A method for positioning obstacles in the transport of fan blades based on aerial survey

Abstract

This invention relates to the fields of wind power engineering technology, transportation safety, and geographic information systems. It discloses a method for locating air obstacles during wind turbine blade transportation based on aerial survey topographic data. The method includes: constructing a 3D terrain model and a 3D vehicle model; obtaining building outline vector data and individual tree point cloud data from the 3D terrain model; obtaining individual tree point cloud data from laser point cloud data; simulating the transportation process to generate a sweep envelope; generating a sweep line dataset covering the entire sweep envelope; performing longitudinal section collision analysis; spatially overlaying the boundary line of the hazard area with the building outline vector data and individual tree point cloud data to determine the target object for obstacle removal; and performing 3D visualization verification and display to locate air obstacles during wind turbine blade transportation. According to the above technical solution, the passability of different routes and lifting schemes can be simulated in advance before transportation, identifying all risk points and providing strong data support for route optimization and transportation plan formulation.

Description

Technical Field

[0001] This invention relates to the fields of wind power engineering technology, transportation safety and geographic information system technology, and more specifically, to a method for locating air obstacles during wind turbine blade transportation based on aerial survey topographic results. Background Technology

[0002] As wind power development extends into complex mountainous areas, the transportation of ultra-long wind turbine blades has become a major challenge for project construction. When transport vehicles travel on winding mountain roads, the lifted blades create a huge dynamic sweeping space in the air, which can easily lead to collisions with trees, buildings, slopes, and even mountainsides on both sides of the road.

[0003] Currently, the industry's assessment of transport obstacles mainly relies on personnel's field experience and simple measuring tools (such as laser rangefinders and tape measures). This method has significant drawbacks: First, the assessment is highly subjective and lacks accuracy; relying on personnel's experience cannot accurately quantify the spatial relationship between blades and complex terrain features. Second, determining transport plans is inefficient and has limited coverage; for long-distance routes and dense forests, manual surveys are difficult and time-consuming. Third, the assessment results are not intuitive and lack guidance; they usually only provide a rough "danger zone" and cannot accurately locate obstacles, let alone pinpoint the specific trees to be felled or the building parts to be demolished. This often leads to two adverse consequences: first, the clearance area is too large, causing unnecessary vegetation damage and demolition compensation, increasing project costs and negatively impacting the ecological environment; second, incomplete clearance leaves hidden danger points, which may lead to collisions during transport, causing transport interruptions, equipment damage, or even casualties.

[0004] Currently, although there are methods for topographic mapping using 3D laser scanning or oblique photography, and there is also general vehicle routing software, there is a lack of specialized solutions that systematically integrate high-precision terrain data, accurate vehicle and cargo models, transportation motion simulation, and quantitative spatial conflict analysis, and directly serve the decision-making process for clearing obstacles in the transportation of wind turbine blades.

[0005] Therefore, there is an urgent need for a scientific, precise, and efficient technical means to accurately locate and analyze air obstacles during blade transportation. Summary of the Invention

[0006] To achieve the above objectives, this application provides a method for locating air obstacles during wind turbine blade transportation based on aerial survey topographic results, comprising the following steps: Construct a 3D terrain model and a 3D vehicle model; obtain building outline vector data and single-tree point cloud data from the 3D terrain model; the 3D terrain model includes: digital elevation model, digital surface model, digital orthophoto, and high-precision real-scene 3D model; the 3D vehicle model includes: 2D top view outline model, 2D side view outline model, and 3D detailed model; the 2D side view outline model also includes a series of discrete blade lifting angle state models. Simulate the transportation process and generate the sweep envelope range; Within the sweep envelope, generate a sweep line dataset covering the entire sweep range; Perform longitudinal section collision analysis; longitudinal section collision analysis is used to determine the mileage intervals with collision risks or hidden dangers on the sweep line data, and generate the boundary lines of the hidden danger areas based on the mileage intervals; Spatial overlay analysis was performed on the boundary line of the hazard area with building outline vector data and single tree point cloud data to determine the target objects for obstacle removal; the target objects included all buildings and multiple trees that were wholly or partially located in the hazard area. Three-dimensional visualization verification and display were conducted to locate air obstacles in the transport of wind turbine blades.

