Power grid planning method, system and device using unmanned aerial vehicle, and storage medium

Abstract

The invention provides a power grid planning method, system and device using an unmanned aerial vehicle, and a storage medium, and the method comprises the steps: collecting control points and environment data through the unmanned aerial vehicle; environment data are input into a neural network model to generate coordinates of all towers between a starting point and an ending point, and in the training process of the neural network model, model parameters are verified and adjusted through calculation of mechanical characteristics borne by the towers and reverse gradient propagation; point lofting is carried out in batches based on tower coordinates, a visual model is used for identifying point lofting errors, and coordinates of the corner tower are remeasured based on control point coordinates; correcting the point lofting error and then carrying out line lofting to obtain a power grid planning result; and establishing twin model storage data corresponding to the power grid planning result. According to the application, the comprehensive functions of power grid planning by using the unmanned aerial vehicle, control point acquisition, new line tower position determination, point lofting and line lofting are designed on the whole, and manual design is comprehensively replaced.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid planning, and in particular to a method, system, device and storage medium for power grid planning using an unmanned aerial vehicle. Background Art

[0002] Using drones for grid planning is a modern, efficient and precise technical means aimed at optimizing the design, installation and maintenance of power transmission lines. It can leverage the drones' unique aerial advantages to provide detailed data support for grid planning.

[0003] In the related technologies, in the traditional power grid planning scheme using drones, the role of drones is limited to the data collection stage, that is, carrying lidar equipment to obtain terrain data, and mainly relying on manual calculation in the process of path planning and tower site selection; with the development of artificial intelligence, scholars have tried to use simpler path planning algorithms and neural network models to solve path planning problems, but the existing technology only considers avoiding obstacles and minimizing the cost path, but does not consider whether the towers meet the physical constraints after the actual situation is established. Some towers that need to be built on slopes or located in other unfavorable positions are prone to tilt, which can easily lead to safety accidents.

[0004] Based on the above analysis of the development status of this technical field, the existing technology lacks a solution that includes path tower site selection, automatic point and line layout, and the use of mechanical characteristics to verify and feedback optimization parameters during the training process of the neural network model used to generate tower coordinates. Summary of the Invention

[0005] The purpose of the present invention is to provide a method, system, device and storage medium for power grid planning using drones, aiming to solve the above-mentioned problems in the prior art.

[0006] According to a first aspect of an embodiment of the present invention, a method for power grid planning using a drone is provided, comprising:

[0007] Use drones to collect control points and environmental data;

[0008] The environmental data is input into the neural network model to generate the coordinates of each tower between the starting point and the end point. During the training process, the neural network model verifies and adjusts the model parameters by calculating the mechanical characteristics of the tower and performing reverse gradient propagation.

[0009] Perform point staking in batches based on tower coordinates, use visual models to identify point staking errors, and remeasure the coordinates of corner towers based on control point coordinates. After correcting point staking errors, perform line staking to obtain power grid planning results.

[0010] Establish a twin model to store data corresponding to the power grid planning results.

[0011] According to a second aspect of an embodiment of the present invention, a system for power grid planning using a drone is provided, comprising:

[0012] The acquisition module is used to collect control points and environmental data using drones;

[0013] A coordinate generation module is used to input environmental data into a neural network model to generate the coordinates of each tower between the starting point and the end point. During the training process, the neural network model verifies and adjusts the model parameters by calculating the mechanical characteristics of the tower and performing reverse gradient propagation.

[0014] The staking and correction module is used to perform batch point staking based on tower coordinates, identify point staking errors using a visual model, and remeasure the coordinates of corner towers based on control point coordinates. After correcting point staking errors, line staking is performed to obtain power grid planning results.

[0015] The storage module is used to establish a twin model storage data corresponding to the power grid planning results.

[0016] According to a third aspect of an embodiment of the present invention, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of the method for planning a power grid using a drone as provided in the first aspect of the present disclosure are implemented.

