Three-dimensional model-based transformer substation unmanned aerial vehicle inspection path planning processing method and system

Abstract

The invention discloses a transformer substation unmanned aerial vehicle inspection path planning processing method and system based on a three-dimensional model. The method comprises the steps of obtaining three-dimensional environment data in a transformer substation; based on the three-dimensional environment data, an obstacle model is established through an obstacle point cloud recognition algorithm, and an obstacle distribution diagram is generated according to the obstacle model; the obstacles comprise power lines and equipment in the transformer substation; constructing an electromagnetic interference field distribution diagram, wherein the electromagnetic interference field distribution diagram is generated by combining the electromagnetic radiation height of the equipment in the transformer substation and the equipment arrangement position information; based on the obstacle distribution diagram and the electromagnetic interference field distribution diagram, generating an initial unmanned aerial vehicle inspection path by using a global path planning algorithm; according to the method, physical obstacles and electromagnetic interference possibly existing in the environment can be effectively handled, and high-safety and high-reliability flight path planning guarantee is provided for unmanned aerial vehicle inspection.

Description

Technical Field

[0001] The present invention relates to the field of data processing for UAV path planning in substations, and particularly to a method and system for processing UAV inspection path planning in substations based on a three-dimensional model. Background Art

[0002] During the inspection process of substations, it is necessary to conduct UAV path inspection and path planning over dense electrical equipment. However, due to the limited internal space and numerous obstacles in substations, UAVs are required to have high-precision positioning capabilities (such as RTK GPS technology) and obstacle avoidance functions to achieve accurate route planning and autonomous flight.

[0003] However, this is not the most dangerous. Specifically, researchers have found that determining target obstacles by combining laser point cloud analysis of obstacles within a safe distance and analyzing their electromagnetic interference is what needs to be most noted in path planning. There are a large number of electrical equipment and strong electric fields in substations, which may interfere with the electronic equipment of UAVs. Therefore, it is necessary to perform high-precision navigation and recognition of laser point clouds and electromagnetic interference analysis to implement effective measures to reduce the impact of electromagnetic interference on flight control. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for processing UAV inspection path planning in substations based on a three-dimensional model, which solves the above-mentioned technical problems pointed out in the prior art.

[0005] The present invention proposes a method for processing UAV inspection path planning in substations based on a three-dimensional model, including the following operating steps: Obtain the three-dimensional environmental data of the substation, where the three-dimensional environmental data includes point cloud data of power lines obtained by lidar measurement and three-dimensional point cloud data of equipment in the substation; Based on the three-dimensional environmental data, establish an obstacle model through an obstacle point cloud recognition algorithm, and generate an obstacle distribution map according to the obstacle model; the obstacles include power lines and equipment in the substation; Construct an electromagnetic interference field distribution map, which is generated by combining the electromagnetic radiation height of equipment in the substation and the equipment layout position information; Based on the obstacle distribution map and the electromagnetic interference field distribution map, use a global path planning algorithm to generate an initial UAV inspection path; During the flight of the UAV, dynamically update the obstacle distribution map and the electromagnetic interference field distribution map based on real-time sensor data, and use a local path optimization algorithm to adjust the flight path.

[0006] Preferably, as an implementable solution; in step S1, obtaining the three-dimensional environmental data of the substation includes: Use a lidar device installed on a drone to scan the internal environment of the substation and collect point cloud data; Preprocess the collected point cloud data, including point cloud denoising, ground segmentation, and point cloud coordinate transformation; Register the three-dimensional point cloud data of the equipment in the substation and the point cloud data of the power lines after preprocessing with the three-dimensional BIM model of the equipment in the substation to generate an obstacle model to be completed.

[0007] Preferably, as an implementable solution; in step S1, based on the three-dimensional environmental data, an obstacle model is established through an obstacle point cloud recognition algorithm, including: Extract the obstacle point cloud through a point cloud segmentation algorithm; the point cloud segmentation algorithm includes a segmentation method based on region growing; Use a machine learning algorithm to classify the extracted obstacle point cloud, divide the obstacles into static obstacles and dynamic obstacles, and record them as the classification results of the obstacles; Based on the classification results of the obstacles, generate an obstacle distribution map and assign a danger weight value to each obstacle; the danger weight value is determined by calling a preset database according to the classification results of the obstacles; the database is used to store the danger weight values corresponding to the classification types of each obstacle; The obstacle model is a three-dimensional model including an obstacle distribution map and marking the classification type of each obstacle.