[0007] The process of generating the swept envelope includes: loading a digital orthophoto as a base map, placing a two-dimensional top-view profile model of the vehicle at the planned starting point, and simulating the straight-line and turning motion of the vehicle along the road centerline or a predetermined trajectory through interactive operation. During the interactive operation, the motion trajectory points of the wheels, body, blades and feature contour points during the vehicle's movement and rotation are recorded. The motion trajectory points are connected to form the dynamic sweep envelope range line of the blades on the entire transportation path, thus constituting the sweep envelope range.

[0008] Methods for generating sweep line datasets include: With the center of the blade flange as the origin, a series of rays are emitted along the trajectory line, starting from the sweep start edge and rotating towards the sweep end edge at fixed angular intervals or adaptive angular steps, forming a set of sweep line datasets covering the entire sweep range.

[0009] The longitudinal section collision analysis includes the following steps: For each sweep line in the sweep line dataset, longitudinal profile sections are cut to generate ground longitudinal profile lines and top longitudinal profile lines of ground features; Perform dynamic correction of vehicle attitude; By comparing the vertical distance relationship between each point on the blade outline of the vehicle's two-dimensional side profile model after dynamic correction of vehicle attitude and the ground longitudinal profile line and the top longitudinal profile line of ground features, collision risk assessment is performed to determine continuous mileage intervals. Map all the hazard mileage intervals recorded on the sweep line back to the plan view, representing them as multiple short lines along the sweep line direction; the short lines represent the hazard status as hazard segments; use a geometric algorithm to sequentially connect the inner and outer endpoints of the hazard segments on adjacent sweep lines to form the boundary line of the hazard area.

[0010] The vehicle attitude dynamic correction includes the following steps: The starting point of the mileage is determined based on the position of the vehicle's two-dimensional side profile model on the plan view, and the current position of the vehicle is defined as the starting point of the sweep line. ; Adjust the target position of the vehicle's two-dimensional side profile model so that the tire contact point and the ground elevation profile line precisely coincide at the target position. Based on the starting point of the sweep line The slope value is used to rotate the entire two-dimensional side profile model of the vehicle so that the tilt angle of the chassis is consistent with the slope.

[0011] The target location is determined by the elevation and slope of the corresponding ground longitudinal profile; this includes: obtaining the center of the vehicle's rear wheels at... The ground elevation at that location is expressed as: ; calculate The slope angle of the terrain at that location α terrain slope angle α Obtained using the central difference method, it is expressed as: , where ΔS is the minute mileage increment.

[0012] When rotating the entire 2D side view profile model of a vehicle, the 2D side view profile model of the vehicle is first defined as follows: The corresponding blade lift angle is β, and a geometric transformation is performed with the grounding reference point as the reference. The geometric transformation process includes translation and rotation. During translation, the grounding point is aligned with the coordinate system. Coincident; during rotation, the tilt angle of the model's chassis axis is equal to α The geometrically transformed 2D side profile model of the vehicle is represented as: .

[0013] Further, the collision risk assessment includes: Set a safe distance threshold If the outline of the blade is lower than the top longitudinal profile line of the ground feature or the ground longitudinal profile line, then "intersection" occurs and is judged as "collision". If the blade outline is higher than the top longitudinal profile line of the ground feature, but the clearance is less than the safety threshold, it is judged as "having a hidden danger".

[0014] Furthermore, 3D visualization verification and display include: Import the vector data and target object location information within the boundary of the potential hazard area into a 3D visualization engine or platform, and perform spatial registration and overlay based on a high-precision real-scene 3D model. The superimposed hazard area is colored semi-transparently, and the target object is highlighted. Select a representative sweep line, place the corresponding detailed 3D model of the vehicle at the starting position on the plane, and adjust the lifting angle of the blades according to the working conditions.