[0017] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, on which an implementation program for information transmission is stored. When the program is executed by a processor, the steps of the method for planning a power grid using a drone provided in the first aspect of the present disclosure are implemented.

[0018] The technical solution provided by the embodiment of the present invention includes the following beneficial effects: an overall method for using drones for power grid planning is designed, including the comprehensive functions of control point acquisition, determination of tower positions for new lines, point layout and line layout, which completely replaces manual design; a neural network model is used to generate the coordinates of each tower between the starting point and the end point, and during the training process, the neural network model verifies and adjusts the model parameters by calculating the mechanical characteristics of the towers through reverse gradient propagation. The neural network model can handle complex nonlinear terrain conditions, and physical constraints can improve the generalization and rationality of the model output. A comprehensive prediction structure of neural network prediction and physical correction is established, which significantly improves the engineering applicability of the model.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 is a flow chart of a method for power grid planning using a drone according to an embodiment of the present invention;

[0022] Figure 2 is a schematic diagram of a power grid planning system using a drone according to an embodiment of the present invention;

[0023] Figure 3 is a schematic diagram of a basic interface of an embodiment of the present invention;

[0024] Figure 4 is a schematic diagram of a path planning interface according to an embodiment of the present invention;

[0025] Figure 5 is a schematic diagram of a point stakeout interface according to an embodiment of the present invention;

[0026] Figure 6 is a schematic diagram of a line lofting interface according to an embodiment of the present invention;

[0027] Figure 7 is a schematic diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below in conjunction with the drawings in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this document.

[0029] Method Example

[0030] According to an embodiment of the present invention, a method for power grid planning using a drone is provided. Figure 1 FIG. 1 is a flow chart of a method for planning a power grid using a drone according to an embodiment of the present invention. Figure 1 As shown, the method for planning a power grid using a drone according to an embodiment of the present invention specifically includes:

[0031] In step S110, a drone is used to collect control point and environmental data, specifically including:

[0032] Use drones for high-altitude surveys before route planning to obtain the actual on-site conditions.

[0033] The three-dimensional coordinates of control points are obtained through the lidar equipment on the drone. The control points are used to provide high-precision benchmark data for subsequent planning to ensure the accuracy of the plan.

[0034] The camera equipment carried by the drone records natural data in the vertical direction and road intersection data in the horizontal direction, and the natural data and road intersection data are standardized as environmental data; the environmental data records information such as terrain and obstacles such as buildings or forests. Standardization can store three-dimensional scenes in data form.

[0035] In step S120, the environmental data is input into the neural network model to generate the coordinates of each tower between the starting point and the end point. During the training process, the neural network model verifies and adjusts the model parameters by calculating the mechanical characteristics of the tower and performing reverse gradient propagation, specifically including:

[0036] The location information of the starting and ending points is determined according to the actual situation of the project, and the line composed of towers needs to be determined in combination with the data obtained from the drone survey.

[0037] The neural network model used in the embodiment of the present invention is a Seq2Seq model. The encoder processes the environmental data, and the decoder uses an RNN or Transformer model to gradually generate a coordinate sequence, such as (x1, y1, z1), (x2, y2, z2) ... (xn, yn, zn). It should be noted that all coordinates in the embodiment of the present invention are three-dimensional coordinates, that is, they include latitude, longitude, and elevation.

[0038] In the application verification phase, inputting environmental data into the trained neural network model can output a coordinate sequence. Preferably, the span constraint distance requirement can be set in the neural network model. However, during the training process, it is necessary to verify the mechanical characteristics of the tower and adjust the model parameters through reverse gradient propagation:

[0039] Generate the initial coordinate sequence of the tower based on the environmental data;

[0040] Since, in the embodiment of the present invention, the mechanical feature is only a quantity that assists in model parameter optimization and does not require high accuracy, in the initial coordinate sequence, a regression algorithm is used based on adjacent coordinate information to simulate the horizontal tension exerted on the intermediate coordinates. The independent variables of the regression algorithm include, but are not limited to, features related to the coordinate information, and the dependent variable is the horizontal tension.