[0008] Preferably, as an implementable solution; construct an electromagnetic interference field distribution map, including: Obtain the electromagnetic radiation characteristic information and equipment layout position information of each piece of equipment in the substation, and based on the electromagnetic radiation characteristic information of the equipment in the substation, calculate the electromagnetic field distribution (especially the electromagnetic radiation height) of the electromagnetic field generated by the equipment in the current substation; Perform grid processing on the calculated electromagnetic field distribution in the space of the substation, so that the space in the substation is divided into several grid units, and each grid unit contains an electromagnetic interference intensity value (obtained by direct matching in the database) and an electromagnetic radiation height; Overlay the gridded electromagnetic field distribution with the current obstacles in the obstacle distribution map and the current equipment layout position information to generate a comprehensive electromagnetic interference field distribution map; the electromagnetic interference field distribution map includes equipment layout position information, electromagnetic radiation height, and influence height; the influence height is the sum of the highest point height of the current equipment where the equipment layout position information is located and the electromagnetic radiation height affected under electromagnetic field interference.

[0009] Preferably, as an implementable solution; generate an initial inspection path based on the comprehensive electromagnetic interference field distribution map, including: In the comprehensive electromagnetic interference field distribution map, search for the global optimal path from the starting point to the target point with reference to the said hazard weight and influence height; Smooth the generated global optimal path, and optimize the continuous flyability of the path using a Bezier curve to obtain the smoothed path; Take the smoothed path as the initial inspection path of the UAV.

[0010] Preferably, as an implementable solution; in the comprehensive electromagnetic interference field distribution map, searching for the global optimal path from the starting point to the target point with reference to the said hazard weight and influence height specifically includes: Search for the global optimal path in the comprehensive electromagnetic interference field distribution map. When executing, use a dynamic weighted cost function combined with the obstacle hazard weight, electromagnetic interference intensity, and flight time for path search; at the same time, set the constraint conditions of the comprehensive cost function; Define the comprehensive cost function as: ; Where: is the cumulative sum of the obstacle d hazard weights of the path segment from node i to node j, ; ( is the hazard weight of the k-th grid cell); takes the maximum value of the electromagnetic interference intensity in the path segment, is the interference intensity of the k-th grid cell; are the node coordinates, and v is the cruising speed of the UAV; is the heuristic function, taking the Euclidean distance from node j to the end point; is the dynamically adjusted weight coefficient; The constraint condition of the said comprehensive cost function is: the height of the path node satisfies where is the influence height of node j (equipment height + radiation influence height); Preferably, as an implementable solution; smoothing the generated global optimal path, and optimizing the continuous flyability of the path using a Bezier curve to obtain the smoothed path specifically includes: Obtain the generated path point sequence and smooth it using a piecewise cubic Bezier curve to ensure that the path is continuously flyable and avoids the electromagnetic interference area; Determine the generation of control points: for each path point according to its influence height generate a vertical safety offset: z{k}=

[0011] Wherein: is a path point the corresponding influence height thereof; represents the actual flight height of the drone at the path point ; is the offset adjustment amount, which is a constant;

[0012] Perform curve fitting of the control points: Input the offset point set into the Bezier curve generator, and the objective function is: ; is the Bezier curve, is the smoothing factor; Generate an initial anti-interference inspection path; Overlay and verify the smoothed path with the electromagnetic interference field distribution map; During the overlay verification, if the ( is a preset threshold), then trigger a local replanning processing operation, and finally output a path that meets the following conditions: The first anti-interference optimization condition; ; The second anti-interference optimization condition: , the total risk weight is lower than the critical threshold; The third anti-interference optimization condition: .

[0013] Preferably, as an implementable solution; the first anti-interference optimization condition is a height constraint condition; the second anti-interference optimization condition is a cumulative risk weight constraint; the third anti-interference optimization condition is an electromagnetic interference intensity constraint.

[0014] The present invention provides an electronic device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for planning a substation drone inspection path based on a three-dimensional model.

[0015] The present invention provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it implements a method for planning a substation drone inspection path based on a three-dimensional model.

