Method and system for automatically updating power grid equipment graph model based on data collected and recorded by unmanned aerial vehicle

Abstract

The invention provides an automatic updating method and system for a power grid equipment graph model based on data collected and recorded by an unmanned aerial vehicle. The unmanned aerial vehicle is used for regularly patrolling a power grid line and collecting image data and position data of power grid equipment in real time; on the basis of a power grid GIS system and identified equipment information, topology key nodes of the power grid are automatically identified, and data collected by the unmanned aerial vehicle are compared and matched with an existing power grid topology structure through a data matching algorithm; providing a user interaction interface for a user to check a matching result, and performing confirmation or necessary adjustment; based on topological key node information, information of other devices on a line is automatically updated, newly collected data is integrated with an existing power grid device graph model, and corresponding power grid device graph model data is automatically updated. According to the method, the key topological nodes of the power grid can be automatically identified, the equipment information on the power grid line is automatically updated, the updating efficiency and accuracy of the power grid equipment graph model are improved, and real-time updating and sharing of data are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid management, and in particular to a method and system for automatically updating a power grid equipment model based on data collected and recorded by an unmanned aerial vehicle. Background Art

[0002] As an important infrastructure in modern society, it is very important to maintain the long-term safe and stable operation of power grid equipment. The power grid diagram model refers to the graphics and models of the power grid system. It is an abstract description of the connection characteristics of the actual distribution network. It is the basis for real-time monitoring and various advanced analysis and application functions of the power grid automation master station system. It fundamentally determines the practical level and construction success of the power grid automation system. At present, one of the main factors affecting the practical operation of the power grid automation system is the accuracy of the distribution network diagram model. The distribution network diagram model has the problems of low quality level, large maintenance workload, and frequent updating and transformation. With the continuous expansion of the scale of the power grid, the traditional manual inspection method cannot meet the needs of efficient and intelligent management of modern power grids. As an emerging intelligent inspection equipment, drones can effectively solve many problems existing in traditional manual inspection methods with their efficient and accurate working characteristics.

[0003] However, relying solely on drones’ high-definition cameras and lidar sensors to collect information about poles and their surrounding equipment is not enough to achieve intelligent power grid management. Considering the flight characteristics of drones, real-time monitoring of flight status and data collection is required. The key topological nodes in the data collected by drones include the flight start point, end point, branch point, and topological discontinuity point. Automatically comparing and matching these key topological nodes with the existing power grid topology is a new challenge. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a highly automated and high-precision method and system for automatically updating power grid equipment models based on data collected and recorded by drones.

[0005] Technical solution: A method for automatically updating a power grid equipment model based on drone-collected data, comprising the following steps:

[0006] (1) Set the flight route of the drone in the target area, use the drone to patrol the power grid lines, and collect image data and location data of power grid equipment in real time;

[0007] (2) Identify topological key nodes based on the collected grid equipment image data and location data, automatically compare and match the topological key nodes with the existing grid topology structure, and obtain automatic matching results;

[0008] (3) Based on the automatic matching results of topological key nodes, the collected power grid equipment image data and location data are integrated with the existing power grid equipment model data to update the power grid equipment model data.

[0009] Specifically, in step (1), the flight route of the UAV is determined by the power grid GIS system and the power grid map data, and the range of the flight route covers the entire target area.

[0010] Specifically, in step (1), the collected image data and location data of the power grid equipment are transmitted in real time to the ground control station via a wireless network for storage.

[0011] Specifically, step (1) also includes: extracting features from the collected power grid equipment image data and location data, and matching and verifying them with corresponding data in the power grid GIS system.