[0041] Calculate the tower base overturning moment and base vertical reaction force as mechanical characteristics based on the horizontal tension value; obtain the horizontal tension, the preset tower height, the tower deadweight and the unit weight of the conductor;

[0042] The base overturning moment M is calculated using Equation 1:

[0043] M=H i h Formula 1;

[0044] Among them, H i It represents the horizontal tension of tower i, and h represents the height of the tower;

[0045] Use formula 2 to calculate the base vertical reaction force R:

[0046]

[0047] Among them, G i represents the deadweight of tower i, ω represents the unit weight of conductor, L i-1,i Indicates the distance between tower i-1 and tower i, L i,i+1 It represents the distance between tower i and tower i+1, where the distance is in the horizontal direction.

[0048] Determine whether the mechanical characteristics exceed the engineering specification threshold as a calculation loss, and use the calculation loss to reverse the gradient to adjust the model parameters. This is equivalent to the gradient back propagation of the loss forcing the network to adjust the parameters so that the model's prediction results conform to physical laws. During the training process, coordinates that are more in line with physical laws are gradually predicted;

[0049] In an embodiment of the present invention, the loss core is that the greater the loss is when it exceeds the engineering specification, the smaller the loss is otherwise. After the iteration is completed, a trained neural network model is obtained. Using this optimized model in the application process can directly generate results that meet physical constraints.

[0050] In step S130, point staking is performed in batches based on the tower coordinates. The visual model is used to identify point staking errors. The coordinates of the corner towers are remeasured based on the control point coordinates. After correcting the point staking errors, line staking is performed to obtain the power grid planning results. Specifically, the following steps are performed:

[0051] The purpose of point staking is to mark the coordinates of the designed tower on the ground. After selecting the route for multi-point staking, the drone automatically selects the nearest tower for staking. This process requires cooperation with ground staff in pile marking during application.

[0052] After point staking is completed, line staking needs to be carried out. Line staking means constructing a line segment by selecting the control starting point and end point, that is, between two towers. After reaching the control end point, the control point is reselected to continue working. Line staking ensures that the conductor does not swing within the maximum deviation range under working conditions such as wind deviation and icing. Line staking marks these ranges to ensure the safety of the construction process.

[0053] During the batch point staking process, the drone transmits the staking images in real time, and the error between the staking points and the coordinates output by the neural network model is compared using the visual model feature point matching algorithm. The staking points with errors greater than a preset distance are selected as the suspected error set. In this embodiment of the present invention, the preset distance is 5 cm.

[0054] The corner towers in the suspected error set were screened and their coordinates were remeasured using the control point-based resection method to obtain the final error set. Since the corner towers bear the reasonable load of two conductors in different directions, even a slight deviation in their setting can lead to a significant increase in tension. Therefore, the resection method was used to reverse-calculate the precise position of the towers using the measured known control points, further verifying whether there was any deviation at the millimeter level.

[0055] Finite micro-offsets are performed on non-corner towers in the final error set, and key points such as corner towers are recalculated using the software PLS-CADD.

[0056] In step S140, a twin model storage data corresponding to the power grid planning result is established, specifically including:

[0057] Use Unity3D models for visual rendering to generate a digital interface including the scene and its power grid planning, and store the three-dimensional coordinates of control points and towers in an Excel format data table.

[0058] The method further comprises:

[0059] In step S150, the sensor device installed at the bottom of the tower monitors data changes in real time and issues an alarm when the data changes abnormally, specifically including:

[0060] Changes are monitored through inclination, tension, and temperature sensors installed near the tower.

