Method and system for rapidly detecting geological settlement dangerous case of unmanned aerial vehicle photovoltaic station

Abstract

The invention relates to the technical field of engineering detection, in particular to an unmanned aerial vehicle photovoltaic station geological settlement dangerous case rapid detection method and system. An unmanned aerial vehicle is used for periodically collecting and processing image data of a photovoltaic station area along an air route to obtain digital elevation models at different times; and comparing the digital elevation models at different times to analyze the settlement condition of the photovoltaic station area. By adopting the unmanned aerial vehicle photography technology, the whole-area monitoring of the photovoltaic station can be realized, the real situation of the ground feature and landform of the subsidence area can be clearly and intuitively reflected, the comprehensiveness of ground subsidence monitoring can be ensured, and the aerial photography efficiency, the modeling speed and the modeling precision are greatly improved; the problems that a traditional subsidence monitoring method is high in cost, low in efficiency and low in precision are solved.

Description

Technical Field

[0001] The present invention relates to the field of engineering detection technology, and in particular to a method and system for quickly detecting geological subsidence hazards in a photovoltaic station using an unmanned aerial vehicle. Background Art

[0002] If a photovoltaic station is built in an area of ​​underground mining or excessive groundwater consumption, geological subsidence is likely to occur. Therefore, it is necessary to strengthen the detection of geological subsidence in the photovoltaic station and discover geological subsidence problems in a timely manner.

[0003] Traditional methods of subsidence monitoring include leveling and GPS technology. Leveling has always been considered the most accurate method of subsidence monitoring. Leveling points are laid out in the monitoring area. Through the adjustment of leveling monitoring data, the settlement information of each point in the area can be obtained. Ground subsidence needs to be monitored periodically. According to the height adjustment value obtained from each period of monitoring, the settlement amount, cumulative settlement amount, settlement rate and other data of this period are calculated to grasp the ground subsidence in the monitoring area. Due to the large monitoring range, this method not only has a large workload and takes a long time, but also requires a lot of manpower costs. GPS technology measures the three-dimensional coordinates of the monitoring points. Through the height difference of the monitoring points at different time periods, it can be concluded whether the ground in the monitoring area has subsided. The GPS monitoring network shape has a great influence on the accuracy of GPS height measurement. The distance between the monitoring point and the subsidence area affects the structure and accuracy of the monitoring network. However, the specific scope of the subsidence area is difficult to grasp. If the monitoring point is selected near the monitoring area, the stability and reliability of the monitoring point will be greatly reduced, and the monitoring results will not reflect the trend of ground subsidence changes in the region.

[0004] It can be seen that the traditional subsidence monitoring method has problems such as high cost, low efficiency and low accuracy. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for rapid detection of geological subsidence hazards in photovoltaic stations using drones, aiming to solve the problems of high cost, low efficiency and low precision of traditional subsidence monitoring methods.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a method for quickly detecting geological subsidence dangers in a photovoltaic station using an unmanned aerial vehicle, comprising:

[0007] S1 Survey the photovoltaic station area, select suitable locations for setting up image control points, set up image control point markers, and measure and record the precise coordinates of each image control point;

[0008] S2 plans and designs the flight path based on the layout of the photovoltaic station area and the location of the image control points;

[0009] The S3 drone flies along the route to take aerial photos, obtain image data of the photovoltaic station area, and obtain the drone's POS data;

[0010] S4 processes the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the surface;

[0011] S5 processes the three-dimensional point cloud data of the surface to generate a digital elevation model;

[0012] S6, repeating steps S3 to S5 at every preset time period to obtain digital elevation models at different times;

[0013] S7 compares the digital elevation models at different times and analyzes the settlement of the photovoltaic station area.

[0014] Among them, survey the photovoltaic station area, select suitable locations for setting up image control points, set up image control point markers, and measure and record the precise coordinates of each image control point, specifically including:

[0015] Combined with high-resolution image data, the photovoltaic station area is surveyed;

[0016] Select easily identifiable locations in the photovoltaic station area to lay out image control point markers;

[0017] RTK technology is used to measure and record the precise coordinates of each image control point.

[0018] Among them, survey the photovoltaic station area, select suitable locations for setting up image control points, set up image control point markers, and measure and record the precise coordinates of each image control point, which also includes:

[0019] Image control point markers use highly recognizable markers that contrast clearly with the surrounding environment.

[0020] The route is planned and designed according to the layout of the photovoltaic station area and the location of the image control points, specifically including:

[0021] The route design adopts the layout of 5n conventional routes and 2n framework routes perpendicular to the conventional routes.