[0016] Compared with the prior art, the embodiments of the present invention have at least the following technical advantages: The present invention provides a method for processing the path planning of an unmanned aerial vehicle (UAV) for substation inspection based on a three-dimensional model. Starting from point cloud data, a lidar device is used to scan the equipment and power lines in the substation to obtain three-dimensional environmental data of the substation. The three-dimensional environmental data includes the point cloud data of the power lines obtained by lidar measurement and the three-dimensional point cloud data of the equipment in the substation. Based on the three-dimensional environmental data, an obstacle model is established through an obstacle point cloud recognition algorithm, and an obstacle distribution map is generated according to the obstacle model. The obstacles include power lines and equipment in the substation. An electromagnetic interference field distribution map is constructed, which is generated by combining the electromagnetic radiation height of the equipment in the substation and the equipment layout position information. S4. Based on the obstacle distribution map and the electromagnetic interference field distribution map, a global path planning algorithm is used to generate an initial UAV inspection path. Analyzing the above technical solution, it can be seen that the above method for processing the path planning of an unmanned aerial vehicle for substation inspection based on a three-dimensional model can accurately reconstruct the actual situation in the substation by obtaining three-dimensional environmental data. By using obstacle point cloud recognition and electromagnetic interference field modeling, the environmental risks are refined into two dimensions (physical and electromagnetic), providing a sufficient basis for path planning. On the basis of global path planning, real-time updates during flight and local path optimization are added to ensure that when the actual scenario changes, the flight path can still be dynamically adjusted to reduce risks.

[0017] Generally speaking, the above method for processing the path planning of an unmanned aerial vehicle for substation inspection based on a three-dimensional model realizes a complete UAV inspection path planning process from environmental data collection, obstacle and electromagnetic interference field modeling, global path planning to dynamic adjustment during flight. Considering physical obstacles and electromagnetic interference comprehensively, it ensures the safe and efficient execution of inspection tasks by the UAV in the substation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a schematic diagram of the main process of the method for processing the path planning of an unmanned aerial vehicle for substation inspection based on a three-dimensional model; Figure 2 It is a schematic diagram of a specific process for obtaining three-dimensional environmental data in the method for processing the path planning of an unmanned aerial vehicle for substation inspection based on a three-dimensional model; Figure 3It is a schematic diagram of a specific process for establishing an obstacle model through an obstacle point cloud recognition algorithm based on the three-dimensional environment data in the substation UAV inspection path planning and processing method based on a three-dimensional model; Figure 4 It is for constructing the electromagnetic interference field distribution in the substation UAV inspection path planning and processing method based on a three-dimensional model Figure 1 Schematic diagram of specific process; Figure 5 It is a schematic diagram of a specific process for generating an initial inspection path in the substation UAV inspection path planning and processing method based on a three-dimensional model; Figure 6 It is a schematic diagram of the principle structure of an electronic device. Specific implementation manners

[0020] Next, the technical solutions of the present invention will be clearly and completely described in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] Next, the present invention will be further described in detail through specific embodiments in conjunction with the accompanying drawings. Embodiment 1

[0022] As Figure 1 shown, the present invention provides a substation UAV inspection path planning and processing method based on a three-dimensional model, including the following operation steps: S1. Obtain the three-dimensional environment data in the substation, where the three-dimensional environment data includes the point cloud data of the power line measured by lidar and the three-dimensional point cloud data of the equipment in the substation; S2. Based on the three-dimensional environment data, establish an obstacle model through an obstacle point cloud recognition algorithm, and generate an obstacle distribution map according to the obstacle model; the obstacles include power lines and equipment in the substation; S3. Construct an electromagnetic interference field distribution map, which is generated by combining the electromagnetic radiation height of the equipment in the substation and the equipment layout position information; S4. Based on the obstacle distribution map and the electromagnetic interference field distribution map, use a global path planning algorithm to generate an initial UAV inspection path; S5. During the flight of the UAV, dynamically update the obstacle distribution map and the electromagnetic interference field distribution map based on real-time sensor data, and use a local path optimization algorithm to adjust the flight path (that is, receive and transmit data in real time, follow the changes in the environment and equipment in the sensor data detected by the sensor, dynamically start the update program, and finally use the local path optimization algorithm to adjust the flight path, that is, determine the path of the current change position according to the change position of the environment and equipment for S4 operation and update processing).