[0061] To sum up, in response to the existing problems, this invention uses a drone power grid planning method and designs a comprehensive method for using drones for power grid planning, including control point collection, new line determination of tower positions, point layout and line layout, which completely replaces manual design; a neural network model is used to generate the coordinates of each tower between the starting point and the end point, and during the training process of the neural network model, the mechanical characteristics of the tower are calculated and the model parameters are adjusted by reverse gradient propagation. The neural network model can handle complex nonlinear terrain conditions, and physical constraints can improve the generalization and rationality of the model output. A comprehensive prediction structure of neural network prediction and physical correction is established, which significantly improves the engineering applicability of the model; a regression algorithm is used to quickly predict horizontal tension to avoid the huge time consumption caused by catenary modeling, and the base overturning moment and base vertical reaction force can be quickly calculated as mechanical characteristics; during the point layout process, the coordinates of the corner towers in the suspected error set are reviewed at the millimeter level to avoid measurement errors at key positions.

[0062] System Example

[0063] According to an embodiment of the present invention, a power grid planning system using drones is provided. Figure 2 FIG. 1 is a schematic diagram of a power grid planning system using a drone according to an embodiment of the present invention. Figure 2 As shown, the power grid planning system using drones according to an embodiment of the present invention specifically includes:

[0064] The acquisition module 20 is used to use a drone to collect control point and environmental data, specifically for:

[0065] The three-dimensional coordinates of the control points are obtained through the laser radar equipment carried by the drone;

[0066] The camera equipment carried by the drone records natural data in the vertical direction and road intersection data in the horizontal direction. The natural data and road intersection data are standardized and used as environmental data.

[0067] The coordinate generation module 22 is used to input the environmental data into the neural network model to generate the coordinates of each tower between the starting point and the end point. During the training process, the neural network model verifies the mechanical characteristics of the tower and adjusts the model parameters through reverse gradient propagation. Specifically, it is used to:

[0068] The model parameters are verified by calculating the mechanical characteristics of the tower and adjusting them through reverse gradient propagation, including:

[0069] Generate the initial coordinate sequence of the tower based on the environmental data;

[0070] In the initial coordinate sequence, a regression algorithm is used to simulate the horizontal tension of the intermediate coordinates based on the adjacent coordinate information. The base overturning moment and base vertical reaction force of the tower are calculated based on the value of the horizontal tension as mechanical characteristics.

[0071] Obtain horizontal tension, pre-set tower height, tower weight and conductor unit weight;

[0072] The base overturning moment M is calculated using Equation 1:

[0073] M=H i h Formula 1;

[0074] Among them, H i It represents the horizontal tension of tower i, and h represents the height of the tower;

[0075] Use Formula 2 to calculate the base vertical reaction force R:

[0076]

[0077] Among them, G i represents the deadweight of tower i, ω represents the unit weight of conductor, L i-1,i Indicates the distance between tower i-1 and tower i, L i,i+1 Indicates the spacing between tower i and tower i+1.

[0078] Whether the mechanical characteristics exceed the engineering specification threshold is judged as the calculation loss, and the calculation loss is propagated through reverse gradient to adjust the model parameters.

[0079] The staking and correction module 24 is used to perform point staking in batches based on the tower coordinates, identify point staking errors using a visual model, and remeasure the coordinates of the corner towers based on the control point coordinates; after correcting the point staking errors, perform line staking to obtain the power grid planning results. Specifically, it is used to:

[0080] During the batch point staking process, the drone transmits the staking images in real time, and uses the visual model feature point matching algorithm to compare the errors between the staking points and the coordinates output by the neural network model. The staking points with errors greater than the preset distance are selected as the suspected error set.

[0081] The corner towers in the suspected error set are screened, and the coordinates of the corner towers are remeasured using the back intersection method based on control points to obtain the final error set.

[0082] The storage module 26 is used to establish a twin model storage data corresponding to the power grid planning results.

[0083] The system further comprises:

[0084] The application module 28 is used to monitor data changes in real time through the sensor equipment installed at the bottom of the tower, and to issue an alarm when the data changes abnormally.

[0085] The above technical solutions of the embodiments of the present invention are illustrated with reference to the following drawings.