[0022] The drone takes aerial photos along the route to obtain image data of the photovoltaic station area and the POS data of the drone, including:

[0023] The drone carrying the equipment is lifted to a preset altitude and flies along the route;

[0024] During the flight, the drone photographs the photovoltaic station, collects high-resolution image data, and records the drone's POS data in real time when each photo is taken.

[0025] The processing of the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the ground surface specifically includes:

[0026] Use the image data and POS data obtained by the drone to perform aerial triangulation and generate initial point cloud data;

[0027] The initial point cloud data is densely matched to generate high-precision three-dimensional surface point cloud data.

[0028] The processing of the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the ground surface also includes:

[0029] Perform accuracy check on the generated high-precision surface three-dimensional point cloud data.

[0030] The processing of the three-dimensional point cloud data of the ground surface to generate a digital elevation model specifically includes:

[0031] Perform point cloud filtering on the three-dimensional point cloud data of the surface to remove noise and outliers;

[0032] Construct point cloud mesh for the three-dimensional point cloud data of the ground surface, and connect the hashed point cloud data into a triangular irregular grid;

[0033] Interpolation calculation is performed on the triangular irregular grid and converted into a digital elevation model in the form of a regular grid.

[0034] The comparison of digital elevation models at different times and analysis of the settlement of the photovoltaic station area specifically include:

[0035] Compare digital elevation models at different times, calculate the elevation differences between different times, and generate differential data;

[0036] Based on the differential data, calculate the settlement of each point in the photovoltaic station area;

[0037] Based on the settlement amount at each point in the PV station area, the settlement center and the area with large settlement rate are identified, and the spatial distribution of settlement is analyzed.

[0038] In a second aspect, the present invention further provides a UAV photovoltaic station geological subsidence danger rapid detection system, comprising a deployment module, a route planning module, a UAV acquisition module, a first processing module, a second processing module and an analysis module; the deployment module, the route planning module, the UAV acquisition module, the first processing module, the second processing module and the analysis module are connected in sequence;

[0039] The layout module is used to survey the photovoltaic station area, select locations suitable for arranging image control points, arrange image control point markers, and measure and record the precise coordinates of each image control point;

[0040] The route planning module is used to plan and design the route according to the layout of the photovoltaic station area and the layout positions of the image control points;

[0041] The drone acquisition module is used to configure the drone to fly along the route to take aerial photos, obtain image data of the photovoltaic station area, and obtain POS data of the drone;

[0042] The first processing module is used to process the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the surface;

[0043] The second processing module is used to process the three-dimensional point cloud data of the surface to generate a digital elevation model;

[0044] The analysis module is used to compare the digital elevation models at different times and analyze the settlement of the photovoltaic station area.

[0045] The invention discloses a method and system for quickly detecting geological subsidence risks in photovoltaic stations using drones. The method adopts drone photography technology to realize full-area monitoring of photovoltaic stations, and can clearly and intuitively reflect the real situation of landforms in the subsidence area, thereby ensuring the comprehensiveness of ground subsidence monitoring. The aerial photography efficiency, modeling speed and modeling accuracy are greatly improved, thus solving the problems of high cost, low efficiency and low accuracy of traditional subsidence monitoring methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0047] Figure 1 The present invention is a flowchart of a method for quickly detecting geological subsidence hazards in a photovoltaic station using an unmanned aerial vehicle.

[0048] Figure 2 It is a specific flow chart of step S1 of the present invention.

[0049] Figure 3 It is a specific flow chart of step S3 of the present invention.

[0050] Figure 4 It is a specific flow chart of step S4 of the present invention.

[0051] Figure 5 It is a specific flow chart of step S5 of the present invention.

[0052] Figure 6 It is a specific flow chart of step S7 of the present invention.

[0053] Figure 7 It is the UAV route planning diagram of the present invention.

[0054] Figure 8 It is a structural schematic diagram of a rapid detection system for geological subsidence hazards in a UAV photovoltaic station of the present invention.

[0055] 1-deployment module, 2-route planning module, 3-UAV collection module, 4-first processing module, 5-second processing module, 6-analysis module. DETAILED DESCRIPTION

[0056] The first embodiment of the present application is:

[0057] See also Figure 1-Figure 7 ,in, Figure 1 The present invention is a flowchart of a method for quickly detecting geological subsidence hazards in a photovoltaic station using an unmanned aerial vehicle. Figure 2 It is a specific flow chart of step S1 of the present invention. Figure 3 It is a specific flow chart of step S3 of the present invention. Figure 4 It is a specific flow chart of step S4 of the present invention. Figure 5 It is a specific flow chart of step S5 of the present invention. Figure 6 It is a specific flow chart of step S7 of the present invention. Figure 7 It is the UAV route planning diagram of the present invention.