[0012] Specifically, in step (2), the key topological nodes include a starting point, an end point, and a branch point. After selecting the power grid equipment in the target area as the starting point, the image data of the starting point power grid equipment collected by the drone is matched with the power grid equipment data in the power grid GIS system through image recognition. After confirming the consistency, the power grid equipment is used as the starting point to start matching the key topological nodes; the end point is the power grid equipment at the end of the power grid line in the power grid GIS system. The image data of the end point power grid equipment collected by the drone is matched with the power grid equipment data in the power grid GIS system through image recognition; the branch point is the node of multiple lines on the same pole in the power grid GIS system. The image data of the branch point collected by the drone is matched with the image data of the branch point in the power grid GIS system through image recognition.

[0013] Preferably, in step (2), the topological key nodes also include topological discontinuity points, which are areas in the target area where the drone cannot collect image data and location data of power grid equipment. The location data of power grid equipment at the topological discontinuity points are calculated using the Kriging interpolation method through the existing nodes in the topological key nodes.

[0014] Specifically, the Kriging interpolation method includes: obtaining the position data of existing nodes in the topological key nodes, obtaining the distance value between the existing nodes and the topological discontinuity point based on the position data of the existing nodes, calculating the semi-variance values ​​at different distances, selecting the semi-variance function model, performing parameter fitting on the semi-variance function model, constructing a set of linear equations based on the existing nodes and the fitted semi-variance function model, solving to obtain the Kriging weights corresponding to each distance value, and using the Kriging interpolation formula based on the Kriging weights and the position data of the existing nodes to calculate the weighted average of all existing nodes for each topological discontinuity point to obtain the power grid equipment data at the topological discontinuity point.

[0015] Specifically, in step (3), without affecting the existing grid equipment model data structure, the difference information between the collected grid equipment image data and position data and the existing grid equipment model data is used as valid data to dynamically update the grid equipment model data.

[0016] Preferably, the above-mentioned method for automatically updating the power grid equipment model also includes the following steps:

[0017] The automatic matching results of topological key nodes are reviewed and confirmed, and the topological key nodes with identification errors or matching errors are modified based on expert experience.

[0018] The present invention also provides a power grid equipment image model automatic update system based on drone-collected data, comprising:

[0019] UAV data collection module: used to set the flight route of the UAV in the target area, use the UAV to patrol the power grid lines, and collect image data and location data of power grid equipment in real time;

[0020] Topology key node matching module: used to identify topology key nodes based on the collected power grid equipment image data and location data, automatically compare and match the topology key nodes with the existing power grid topology structure, and obtain automatic matching results;

[0021] Power grid equipment model automatic update module: used to integrate the collected power grid equipment image data and location data with the existing power grid equipment model data based on the automatic matching results of topological key nodes, and update the power grid equipment model data.

[0022] Beneficial effects: Compared with the prior art, the significant effects of the present invention are: using drones equipped with high-precision cameras and sensors to conduct regular inspections of power grid lines and collect image data and location data of power grid equipment (towers and their surrounding equipment) in the target area in real time, which greatly reduces labor costs and has accuracy and real-time performance; based on the power grid GIS system data and the identified equipment information, the topological key nodes of the power grid are automatically identified, and compared and matched with the existing power grid topological structure, and the expert experience review and confirmation link of the automatic matching results is added, which greatly improves the reliability of the matching results; based on the topological key nodes, the newly collected data is integrated with the existing power grid equipment model data, and the corresponding power grid equipment model data is automatically updated, realizing accurate, efficient and automatic updating of the power grid equipment model. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of the method of embodiment 1 of the present invention.

[0024] Figure 2 This is a flowchart of topological key node matching in Example 1 of the present invention.

[0025] Figure 3 It is a schematic diagram of the automatic update result of the power grid equipment model in Example 1 of the present invention. DETAILED DESCRIPTION

[0026] The present invention is further described below in conjunction with the accompanying drawings.

[0027] Example 1

[0028] See also Figure 1 As shown, this embodiment provides a method for automatically updating a power grid equipment model based on drone-collected data, comprising the following steps:

[0029] (1) Set the flight route of the drone in the target area, use the drone to patrol the power grid lines, and collect image data and location data of power grid equipment in real time.