[0086] Figure 3 is a schematic diagram of the basic interface of an embodiment of the present invention, such as Figure 3 As shown, the basic function interface of the drone is displayed; Figure 4 is a schematic diagram of a path planning interface according to an embodiment of the present invention. Figure 4 As shown in the figure, the process of verifying the results output by the neural network model in a real environment is demonstrated;

[0087] Figure 5 is a schematic diagram of the point layout interface of an embodiment of the present invention, such as Figure 5 As shown, a multi-point stakeout mode is demonstrated; Figure 6 is a schematic diagram of the line layout interface of an embodiment of the present invention, such as Figure 6 As shown, the process of constructing line segments is demonstrated.

[0088] Preferably, the drone also provides an angle dividing line prompt when implementing the point staking function. When the point is staked out to the corner tower, the system interface prompts the direction of the angle dividing line for accurate position marking; and the deviation of the drone from the baseline must be within the specified range. In the embodiment of the present invention, the deviation range from the baseline is within 1 meter.

[0089] Preferably, from the perspective of portability and ease of use, the drones in the embodiments of the present invention are DJI Mavic 3 and DJI Matrice 4 series drones. If there is a need for cross-section survey, DJI Matrice 4 is the most preferred because this series of drones has a laser ranging function.

[0090] To sum up, in response to the existing problems, this invention uses a drone power grid planning system and designs a comprehensive method for using drones for power grid planning, including control point collection, new line determination of tower positions, point layout and line layout, which completely replaces manual design; a neural network model is used to generate the coordinates of each tower between the starting point and the end point, and during the training process of the neural network model, the mechanical characteristics of the tower are calculated and the model parameters are adjusted by reverse gradient propagation. The neural network model can handle complex nonlinear terrain conditions, and physical constraints can improve the generalization and rationality of the model output. A comprehensive prediction structure of neural network prediction and physical correction is established, which significantly improves the engineering applicability of the model; a regression algorithm is used to quickly predict horizontal tension to avoid the huge time consumption caused by catenary modeling, and the base overturning moment and base vertical reaction force can be quickly calculated as mechanical characteristics; during the point layout process, the coordinates of the corner towers in the suspected error set are reviewed at the millimeter level to avoid measurement errors at key positions.

[0091] Electronic device embodiment

[0092] Figure 7 is a schematic diagram of an electronic device according to an embodiment of the present invention. Electronic device 700 may include at least one processor 710 and memory 720. Processor 710 can execute instructions stored in memory 720. Processor 710 is communicatively coupled to memory 720 via a data bus. In addition to memory 720, processor 710 may also be communicatively coupled to input device 730, output device 740, and communication device 750 via the data bus.

[0093] The processor 710 may be any conventional processor, such as a commercially available CPU. The processor may also include a graphics processor (GPU), a field programmable gate array (FPGA), a system on chip (SOC), an application specific integrated circuit (ASIC), or a combination thereof.

[0094] The memory 720 may be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0095] In an embodiment of the present disclosure, executable instructions are stored in the memory 720, and the processor 710 can read the executable instructions from the memory 720 and execute the instructions to implement all or part of the steps of any of the drone power grid planning methods in the above exemplary embodiments.

[0096] Computer readable storage medium embodiments

[0097] In addition to the above-mentioned methods and systems, exemplary embodiments of the present disclosure may also be a computer program product or a computer-readable storage medium storing the computer program product. The computer product includes computer program instructions, which can be executed by a processor to implement all or part of the steps described in any of the above-mentioned exemplary embodiments of the method for grid planning using drones.

[0098] The computer program product may be written in any combination of one or more programming languages ​​to write program code for performing the operations of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, etc., as well as conventional procedural programming languages ​​such as "C" or similar programming languages ​​and scripting languages ​​(e.g., Python). The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0099] Computer-readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or components, or any combination thereof. More specific examples of readable storage media include: static random access memory (SRAM) electrically connected with one or more wires, electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk, or any suitable combination thereof.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for power grid planning using drones, characterized in that: include: Use drones to collect control points and environmental data; Inputting the environmental data into a neural network model to generate the coordinates of each tower between a starting point and an end point, wherein during the training process, the neural network model verifies and adjusts the model parameters by calculating the mechanical characteristics of the tower and performing reverse gradient propagation; Point staking is performed in batches based on the tower coordinates, point staking errors are identified using a visual model, and the coordinates of the corner towers are remeasured based on the control point coordinates; line staking is performed after correcting the point staking errors to obtain a power grid planning result; Establish a twin model storage data corresponding to the power grid planning results.