[0015] According to this invention, obstacle assessment can be elevated from an experience-based level to a quantitative analysis level, improving survey efficiency; it enables precise location and clear targeting, avoiding indiscriminate, one-size-fits-all obstacle removal and maximizing the protection of the ecological environment. Through this invention, the passability of different routes and lifting schemes can be simulated in advance before transportation, identifying all risk points and providing strong data support for route optimization and transportation plan development. Attached Figure Description

[0016] Figure 1 This is a step diagram of the method for locating air obstacles during wind turbine blade transportation based on aerial survey topographic results, according to an embodiment of the present invention. Figure 2 This is a data processing relationship diagram of the wind turbine blade transport obstacle positioning method provided in the embodiments of the present invention; Figure 3 This is a schematic diagram of a two-dimensional side view contour model of a blade transport vehicle provided according to an embodiment of the present invention; Figure 4 This is a schematic diagram of simulating vehicle steering and generating a blade sweep envelope range line on a digital orthophoto image according to an embodiment of the present invention. Figure 5 This is a schematic diagram of longitudinal section collision analysis provided according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the boundary line of the potential hazard area provided in an embodiment of the present invention; Figure 7 This is a three-dimensional visualization effect diagram of the potential hazard area provided by an embodiment of the present invention. Detailed Implementation

[0017] The specific implementation of the present invention will now be described in detail with reference to the accompanying drawings.

[0018] The wind turbine blade transport obstacle positioning method provided by this invention is as follows: Figure 1 As shown, it includes the following steps: Step S100: Construct a 3D terrain model and a 3D vehicle model; obtain building outline vector data and single-tree point cloud data from the 3D terrain model; When constructing the 3D terrain model, aerial surveys are conducted using UAVs equipped with oblique photography cameras and LiDAR to collect high-resolution images and point cloud data of wind farm areas and planned transportation corridors. The high-resolution images and point cloud data are processed to generate the 3D terrain model, which includes: a digital elevation model (DEM) representing the terrain undulations, a digital surface model (DSM) representing the surface and the tops of above-ground objects, a digital orthophoto (DOM) representing the planar base map, and a high-precision real-scene 3D model (such as OSGB format). Simultaneously, based on the high-resolution images and point cloud data, building outlines are extracted to generate building polygon vector layers for extracting building outline vector data. After filtering and classifying the point cloud data, density-based clustering (such as DBSCAN) or deep learning-based tree segmentation algorithms are applied to identify and locate each individual tree, generating tree point vector layers for extracting individual tree point cloud data.

[0019] When constructing the 3D model of the vehicle, a parametric vehicle-blade model library is established based on the actual design drawings and dimensional parameters of the blade transport vehicle and the blades. Specifically, the blade transport vehicle consists of a tractor and a dedicated blade transfer bridge, and the dimensional parameters include specific data such as vehicle length, vehicle width, wheelbase, track width, blade length, center of rotation, and hydraulic lifting range. The vehicle-blade model library includes: a 2D top-view profile model of the vehicle for planar path simulation, a 2D side-view profile model of the vehicle for longitudinal section collision analysis, and a detailed 3D model of the vehicle with blade lifting animation for the final 3D scene display. The 2D side-view profile model of the vehicle also includes a series of discrete blade lifting angle state models, such as... Figure 3 As shown, the blade lifting angle is in 5° intervals, with a support range from 10° to 35°.

[0020] Taking a 300-meter-long blade transport road in a mountainous wind farm as an example: When constructing the 3D terrain model, a Pegasus D2000 UAV equipped with a CAM3000 aerial survey camera and a LiDAR2000 lidar was used to conduct aerial surveys of the road and a 150-meter radius on both sides, generating DOM, DEM, DSM, and a real-scene 3D model with a resolution of 3 cm. The point cloud data was denoised, and ground, vegetation, and buildings were classified and identified. A point cloud-based clustering algorithm (such as DBSCAN) was used for individual tree segmentation, identifying approximately 300 independent trees and obtaining their center point coordinates and tree height attributes. When constructing the 3D vehicle model, based on the technical parameters of the 6-axle blade transport vehicle and the 100-meter-long blades, a top-view outline and side-view outlines with blade lifting angles of 15°, 20°, 25°, and 30° were created in AutoCAD as the vehicle's 2D top-view outline model and 2D side-view outline model.