[0030] The drone used in this embodiment is equipped with a high-precision camera and a lidar sensor, which can collect high-definition image data and accurate three-dimensional position data of power grid equipment (poles and towers and their peripheral equipment) in real time during flight. The drone has strong flexibility and maneuverability, and can cover areas in the target area that are difficult to reach by traditional manual patrols, thereby ensuring the comprehensiveness and accuracy of the data.

[0031] Before the UAV conducts patrol operations, the flight route of the UAV is planned according to the specific conditions of the power grid lines to be patrolled using the power grid GIS system and power grid model data to ensure that the flight route fully covers the target area. The accuracy and stability of the flight route are guaranteed by GPS and INS (inertial navigation system). The flight altitude and speed are adjusted according to the specific conditions of the power grid lines to obtain the best image clarity and data accuracy. The collected image data and location data are transmitted to the ground control station in real time via the wireless network. The ground control station monitors the flight status and data collection of the UAV in real time to ensure the integrity and quality of the data.

[0032] After the drone data collection is completed, the collected power grid equipment image data and location data are processed. Taking the pole tower as an example, the type, location, height, material and other detailed information of the pole tower are extracted from the image data through feature extraction and pattern recognition, and matched and verified with the data in the power grid GIS system. The processing of image data includes noise removal, edge detection, graphics enhancement, feature extraction, etc. The processed data is stored and retrieved through an efficient data management platform. Commonly used data management platforms include Oracle, MySQL, Redis, MongoDB, etc. The present invention mainly uses Oracle+MongoDB, and finally obtains high-precision pole tower information as the basis for subsequent power grid equipment model updates.

[0033] (2) Identify topological key nodes based on the collected grid equipment image data and location data, and automatically compare and match the topological key nodes with the existing grid topology structure to obtain automatic matching results.

[0034] See also Figure 2 As shown, in the present invention, the topological key nodes include the starting point, the end point, the branch point, and the topological discontinuity point. The starting point of the inspection is usually confirmed by the drone pilot according to the inspection work order, which is generally the power station or the first tower led out of the power station. After confirming the power grid equipment as the starting point in the target area, the image data of the starting point power grid equipment collected by the drone is matched with the power grid equipment data in the power grid GIS system through image recognition. After confirming that their positions and types are consistent, the power grid equipment is used as the starting point to start matching the topological key nodes.

[0035] The end point of the inspection is the power grid equipment at the end of the power grid line in the power grid GIS system, usually a station building or a pole tower. In some actual scenarios, there is also a user access point. The equipment is located in a building, and the corresponding equipment range needs to be adjusted during matching. The image data of the end point power grid equipment collected by the drone is matched with the power grid equipment data in the power grid GIS system through image recognition to confirm each end point.

[0036] The branch point is a node in the power grid GIS system where multiple lines are installed on the same pole, that is, multiple lines are branched to multiple pole tower devices. The branch point image data collected by the drone is matched with the branch point image data in the power grid GIS system through image recognition.

[0037] Topological discontinuities are areas in the target area where drones cannot collect image data and location data of power grid equipment, such as no-fly zones or areas where cables are buried underground. Such areas are identified and marked through GIS data and preset flight restriction zones. For underground cables, the location of topological discontinuities can be inferred through the line direction and existing underground cable data. The location data of power grid equipment at topological discontinuities can also be calculated through existing nodes in topological key nodes using the Kriging interpolation method to ensure the integrity and accuracy of the overall topological structure.