[0023] The present invention provides a method for planning and processing the inspection path of a substation UAV based on a three-dimensional model. Starting from point cloud data, it uses a lidar device to scan the equipment and power lines in the substation to obtain three-dimensional environment data of the substation. The three-dimensional environment data includes point cloud data of power lines obtained by lidar measurement and three-dimensional point cloud data of equipment in the substation; based on the three-dimensional environment data, an obstacle model is established through an obstacle point cloud recognition algorithm, and an obstacle distribution map is generated according to the obstacle model; the obstacles include power lines and equipment in the substation. Construct an electromagnetic interference field distribution map, which is generated by combining the electromagnetic radiation height of equipment in the substation and the equipment layout position information; S4. Based on the obstacle distribution map and the electromagnetic interference field distribution map, use a global path planning algorithm to generate an initial UAV inspection path. Analyzing the above technical solution, it can be seen that the above method for planning and processing the inspection path of a substation UAV based on a three-dimensional model can accurately reconstruct the actual situation in the substation by obtaining three-dimensional environment data. By using obstacle point cloud recognition and electromagnetic interference field modeling, the environmental risks are refined into two dimensions (physical and electromagnetic), providing a sufficient basis for path planning. Adding real-time updates and local path optimization during the flight on the basis of global path planning ensures that when the actual scenario changes, the flight path can still be dynamically adjusted to reduce risks.

[0024] Generally speaking, the above method for planning and processing the inspection path of a substation UAV based on a three-dimensional model realizes a complete UAV inspection path planning process from environmental data collection, obstacle and electromagnetic interference field modeling, global path planning to dynamic adjustment during flight. Considering physical obstacles and electromagnetic interference comprehensively, it ensures the safe and efficient execution of inspection tasks by the UAV in the substation.

[0025] As Figure 2 shown, in the above embodiment, obtaining the three-dimensional environment data of the substation in step S1 includes: S11. Use a lidar device installed on the UAV to scan the internal environment of the substation and collect point cloud data; S12. Preprocess the collected point cloud data, including point cloud denoising, ground segmentation, and point cloud coordinate transformation. S13. Register the pre - processed 3D point cloud data of the equipment in the substation and the point cloud data of the power lines with the 3D BIM model of the equipment in the substation to generate an obstacle model to be completed.

[0026] It should be noted that point cloud registration is the process of aligning multiple 3D point cloud data sets obtained from different perspectives or times into the same coordinate system. After the obstacle model to be completed undergoes the processing of S21 - S23, it is possible to add the extraction of obstacle point clouds and danger weights, etc., and finally generate an obstacle model containing an obstacle distribution map.

[0027] For the technical solutions of the above - mentioned S11 - S13 steps, it uses the lidar device carried by the UAV to obtain the original point cloud data inside the substation; registering the pre - processed point cloud data with the BIM model of the equipment can effectively make up for the possible deviations in a single data source and construct an accurate 3D environment model.

[0028] See Figure 3 , in the above - mentioned embodiment, in step S1, based on the 3D environment data, establishing an obstacle model through an obstacle point cloud recognition algorithm includes: S21. Extract obstacle point clouds through a point cloud segmentation algorithm (such as a region - growing - based segmentation method); the point cloud segmentation algorithm includes a region - growing - based segmentation method. S22. Use a machine learning algorithm to classify the extracted obstacle point clouds, divide the obstacles into static obstacles and dynamic obstacles, and record them as the classification results of the obstacles. S23. Based on the classification results of the obstacles, generate an obstacle distribution map and assign a danger weight to each obstacle; the danger weight is determined by calling a preset database according to the classification results of the obstacles; the database is used to store the danger weights corresponding to the classification types of each obstacle. Based on the obstacle model is a 3D model containing an obstacle distribution map and marking the classification type of each obstacle.

[0029] For the above - mentioned technical solutions, it uses a region - growing - based point cloud segmentation algorithm to extract obstacles in the environment; distinguishes obstacles into static and dynamic through machine learning methods, so as to clarify the different risks that may be caused by each; at the same time, it calls the hazard weights according to the preset database, so that the risk of each obstacle can be quantified, and finally generates an obstacle distribution map based on this, providing a technical basis for subsequent global path search.