[0021] Step S110: Simulate the transportation process and generate the sweep envelope range; Transportation process simulation is achieved using GIS software or a self-developed simulation platform, combined with digital orthophotos and a two-dimensional top-view outline model of the vehicle (e.g.) Figure 2 As shown in step S201, the steps include: loading a digital orthophoto as a base map, placing a two-dimensional top-view contour model of the vehicle at the planned starting point, and simulating the vehicle's straight-line and turning movements along the road centerline or a predetermined trajectory through interactive operation; during the interactive operation, recording the motion trajectory points of the wheels, body, blades, and feature contour points during the vehicle's movement and rotation, and connecting the motion trajectory points to form the dynamic sweep envelope range line of the blades along the entire transportation path, constituting the sweep envelope range (e.g., ...). Figure 4 (As shown). Generally, the shape of the sweep envelope is approximately fan-shaped.

[0022] The transportation process simulation can generate multiple route options.

[0023] Step S120: Within the sweep envelope, with the center of the blade flange as the origin, along the trajectory line, starting from the sweep start edge, a series of rays are rotated and emitted towards the sweep end edge at fixed angular intervals (such as 3 degrees or 1 degree) or adaptive angular steps to form a set of sweep line datasets covering the entire sweep range.

[0024] Step S130: Perform longitudinal section collision analysis, that is: determine the mileage intervals with collision risks or hidden dangers on the sweep line data, and generate the boundary line of the hidden danger area based on the mileage intervals; Longitudinal section collision analysis is achieved based on the digital elevation model and digital surface model in the 3D terrain model, and the 2D side profile model of the vehicle (e.g.) Figure 2 As shown in step S202), the specific process includes the following steps: 1) For each sweep line in the sweep line dataset, spatial analysis methods are used to perform longitudinal profile sections, generating ground longitudinal profile lines and top longitudinal profile lines of ground features. The ground longitudinal profile lines are generated by extracting elevation values ​​along the sweep lines using a digital elevation model to reflect terrain undulations. Simultaneously, such as... Figure 5 As shown, the top profile line of the ground features is generated by extracting elevation values ​​along the same sweep line based on the digital surface model, and is used to reflect the top height of the ground and above-ground features (such as tree canopies and roofs).

[0025] When performing longitudinal section cutting, the section is cut along the sweep line based on the DEM. i Partially, the ground elevation profile was obtained. Where S is the mileage along the sweep line; based on the DSM, the elevation profile of the top of the land cover is obtained by cutting along the same sweep line. (Including tree canopy, roof, etc.).

[0026] 2) Perform dynamic vehicle attitude correction; When overlaying a 2D side-view profile model of a vehicle with a 3D terrain model, simply placing the 2D side-view profile model flat will produce significant errors in mountainous scenarios. Therefore, this invention employs a dynamic attitude correction algorithm for dynamic vehicle attitude correction.

[0027] First, based on the position of the vehicle's two-dimensional side profile model on the plan view, determine its starting point on the longitudinal profile view, and define the vehicle's current position as the sweep line starting point. ; Then, adjust the target position of the vehicle's two-dimensional side profile model so that the vehicle's tire contact point and the ground elevation section line at that target position are precisely aligned. The target location is determined by the elevation value and slope (i.e., tangent slope) of the corresponding ground longitudinal profile; the center of the vehicle's rear wheel (or a designated ground contact reference point) is obtained. The ground elevation at that location is expressed as: ;calculate The slope angle of the terrain at that location α terrain slope angle α It can be obtained through the central difference method, expressed as: , where ΔS is a small mileage increment; Finally, based on the starting point of the sweep line The slope value is used to rotate the entire two-dimensional side profile model of the vehicle so that the tilt angle of its chassis (or wheel axle) is consistent with the slope, thereby realistically simulating the vehicle's driving posture on a slope. When rotating the entire 2D side view profile model of a vehicle, the 2D side view profile model of the vehicle is first defined as follows: The corresponding blade lift angle is β. A geometric transformation is performed using its grounding reference point as a reference. The geometric transformation process includes translation and rotation. During translation, the grounding point is aligned with the coordinate system. Coincident; during rotation, make the tilt angle of the model's chassis axis (or horizontal reference line) equal to... α The geometrically transformed 2D side profile model of the vehicle is represented as: Geometric transformations ensure that the attitude of the vehicle's two-dimensional side profile model strictly matches the actual terrain slope.