[0038] The application of the Kriging interpolation method in the present invention is described below: Taking a pole tower as an example, the position data of a known pole tower (i.e., a node), such as three-dimensional coordinate data, is obtained, and then according to the marked no-fly zone or cable underground area, based on the position data of the known pole tower, the distance value between the known pole tower and the topological discontinuity point is obtained, and the semi-variance value γ(h) at different distances h is calculated, and a suitable semi-variance function model (such as a spherical model or a Gaussian model) is selected, and the semi-variance function model is parameter-fitted. A linear equation group is constructed according to the data of the known pole tower and the fitted semi-variance function model, and the Kriging weight corresponding to each distance value is obtained by solving the Kriging weight and the position data of the known pole tower. The Kriging interpolation formula is used to calculate the weighted average of all known pole towers for each topological discontinuity point, and the pole tower position and related data at the topological discontinuity point are obtained. The Kriging interpolation formula is as follows:

[0039]

[0040] Where: Z(x0) is the estimated value at the point x0 to be estimated; Z(x i ) is a known point x i The observed value at i is the Kriging weight at the point x0 to be estimated; n is the number of known points;

[0041] Taking the Gaussian model as an example, the semivariogram model formula is as follows:

[0042]

[0043] Where: γ(h) is the semivariogram function, h is the distance between two points; c0 is the nugget, which indicates the variation at a small scale; c is the sill, which indicates the overall variation of the data; a is the range, which indicates the distance at which there is no correlation between the data. The parameters c0, c, and a are determined by fitting the known node h and the corresponding semivariogram value γ(h).

[0044] (3) Review and confirm the automatic matching results of topological key nodes, and modify the topological key nodes that are misidentified or mismatched based on expert experience.

[0045] In order to ensure the accuracy and reliability of the data, a data review and confirmation link is introduced. Through the user interaction interface, intuitive data comparison is carried out to display the image data collected by the drone, the identified topological key nodes and the matched power grid GIS data. Users can view the specific node information through this interface, confirm or adjust the results of automatic identification, and manually modify the node position or feature information when identification errors or inaccurate matching are found. Users can also add new nodes or delete erroneous nodes according to actual needs to ensure the accuracy and completeness of the data.

[0046] At the same time, each review and adjustment operation of the user automatically generates an audit log, which facilitates traceability and provides a reference for subsequent algorithm optimization.

[0047] (4) Based on the automatic matching results of topological key nodes, the collected power grid equipment image data and location data are integrated with the existing power grid equipment model data to update the power grid equipment model data.

[0048] After the data review and confirmation is completed, without affecting the existing power grid equipment model data structure, the difference information between the collected power grid equipment image data and position data and the existing power grid equipment model data is used as valid data to dynamically update the power grid equipment model data.

[0049] Please refer to Figure 3 As shown in the figure, the gray dots are the original tower coordinate positions, the green and red dots are the tower coordinate positions obtained by recording and re-matching the drone in this embodiment, and the green nodes are topological key nodes. From the comparison between the original model and the updated model, it can be seen that through the automatic updating method of the power grid equipment model provided by this embodiment, after the drone is used to collect data on the power grid lines in the target area, the automatic updating of the power grid equipment model is realized quickly and accurately.

[0050] Example 2

[0051] This embodiment provides a power grid equipment image model automatic updating system corresponding to the power grid equipment image model automatic updating method based on drone recorded data described in Embodiment 1, including the following modules:

[0052] UAV data collection module: used to set the flight route of the UAV in the target area, use the UAV to patrol the power grid lines, and collect image data and location data of power grid equipment in real time;

[0053] Topology key node matching module: used to identify topology key nodes based on the collected power grid equipment image data and location data, automatically compare and match the topology key nodes with the existing power grid topology structure, and obtain automatic matching results;

[0054] Data review module: used to review and confirm the automatic matching results of topological key nodes, and modify the topological key nodes with identification errors or matching errors based on expert experience;

[0055] Power grid equipment model automatic update module: used to integrate the collected power grid equipment image data and location data with the existing power grid equipment model data based on the automatic matching results of topological key nodes, and update the power grid equipment model data.