[0030] See Figure 4 , in the above - mentioned embodiment, in step S3, constructing an electromagnetic interference field distribution map includes: S31. Obtain the electromagnetic radiation characteristic information and equipment layout position information of the equipment in each substation. Based on the electromagnetic radiation characteristic information of the equipment in the substation, calculate the electromagnetic field distribution (especially the electromagnetic radiation height) of the electromagnetic field generated by the equipment in the current substation. S32. Perform grid processing on the calculated electromagnetic field distribution in the space of the substation, so that the space in the substation is divided into several grid cells. Each grid cell contains an electromagnetic interference intensity value (obtained by direct matching in the database) and an electromagnetic radiation height. S33. Superimpose the grid-based electromagnetic field distribution with the current obstacles in the obstacle distribution map and the current equipment layout position information to generate a comprehensive electromagnetic interference field distribution map. The electromagnetic interference field distribution map includes equipment layout position information, electromagnetic radiation height, and influence height. The influence height is the sum of the highest point height of the current equipment where the equipment layout position information is located and the influence on the electromagnetic radiation height under electromagnetic field interference.

[0031] In the above technical solution, through grid processing, the continuous electromagnetic field distribution is changed into discrete grid cells, and each cell contains specific interference values. The grid-based interference data is superimposed with the obstacle distribution and equipment information, thereby constructing a comprehensive map reflecting the actual interference situation, providing key reference information for globally searching for the optimal path. The superposition of this map with the obstacle distribution map enables path planning to not only consider physical obstacles but also effectively avoid potential risk areas caused by electromagnetic interference.

[0032] See Figure 5 , in the above embodiment, in step S4, generating an initial inspection path based on the comprehensive electromagnetic interference field distribution map includes: S41. In the comprehensive electromagnetic interference field distribution map, search for the globally optimal path from the starting point to the target point with reference to the risk weight value and influence height. S42. Smooth the generated globally optimal path, and use a Bezier curve to optimize the continuous flyability of the path to obtain a smoothed path. S43. Use the smoothed path as the initial inspection path of the unmanned aerial vehicle.

[0033] In the above technical solution, a Bezier curve is used to smooth the initial path, making the path turn more naturally and the curve smoother, meeting the physical requirements of the continuous flight of the unmanned aerial vehicle, thereby minimizing the safety hazards caused by path discontinuity.

[0034] In the global optimization process, the obstacle risk weight value and the influence height of electromagnetic interference are comprehensively considered, so that the searched path optimizes the flight time and distance as much as possible on the basis of minimizing risks.

[0035] In the above embodiments, in the comprehensive electromagnetic interference field distribution map, the global optimal path from the starting point to the target point is searched with reference to the risk weight and the influence height, specifically including: S411. Search for the global optimal path in the comprehensive electromagnetic interference field distribution map. When executing, a dynamic weighted cost function is used to combine the obstacle risk weight, the electromagnetic interference intensity, and the flight time for path search; at the same time, the constraint conditions of the comprehensive cost function are set. Define the comprehensive cost function as: ; Where: is the cumulative sum of the obstacle d risk weights of the path segment from node i to node j. ; ( is the risk weight of the k-th grid cell); , take the maximum value of the electromagnetic interference intensity in the path segment. is the interference intensity of the k-th grid cell; are the node coordinates, and v is the cruising speed of the UAV; is the heuristic function, taking the Euclidean distance from node j to the end point; is the dynamically adjusted weight coefficient; The constraint condition of the comprehensive cost function is: the height of the path node satisfies , where is the influence height of node j (equipment height + radiation influence height); For each path point, by calculating the vertical safety offset, ensure that the path meets the equipment influence height requirements in terms of height; use Bezier curve for smooth fitting, and then superimpose and verify the generated path with the comprehensive electromagnetic interference field distribution. If it is detected that the local risk exceeds the preset threshold, trigger local replanning, and use multi-dimensional anti-interference optimization conditions to correct the path again.

[0036] In the above technical solution, the dynamic weighted cost function is used to quantify the risks and interference intensities of obstacles in the path, ensuring that risk points can be fully considered during the search process; at the same time, a heuristic function (such as Euclidean distance) from the current node to the end point is introduced to ensure that the path search takes into account both distance and risk; at the same time, the path node height constraint is set to prevent the risks brought by low flight or excessive flight, realizing double guarantees of flight height and safety, ensuring that the UAV flight path can avoid high-risk areas while meeting the requirements of flight energy consumption and time.