[0028] 3) Compare the vertical distances between each point on the blade outline of the vehicle's two-dimensional side view profile model after dynamic vehicle attitude correction and the ground longitudinal profile line and the top longitudinal profile line of ground features, and perform the following: Figure 2 The collision risk assessment shown in step S203 is as follows: First, a safe distance threshold is set. (e.g., 10 meters): If the blade outline is lower than the top longitudinal profile line of the ground feature or the ground longitudinal profile line, i.e., "intersection" occurs, it is judged as "collision"; if the blade outline is higher than the top longitudinal profile line of the ground feature, but the clearance is less than the safety threshold, it is judged as "potential hazard exists"; record the starting and ending mileage values ​​of "collision" or "potential hazard exists" on each sweep line.

[0029] In specific implementation, define This is the longitudinal profile of the ground. The top longitudinal profile of the ground features; the model Blade profile function and and Perform spatial relationship comparison; for any point on the sweep line S The decision logic is as follows: Define the vertical distance difference: , Status ← "Collision" Status ← “Risk” else Status ← “Safe” Scan the entire sweep line, identify and record the continuous mileage intervals corresponding to all collision and potential hazard conditions. , ].

[0030] 4) Map all hazard mileage intervals recorded on the sweep lines back onto the planar map, representing them as multiple short lines along the sweep line direction. These short lines can serve as hazard segments to express the hazard status. Use geometric algorithms (such as convex hull algorithm, polygon buffer algorithm, or spline curve fitting) to sequentially connect the inner and outer endpoints of hazard segments on adjacent sweep lines, ultimately forming one or more closed polygons, constituting the boundary line of the hazard area (e.g., Figure 6 As shown). The area within the boundary line of the hazard area is the hazard area (e.g., ...). Figure 2 (See step S204).

[0031] Step S140: Perform spatial overlay analysis on the boundary line of the hazard area with the building outline vector data and single-tree point cloud data to determine the target object for obstacle removal; The target objects include all buildings and multiple trees that are wholly or partially located in the hazard area; in this step, identification attributes are added to the target objects, and these marked features are the specific targets that need to be checked, pruned or removed.

[0032] In the case provided by this invention, the movement of a vehicle along the centerline of a road is simulated in dedicated simulation software to generate a swept envelope. Eighteen swept lines are generated at 5-degree intervals. Each line is sectioned in the DEM / DSM to generate a longitudinal profile, which is then analyzed in conjunction with a 30° vehicle side view model. During the analysis, for each longitudinal profile, the vehicle position point and the corresponding mileage starting point are first calculated. The elevation Z0 and slope α of the vehicle position point on the DEM ground line are calculated by the difference between the front and rear points. When placing the vehicle model, the center point of its rear wheel is translated to coordinates (0, Z0), and then rotated around this point by an angle α to make the vehicle body parallel to the slope. An 8-meter safety threshold is set. Hazardous sections are identified on the eight swept lines. A polygon of the hazardous area is generated by convex hull calculation, and spatial analysis determines that the area involves 67 trees.

[0033] Step S150: Perform 3D visualization verification and display to locate air obstacles in the wind turbine blade transport process; 3D visualization verification and display refers to: superimposing the boundary line of the hidden danger area, the vector data of the building outline, the point cloud data of a single tree, and the target object onto the high-precision real-scene 3D model, placing the 3D model of the vehicle on the terrain 3D model, adjusting the 2D side view outline model of the vehicle, and realizing the positioning of the wind turbine blade transport obstacle.