2. The method according to claim 1, characterized in that The method further comprises: Data changes are monitored in real time through sensor equipment installed at the bottom of the tower, and an alarm is issued when the data changes abnormally.

3. The method according to claim 1, characterized in that The use of drones to collect control point and environmental data specifically includes: The three-dimensional coordinates of the control points are obtained through the laser radar equipment carried by the drone; The camera equipment carried by the UAV records natural data in the vertical direction and road intersection data in the horizontal direction, and the natural data and the road intersection data are standardized as environmental data.

4. The method according to claim 1, wherein The verification and reverse gradient propagation adjustment of the model parameters by calculating the mechanical characteristics of the tower specifically includes: generating an initial coordinate sequence of the tower based on the environmental data; In the initial coordinate sequence, a regression algorithm is used to simulate the horizontal tension of the intermediate coordinates based on adjacent coordinate information, and the base overturning moment and base vertical reaction force of the tower are calculated according to the value of the horizontal tension as mechanical characteristics; It is determined whether the mechanical characteristics exceed the engineering specification threshold as a calculation loss, and the calculation loss is reversely propagated through the gradient to adjust the model parameters.

5. The method according to claim 4, characterized in that The calculation of the base overturning moment and the base vertical reaction force of the tower as mechanical characteristics based on the value of the horizontal tension specifically includes: Obtaining the horizontal tension, the preset tower height, the tower weight, and the unit weight of the conductor; The base overturning moment M is calculated using Equation 1: M=H i h Formula 1; Among them, H i It represents the horizontal tension of tower i, and h represents the height of the tower; Use formula 2 to calculate the base vertical reaction force R: Among them, G i represents the deadweight of tower i, ω represents the unit weight of conductor, L i-1,i Indicates the distance between tower i-1 and tower i, L i,i+1 Indicates the spacing between tower i and tower i+1.

6. The method according to claim 1, wherein The step of performing point staking out in batches based on the tower coordinates, identifying point staking out errors using a visual model, and re-measuring the coordinates of the corner towers based on the control point coordinates specifically includes: During the batch point staking process, the drone transmits the staking images in real time, and uses the visual model feature point matching algorithm to compare the errors between the staking points and the coordinates output by the neural network model. The staking points with errors greater than the preset distance are selected as the suspected error set. The corner towers in the suspected error set are screened, and the coordinates of the corner towers are remeasured using a control point-based resection method to obtain a final error set.

7. A power grid planning system using drones, characterized in that: include: The acquisition module is used to collect control points and environmental data using drones; A coordinate generation module, configured to input the environmental data into a neural network model to generate the coordinates of each tower between a starting point and an end point, wherein during the training process of the neural network model, the model parameters are verified by calculating the mechanical characteristics of the towers and adjusting the model parameters through reverse gradient propagation; The layout and correction module is used to perform point layout in batches based on the tower coordinates, identify point layout errors using a visual model, and re-measure the coordinates of the corner towers based on the control point coordinates; after correcting the point layout errors, line layout is performed to obtain the power grid planning results; A storage module is used to establish a twin model storage data corresponding to the power grid planning results.

8. The system according to claim 7, characterized in that The system further comprises: The application module is used to monitor data changes in real time through sensor equipment installed at the bottom of the tower and issue an alarm when the data changes abnormally.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of the method for planning a power grid using a drone are implemented as claimed in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores an implementation program for information transmission, and when the program is executed by the processor, the steps of the method for planning a power grid using a drone are implemented as described in any one of claims 1 to 6.

Images