[0037] In the above embodiments, the generated globally optimal path is smoothed, and the continuous flyability of the path is optimized by using a Bezier curve to obtain a smoothed path, which specifically includes: S421. Obtain the generated sequence of path points , and smooth it with a piecewise cubic Bezier curve to ensure that the path is continuously flyable and avoid the electromagnetic interference area; Determine the generation of control points: For each path point , generate a vertical safety offset according to its influence height : z{k}=

[0038] where: is the influence height corresponding to the path point ; represents the actual flight height of the UAV at the path node (or position point) ; is the offset adjustment amount, which is a constant;

[0039] Execute the curve fitting of the control points: Input the offset point set into the Bezier curve generator, and the objective function is: is the Bezier curve, is the smoothing factor; S422. Generate an initial anti-interference inspection path; superimpose and verify the smoothed path with the electromagnetic interference field distribution map; during the superimposed verification, if the ( is a preset threshold), then trigger the local replanning processing operation, and finally output a path that meets the height constraint condition, the cumulative hazard weight constraint, and the electromagnetic interference intensity constraint condition: wherein, the first anti-interference optimization condition is the height constraint condition; the second anti-interference optimization condition is the cumulative hazard weight constraint; the third anti-interference optimization condition is the electromagnetic interference intensity constraint; The first anti-interference optimization condition; ; The second anti-interference optimization condition: , the total hazard weight is lower than the critical threshold; The third anti-interference optimization condition: .

[0040] In the height constraint condition, : represents the actual flight height of the UAV at the path node (or position point) ; : is the logical symbol "for all ", indicating that this constraint applies to every node on the path; In the height constraint, the height is composed of the superposition of the physical height of the device and the height affected by electromagnetic radiation (see S33), representing the minimum safe flight height of the UAV at position . Ensure that the UAV always flies above the influence range of the electromagnetic interference field to avoid communication interruption or sensor failure caused by electromagnetic interference. At the same time, it can also be dynamically adjusted according to this: if a certain path segment does not meet this condition, the path needs to be adjusted by vertical climbing or horizontal bypassing.

[0041] Cumulative hazard weight value constraint ; In the above cumulative hazard weight value constraint condition, the hazard weight of the path segment is the sum of the obstacle hazard weights of all grid cells on it ), reflecting the comprehensive risk of this path segment (such as being close to high-voltage equipment, easy-to-collide areas, etc.). The preset critical threshold is preset according to the task safety requirements, for example, the high-voltage area is relatively low, and it can be relaxed in open areas. Dynamically adjust the path through the cost function (S41), and preferentially select the path segment with a lower hazard weight.

[0042] In the electromagnetic interference intensity constraint condition, the electromagnetic interference intensity is: take the maximum value of the electromagnetic interference intensity of all grid cells in the path segment to avoid the UAV getting out of control caused by local strong interference. The threshold ( ) is set according to the UAV's anti-electromagnetic interference ability (such as the tolerance threshold of the communication module).

[0043] If a certain path segment , trigger local replanning, and avoid the strong interference area by increasing the height or bypassing.

[0044] Through the setting of multiple constraints in the above technical solution, the system can eliminate unsafe options at the initial stage of path planning, ensuring that the final inspection path meets all safety standards at the same time. The height constraint ensures that the UAV always maintains a sufficient vertical safety distance during flight; the cumulative hazard weight value constraint ensures that the obstacle risk of the overall path is within a controllable range; the electromagnetic interference intensity constraint directly avoids high-interference areas and reduces the risk of equipment errors or signal interruption caused by complex electromagnetic environments; It is decomposed into three independent conditions, which is convenient for step-by-step detection and layer-by-layer optimization during path planning and local replanning, making the entire planning process more rigorous and safe. Embodiment 2

[0045] Based on the same inventive concept as the above method embodiments, an embodiment of the present invention further provides a three-dimensional model-based substation UAV inspection path planning and processing system for implementing the above three-dimensional model-based substation UAV inspection path planning and processing method of the present invention. Since the principle of solving problems in this electronic device embodiment is similar to the method, it at least has all the beneficial effects brought by the technical solutions of the above embodiments, and will not be elaborated here one by one.