Claims

1. A method for automatically updating a power grid equipment model based on drone-collected data, characterized in that: The following steps are involved: (1) Set the flight route of the drone in the target area, use the drone to patrol the power grid lines, and collect image data and location data of power grid equipment in real time; (2) Identify topological key nodes based on the collected grid equipment image data and location data, automatically compare and match the topological key nodes with the existing grid topology structure, and obtain automatic matching results; (3) Based on the automatic matching results of topological key nodes, the collected power grid equipment image data and location data are integrated with the existing power grid equipment model data to update the power grid equipment model data.

2. The method for automatically updating the power grid equipment model according to claim 1, characterized in that: In the step (1), the flight route of the UAV is determined by the power grid GIS system and the power grid map data, and the range of the flight route covers the entire target area.

3. The method for automatically updating the power grid equipment model according to claim 1, characterized in that: In the step (1), the collected image data and location data of the power grid equipment are transmitted in real time to the ground control station via a wireless network for storage.

4. The method for automatically updating the power grid equipment model according to claim 1, characterized in that: The step (1) also includes: extracting features from the collected power grid equipment image data and location data, and matching and verifying them with corresponding data in the power grid GIS system.

5. The method for automatically updating the power grid equipment model according to claim 1, characterized in that: In the step (2), the key topological nodes include a starting point, an end point, and a branch point. After selecting the power grid equipment in the target area as the starting point, the image data of the starting point power grid equipment collected by the drone is matched with the power grid equipment data in the power grid GIS system through image recognition. After confirming the consistency, the power grid equipment is used as the starting point to start matching the key topological nodes; The end point is the power grid equipment at the end of the power grid line in the power grid GIS system. The image data of the end point power grid equipment collected by the drone is matched with the power grid equipment data in the power grid GIS system through image recognition; the branch point is the node with multiple lines on the same pole in the power grid GIS system. The branch point image data collected by the drone is matched with the branch point image data in the power grid GIS system through image recognition.

6. The method for automatically updating the power grid equipment model according to claim 5, characterized in that: In the step (2), the topological key nodes also include topological discontinuity points, which are areas in the target area where the drone cannot collect image data and location data of power grid equipment. The power grid equipment location data at the topological discontinuity points are calculated using the Kriging interpolation method through the existing nodes in the topological key nodes.

7. The method for automatically updating the power grid equipment model according to claim 6, characterized in that: The Kriging interpolation method includes: obtaining the position data of existing nodes in the topological key nodes, obtaining the distance value between the existing nodes and the topological discontinuity point based on the position data of the existing nodes, calculating the semivariance values ​​at different distances, selecting a semivariance function model, performing parameter fitting on the semivariance function model, constructing a linear equation group according to the existing nodes and the fitted semivariance function model, solving to obtain the Kriging weights corresponding to each distance value, and using the Kriging interpolation formula according to the Kriging weights and the position data of the existing nodes to calculate the weighted average of all existing nodes for each topological discontinuity point to obtain the power grid equipment data at the topological discontinuity point.

8. The method for automatically updating the power grid equipment model according to claim 1, characterized in that: In the step (3), without affecting the existing grid equipment model data structure, the difference information between the collected grid equipment image data and position data and the existing grid equipment model data is used as valid data to dynamically update the grid equipment model data.

9. The method for automatically updating the power grid equipment model according to claim 1, characterized in that: The following steps are also included: The automatic matching results of topological key nodes are reviewed and confirmed, and the topological key nodes with identification errors or matching errors are modified based on expert experience.

10. A power grid equipment model automatic update system based on drone data collection, characterized in that: include: UAV data collection module: used to set the flight route of the UAV in the target area, use the UAV to patrol the power grid lines, and collect image data and location data of power grid equipment in real time; Topology key node matching module: used to identify topology key nodes based on the collected power grid equipment image data and location data, automatically compare and match the topology key nodes with the existing power grid topology structure, and obtain automatic matching results; Power grid equipment model automatic update module: used to integrate the collected power grid equipment image data and location data with the existing power grid equipment model data based on the automatic matching results of topological key nodes, and update the power grid equipment model data.

Images