relationship; Extracting recording similarity feature vectors according to the basic attribute information; A cluster analysis is performed on the equipment set in the low-voltage power grid area according to the recorded similarity feature vector to obtain a clustering result, wherein the clustering result includes multiple equipment clusters, and multiple sub-areas of the low-voltage power grid area are obtained using the multiple equipment clusters.

3. The method for automatic modeling of low-voltage power grid data based on visual acquisition and recording according to claim 2, characterized in that: The method for extracting the recording similarity feature vector according to the basic attribute information includes: Defining an initialized feature vector dimensional space, the initialized feature vector dimensional space including feature items and feature values ​​corresponding to each feature item, wherein the feature items include device type code, device volume level, average distance to adjacent devices, electrical connection topology depth, and installation direction classification; The eigenvalue corresponding to each eigenvalue is normalized to obtain a normalized eigenvector dimensional space, and the similarity of the eigenvectors in the normalized eigenvector dimensional space is calculated by cosine similarity to extract the recording similarity eigenvector.

4. The method for automatic modeling of low-voltage power grid data based on visual acquisition and recording according to claim 1, characterized in that: Constructing a plurality of local coordinate systems corresponding to the plurality of sub-regions, the method comprising: Calculating the bounding boxes of the multiple sub-regions to obtain a set of spatial boundary coordinates and a set of spatial center coordinates; Using principal component analysis to fit the device distribution coordinates of each of the multiple sub-areas to extract a first principal axis direction, using the first principal axis direction as the X-axis and using a three-dimensional orthogonal rule to define the Y-axis and the Z-axis, and outputting a three-dimensional coordinate axis; The spatial center coordinates are used as the origin of a local coordinate system, and the three-dimensional coordinate axes are used as the coordinate axes of the local coordinate system to construct a plurality of local coordinate systems corresponding to the plurality of sub-regions.

5. The method for automatic modeling of low-voltage power grid data based on visual acquisition and recording according to claim 1, characterized in that: Configure a visual recording task template, which includes configuring the recording path, viewing angle parameters, image clarity, recording distance interval, point cloud density threshold, and coverage angle range.

6. The method for automatic modeling of low-voltage power grid data based on visual acquisition and recording according to claim 5, characterized in that: The visual data collection task template collects data from the power grid equipment to be collected in the task to be collected according to the at least one matching local coordinate system, and the method includes: The visual recording task template obtains at least one recording task instruction set corresponding to the at least one matching local coordinate system; The at least one matching local coordinate system is bound to the corresponding at least one recording task instruction set and sent to the recording terminal device of the first user. The recording terminal device records data of the power grid equipment to be recorded for the task to be recorded to obtain at least one set of equipment modeling data.

7. The method for automatic modeling of low-voltage power grid data based on visual acquisition and recording according to claim 1, characterized in that: Constructing a multi-region coordinate system conversion model based on the multiple local coordinate systems, the method includes: Obtain key parameters of each local coordinate system, including the three-dimensional coordinates of the origin in the global coordinate system, the coordinate axis direction vector, the local coordinate unit, and the angle definition method; According to the key parameters of each local coordinate system and the transformation formula of the pre-constructed global coordinate system, a local-global coordinate transformation matrix is ​​constructed. The local-global coordinate transformation matrix is ​​used to store the corresponding sub-region index to establish a multi-region coordinate system transformation model.

8. The method for automatic modeling of low-voltage power grid data based on visual acquisition and recording according to claim 7, characterized in that: The method of performing coordinate system transformation on the at least one matching local coordinate system by using the multi-region coordinate system transformation model includes: Obtaining key conversion parameters of the at least one matching local coordinate system using the multi-region coordinate system conversion model; Obtaining initial device modeling conversion data corresponding to at least one set of device modeling data according to the conversion key parameters; Identifying a cross-region anchor device, where the cross-region anchor device is a public device located at the intersection of two or more sub-regions; Calculating, by the cross-region anchor point device, spatial deviation vectors of the initial device modeling conversion data corresponding to the at least one set of device modeling data, including rotation deviation vectors and translation deviation vectors; The initial device modeling conversion data is realigned according to the spatial deviation vector to obtain updated device modeling conversion data.

9. The method for automatic modeling of low-voltage power grid data based on visual acquisition and recording according to claim 1, characterized in that: Generating a visual modeling simulation result of the task to be recorded according to the device modeling conversion data; The visual modeling simulation result is outputted through an interactive simulation module, and the interactive simulation module performs a visual display response by receiving the interactive operation of the first user.

10. The low-voltage power grid data automatic modeling system based on visual recording is characterized by: A system for implementing the method for automatic modeling of low-voltage power grid data based on visual collection and recording according to any one of claims 1 to 9, comprising: A sub-area division module is used to divide the low-voltage power grid area into multiple sub-areas and construct multiple local coordinate systems corresponding to the multiple sub-areas; A coordinate system determination module, configured to analyze the power grid equipment to be recorded of the task to be recorded of the first user, and determine at least one matching local coordinate system from the multiple local coordinate systems according to the type of the power grid equipment to be recorded; A data collection module is used to configure a visual collection task template, wherein the visual collection task template collects data from the power grid devices to be collected in the task to be collected according to the at least one matching local coordinate system to obtain at least one set of device modeling data; A simulation result generation module is used to construct a multi-region coordinate system conversion model based on the multiple local coordinate systems, use the multi-region coordinate system conversion model to perform coordinate system conversion on at least one matching local coordinate system, obtain device modeling conversion data corresponding to at least one set of device modeling data, and generate a visual modeling simulation result of the task to be recorded according to the device modeling conversion data.

Images