[0046] An embodiment of the present invention provides a three-dimensional model-based substation UAV inspection path planning and processing system, including: An acquisition module that obtains three-dimensional environmental data inside the substation, and the three-dimensional environmental data includes point cloud data of power lines obtained by lidar measurement and three-dimensional point cloud data of equipment inside the substation; A modeling module that, based on the three-dimensional environmental data, establishes an obstacle model through an obstacle point cloud recognition algorithm and generates an obstacle distribution map according to the obstacle model; the obstacles include power lines and equipment inside the substation; A generation processing module that constructs an electromagnetic interference field distribution map, and the electromagnetic interference field distribution map is generated by combining the electromagnetic radiation height of equipment inside the substation and the equipment layout position information; A path output module that, based on the obstacle distribution map and the electromagnetic interference field distribution map, uses a global path planning algorithm to generate an initial UAV inspection path; A path update module that, during the flight of the UAV, dynamically updates the obstacle distribution map and the electromagnetic interference field distribution map based on real-time sensor data, and adjusts the flight path using a local path optimization algorithm. Embodiment Three

[0047] The present invention provides an electronic device, as Figure 6 shown. This computing device may include a storage component 31 and a processing component 32; the storage component 31 stores one or more computer instructions, and among them, the one or more computer instructions are called and executed by the processing component 32 to implement a three-dimensional model-based substation UAV inspection path planning and processing method provided in Embodiment One.

[0048] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above method. Embodiment Four

[0049] The present invention provides a computer storage medium storing a computer program, which when executed by a computer, implements a method for processing the path planning of an unmanned aerial vehicle (UAV) for substation inspection based on a three-dimensional model provided in Embodiment 1.

[0050] In summary, the present invention provides a method for processing the path planning of an unmanned aerial vehicle (UAV) for substation inspection based on a three-dimensional model. Through a full-process technical solution that hierarchically and gradually constructs from data acquisition, preprocessing, obstacle and interference modeling to global and local path planning based on multi-dimensional evaluation indicators, the fine and safe planning of the UAV inspection path in the complex environment of the substation is finally achieved. The above method can effectively cope with the possible physical obstacles and electromagnetic interference in the environment, providing a high-security and high-reliability flight path planning guarantee for UAV inspection.

[0051] 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 them; those of ordinary skill in the art can modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A substation drone inspection path planning and processing method based on a three-dimensional model, characterized in that: The steps are as follows: Acquire three-dimensional environmental data in the substation, wherein the three-dimensional environmental data includes point cloud data of power lines and three-dimensional point cloud data of equipment in the substation obtained by laser radar measurement; Based on the three-dimensional environment data, an obstacle model is established through an obstacle point cloud recognition algorithm, and an obstacle distribution map is generated according to the obstacle model; The obstacles include power lines and equipment in substations; Constructing an electromagnetic interference field distribution map, wherein the electromagnetic interference field distribution map is generated by combining the electromagnetic radiation height of the equipment in the substation and the equipment layout location information; Based on the obstacle distribution map and the electromagnetic interference field distribution map, an initial UAV inspection path is generated using a global path planning algorithm; During the flight of the UAV, the obstacle distribution map and electromagnetic interference field distribution map are dynamically updated based on real-time sensor data, and the flight path is adjusted using a local path optimization algorithm.

2. The substation drone inspection path planning and processing method based on a three-dimensional model according to claim 1 is characterized in that: Obtain 3D environmental data within the substation, including: Use a lidar device mounted on a drone to scan the substation's internal environment and collect point cloud data; Preprocess the collected point cloud data, including point cloud denoising, ground segmentation and point cloud coordinate conversion; The pre-processed 3D point cloud data of the substation equipment and the point cloud data of the power lines are aligned with the 3D BIM model of the substation equipment to generate the obstacle model to be completed.

3. The substation drone inspection path planning and processing method based on a three-dimensional model according to claim 2 is characterized in that: Based on the three-dimensional environment data, an obstacle model is established by an obstacle point cloud recognition algorithm, including: Extracting obstacle point clouds through a point cloud segmentation algorithm; the point cloud segmentation algorithm includes a segmentation method based on region growing; Use machine learning algorithms to classify the extracted obstacle point clouds, divide the obstacles into static obstacles and dynamic obstacles, and record them as the obstacle classification results; Based on the classification results of the obstacles, an obstacle distribution map is generated and a danger weight is assigned to each obstacle; the danger weight is determined by calling a preset database according to the classification results of the obstacles; the database is used to store the danger weight corresponding to the classification type of each obstacle; The obstacle model is a three-dimensional model including an obstacle distribution map and a classification type marking each obstacle.