[0034] In this step, the vector data of the boundary line of the potential hazard area and the location information of the target object are imported into a 3D visualization engine or platform. Precise spatial registration and overlay are then performed based on a high-precision real-scene 3D model. The overlaid potential hazard area is then semi-transparently colored (e.g., filled with semi-transparent red), and the target object is highlighted (e.g., by changing the model color to a warning color, adding a glowing outline, or a prominent icon). Simultaneously, a representative sweep line (e.g., the sweep line with the largest potential hazard mileage range) is selected, and the corresponding detailed 3D model of the vehicle is accurately placed at its starting point on the plane. The blade lifting angle is adjusted according to the operating conditions during the analysis. Figure 7 As shown, through the 3D scene roaming, zooming, and multi-angle observation functions, users can intuitively view the posture of the vehicle and blades in the real geographical environment, as well as the spatial distance and orientation relationship between them and the highlighted obstacles, thus realizing the positioning of air obstacles during the transportation of wind turbine blades.

[0035] Based on this, orthophoto maps and two-dimensional obstacle analysis maps can be output, which overlay the transport path, blade sweep envelope, boundary of the hazard area and specific affected ground objects, to show the obstacles and their locations at different lifting angles.

[0036] In the case provided by this invention, the polygon of the potential hazard area and the marked features are imported into ContextCaptureViewer, and the potential hazard area is rendered as a semi-transparent red. A typical sweep line of risk is selected, and a 3D vehicle model is precisely placed at the starting position in the actual 3D scene. The blades are raised by 30°, allowing users to observe from any angle. The spatial relationship between the blade tip and the highlighted tree canopy in front is clear and intuitive, and the blade sweeps across the trees, consistent with the 2D analysis results. Based on this, a 2D obstacle analysis map, a list of affected features containing detailed information on 67 trees, and a 3D scene package are generated to help staff quickly understand the location of the risk, efficiently locate obstacles during transportation, and confirm the obstacle removal plan.

[0037] This invention, based on centimeter-level precision UAV aerial survey data and analysis models, elevates obstacle assessment from an experience-based level to a quantitative analysis level. Through automatic simulation and analysis, it can quickly process transport routes tens of kilometers long, covering areas inaccessible by human labor, greatly improving survey efficiency. In terms of accuracy, it provides precise positioning and clear target identification, directly pinpointing specific buildings or trees, avoiding indiscriminate obstacle removal and maximizing the protection of the ecological environment. Furthermore, this invention allows for pre-simulation of the passability of different routes and lifting schemes before transport, identifying all risk points in advance and providing strong data support for route optimization and transport plan development.

[0038] The above-disclosed embodiments are merely a few specific examples of the present invention. However, the present invention is not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for locating air obstacles during wind turbine blade transport based on aerial survey topographic data, characterized in that, Includes the following steps: Construct 3D terrain and vehicle models; Building outline vector data and single-tree point cloud data are obtained from the terrain 3D model; wherein, the terrain 3D model includes: digital elevation model, digital surface model, 3D laser point cloud, digital orthophoto, and high-precision real-scene 3D model; the vehicle 3D model includes: vehicle 2D top view outline model, vehicle 2D side view outline model, and vehicle 3D detailed model; wherein, the vehicle 2D side view outline model also includes a series of discrete blade lift angle state models; Simulate the transportation process and generate the sweep envelope range; Within the sweep envelope, a sweep line dataset covering the entire sweep range is generated; Perform longitudinal section collision analysis; the longitudinal section collision analysis is used to determine the mileage intervals on the sweep line data where there is a collision risk or hidden danger, and generate the boundary line of the hidden danger area based on the mileage interval; Spatial overlay analysis is performed on the boundary line of the hazard area, building outline vector data, and single tree point cloud data to determine the target objects for obstacle removal; the target objects include all buildings and multiple trees that are wholly or partially located in the hazard area. Three-dimensional visualization verification and display were conducted to locate air obstacles in the transport of wind turbine blades.

2. The method for locating air obstacles during wind turbine blade transportation based on aerial survey topographic results according to claim 1, characterized in that, The generated sweep envelope range includes: loading the digital orthophoto as a base map, placing the two-dimensional top-view profile model of the vehicle at the planned starting point, and simulating the straight-line and turning motion of the vehicle along the road centerline or a predetermined trajectory through interactive operation; During the interactive operation, the motion trajectory points of the wheels, body, blades and feature contour points during the vehicle's movement and rotation are recorded. The motion trajectory points are connected to form the dynamic sweep envelope range line of the blades on the entire transportation path, thus constituting the sweep envelope range.