4. The substation drone inspection path planning and processing method based on a three-dimensional model according to claim 3 is characterized in that: Construct an electromagnetic interference field distribution map, including: Obtain electromagnetic radiation characteristic information and equipment layout location information of each substation equipment, and calculate the electromagnetic field distribution of the electromagnetic field generated by the current substation equipment based on the electromagnetic radiation characteristic information of the equipment in the substation; Gridding the calculated electromagnetic field distribution in the space inside the substation, so that the space inside the substation is divided into a plurality of grid units, each grid unit including an electromagnetic interference intensity value and an electromagnetic radiation height; The gridded electromagnetic field distribution is superimposed with the current obstacles in the obstacle distribution map and the current equipment layout location information to generate a comprehensive electromagnetic interference field distribution map; the electromagnetic interference field distribution map includes the equipment layout location information and the electromagnetic radiation height and the impact height; the impact height is the sum of the highest point height of the current equipment where the equipment layout location information is located and the height of the electromagnetic radiation affected under electromagnetic field interference.

5. The substation drone inspection path planning and processing method based on a three-dimensional model according to claim 4 is characterized in that: Generate an initial inspection path based on a comprehensive electromagnetic interference field distribution map, including: In the comprehensive electromagnetic interference field distribution map, a global optimal path from a starting point to a target point is searched with reference to the hazard weight and the impact height; The generated global optimal path is smoothed, and the continuous flyability of the path is optimized using Bezier curve to obtain the smoothed path; The smoothed path is used as the initial inspection path of the UAV.

6. The substation drone inspection path planning and processing method based on a three-dimensional model according to claim 4 is characterized in that: In the comprehensive electromagnetic interference field distribution map, the global optimal path from the starting point to the target point is searched with reference to the hazard weight and the impact height, specifically including: Search for the global optimal path in the comprehensive electromagnetic interference field distribution map. When executing, a dynamic weighted cost function is used to combine the obstacle hazard weight, electromagnetic interference intensity and flight time to search for the path; at the same time, set the constraints of the comprehensive cost function; The comprehensive cost function is defined as: ； in: is the cumulative sum of the danger weights of obstacles d on the path segment from node i to node j, ; is the danger weight of the kth grid cell; , take the maximum value of the electromagnetic interference intensity in the path segment, is the interference intensity of the kth grid unit; is the node coordinate, v is the cruising speed of the UAV; is the heuristic function, taking the Euclidean distance from node j to the end point; To dynamically adjust the weight coefficient; The constraint condition of the comprehensive cost function is: the height of the path node satisfies ,in is the influence height of node j.

7. The substation drone inspection path planning and processing method based on a three-dimensional model according to claim 6 is characterized in that: The generated global optimal path is smoothed, and the continuous flyability of the path is optimized using Bezier curves to obtain the smoothed path, including: Get the generated path point sequence ,Using piecewise cubic Bezier curve smoothing to ensure the path is continuous and flyable and avoid electromagnetic interference areas; Determine control point generation: for each path point , according to its impact height Generate vertical clearance offset: z{k}= ; in: For waypoints The corresponding impact height; Indicates that the drone is at a path point The actual flight altitude at is the offset adjustment, which is a constant; Perform curve fitting of the control points: Enter the Bezier curve generator, the objective function is: ; is a Bezier curve, is the smoothing factor; Generate an initial anti-interference inspection path; superimpose the smoothed path with the electromagnetic interference field distribution map for verification; in the superposition verification, if the current segment of the path , then the local replanning process is triggered, and the path that meets the following conditions is finally output: is the preset threshold; The first anti-interference optimization condition; ; The second anti-interference optimization condition: , the total risk weight is lower than the critical threshold; The third anti-interference optimization condition: .

8. The substation drone inspection path planning and processing method based on a three-dimensional model according to claim 7 is characterized in that: The first anti-interference optimization condition is a height constraint condition; the second anti-interference optimization condition is a cumulative hazard weight constraint; and the third anti-interference optimization condition is an electromagnetic interference intensity constraint.

9. An electronic device, characterized in that: It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a substation drone inspection path planning and processing method based on a three-dimensional model as described in any one of claims 1-8.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a substation drone inspection path planning and processing method based on a three-dimensional model is implemented as described in any one of claims 1 to 8.

Images