3. The method of claim 1, wherein the method is implemented on a mobile device. The method for generating the sweep line dataset includes: With the center of the blade flange as the origin, a series of rays are emitted along the trajectory line, starting from the sweep start edge and rotating towards the sweep end edge at fixed angular intervals or adaptive angular steps, forming a set of sweep line datasets covering the entire sweep range.

4. The method of claim 1, wherein the method is used for a fan blade transport obstacle location based on aerial survey data. The longitudinal section collision analysis includes the following steps: For each sweep line in the sweep line dataset, longitudinal section sections are cut to generate ground longitudinal section lines and top longitudinal section lines of ground features; Perform dynamic correction of vehicle attitude; By comparing the vertical distance relationship between each point on the blade outline of the vehicle's two-dimensional side profile model after dynamic correction of vehicle posture and the longitudinal profile line of the ground and the top longitudinal profile line of ground features, collision risk is assessed and the continuous mileage range of potential collision hazards is determined. Map all the hazard mileage intervals recorded on the sweep line back to the plan view, representing them as multiple short lines along the sweep line direction; these short lines represent the hazard status as hazard segments; and use a geometric algorithm to sequentially connect the inner and outer endpoints of hazard segments on adjacent sweep lines to form the boundary line of the hazard area.

5. The method for locating air obstacles during wind turbine blade transportation based on aerial survey topographic results according to claim 4, characterized in that, The vehicle attitude dynamic correction includes the following steps: The start point of the mileage is determined according to the position of the two-dimensional side view profile model of the vehicle on the plan view, and the current position of the vehicle is defined as the start point of the sweep line ; Adjust the target position of the vehicle's two-dimensional side profile model so that the tire contact point and the ground elevation section line precisely coincide at the target position; According to the slope value of the start point of the sweep line The vehicle two-dimensional side view profile model is rotated as a whole to keep the inclination angle of the vehicle body chassis consistent with the slope.

6. The method of claim 5, wherein the method is implemented on a mobile device. The target location is determined by the elevation value and slope of the corresponding ground longitudinal profile line, including: acquiring a ground elevation at which a center of a rear wheel of the vehicle is located at is expressed as: ; Computing the terrain slope angle at the location α the terrain slope angle α is obtained by central difference method and is expressed as where ΔS is a small distance increment.

7. The method for locating air obstacles during wind turbine blade transport based on aerial survey topographic results according to claim 5, characterized in that, When rotating the entire two-dimensional side view contour model of the vehicle, the two-dimensional side view contour model of the vehicle is first defined as follows: The corresponding blade lift angle is β, and a geometric transformation is performed with the grounding reference point as the reference. The geometric transformation process includes translation and rotation. During translation, the grounding point is aligned with the coordinate system. Coincident; during rotation, the tilt angle of the model's chassis axis is equal to α The geometrically transformed 2D side profile model of the vehicle is represented as: .

8. The method for locating air obstacles during wind turbine blade transport based on aerial survey topographic results according to claim 4, characterized in that, The collision risk assessment includes: Set a safe distance threshold If the outline of the blade is lower than the top longitudinal profile line of the ground feature or the ground longitudinal profile line, then "intersection" occurs and is judged as "collision". If the blade outline is higher than the top longitudinal profile line of the ground feature, but the clearance is less than the safety threshold, it is judged as "having a hidden danger".

9. The method for locating air obstacles during wind turbine blade transportation based on aerial survey topographic results according to claim 1, characterized in that, The 3D visualization verification and display includes: The vector data and target object location information within the boundary of the potential hazard area are imported into a 3D visualization engine or platform, and spatial registration and overlay are performed based on the high-precision real-scene 3D model. The superimposed hazard area is colored semi-transparently, and the target object is highlighted. Select a representative sweep line, place the corresponding detailed 3D model of the vehicle at the starting position on the plane, and adjust the lifting angle of the blades according to the working conditions.

Images