[0058] The present invention provides a method for quickly detecting geological subsidence danger of a UAV photovoltaic station, comprising:

[0059] S1 Survey the photovoltaic station area, select suitable locations for setting up image control points, set up image control point markers, and measure and record the precise coordinates of each image control point;

[0060] S11 combines high-resolution image data to survey the photovoltaic station area;

[0061] S12 Select easily identifiable locations in the photovoltaic station area to arrange image control point markers;

[0062] Image control point markers should be highly recognizable and have a clear contrast with the surrounding environment;

[0063] S13 uses RTK technology to measure and record the precise coordinates of each image control point.

[0064] In this embodiment, the photovoltaic station area can be manually surveyed on the spot, or high-resolution image data can be used for survey to fully understand the topography and other conditions of the photovoltaic station area, and then select suitable locations for the deployment of image control points. These locations should be easy to identify, not easily damaged, and provide sufficient coverage; image control point markers should be highly recognizable markers, such as metal plates, reflectors, etc. After the image control points are deployed, RTK technology is used to measure and record the precise coordinates of each image control point. The measurement period is a fast static measurement, generally 15s or 30s, and the coordinate accuracy is 2cm in the plane and 5cm in elevation, which meets the 1:1000 mapping accuracy requirement; the deployment of image control points is to provide accurate geographic location information for later calibration and control of geometric deformation of the image, to ensure the accuracy of the final generated digital elevation model.

[0065] S2 plans and designs the flight path based on the layout of the photovoltaic station area and the location of the image control points;

[0066] The route design adopts the layout of 5n conventional routes and 2n framework routes perpendicular to the conventional routes. The following effects can be achieved

[0067] Stereoscopic observation: The vertically intersecting route layout helps to build a nearly three-dimensional observation system. By acquiring data in different directions, we can have a more comprehensive understanding of the changes in geological subsidence in three-dimensional space. For example, if we find that a certain area may have signs of subsidence during conventional route detection, combined with the vertical data obtained by the framework route, we can more accurately determine key information such as the depth, range and shape of the subsidence.

[0068] Adapt to complex geology and photovoltaic site layout; Response to complex geological conditions: The geological conditions of photovoltaic sites may be complex and diverse, such as slopes, gullies, or areas with different soil types. The layout of 5n conventional routes and 2n framework routes can better adapt to such complex terrain. For slope areas, vertically intersecting routes can capture changes in the slope surface from different angles; for irregular terrain such as gullies, this route layout can ensure that drones can go deep into these special areas for inspection, avoiding missing settlement risks caused by special geological structures.

[0069] Photovoltaic station facilities: There are a large number of photovoltaic arrays, inverters and other equipment in photovoltaic stations. The layout of these equipment will also affect the detection of geological subsidence. Crossing routes can flexibly bypass these equipment while ensuring effective detection of the surrounding areas of the equipment. For example, in the narrow passages between photovoltaic arrays and around the foundations of equipment, vertical crossing routes can more accurately detect whether geological subsidence occurs in these key locations, thereby promptly discovering potential risks that may threaten the safe operation of equipment.

[0070] In this embodiment, the current layout methods include two methods: one is the uniform layout of conventional routes one by one; the other is to reduce the layout of control points by planning a construction route perpendicular to the conventional route. In this case, it is only necessary to arrange control points at the four corners of the mission area. In fact, through the design of the construction route, the overlapping dimension of the image can be effectively increased, and the stability of the regional network can be further improved. Therefore, this paper adopts the construction route planning and design scheme to obtain the aerial survey data of the mission area. In the process of establishing the error equation, it is necessary to comprehensively consider the conventional route and the construction route.

[0071] Taking the elevation direction as an example, let the two adjacent conventional routes be i and i+1, and the additional framework route be j, then the error equation can be expressed as:

[0072] Original error equation

[0073] It is known that the original error equation is based on the elevation direction. Let two adjacent conventional routes be i and i+1, and an additional framework route be j. The error equation is expressed as:

[0074]

[0075] where v i 、v i+1 、v j are the error terms caused by the deviation between the corresponding route observation value and the average elevation value, ZT i , ZT i+1 , ZT j are the elevation observations obtained from routes i, i+1, and j, respectively, and q i ,q i+1 ,q j is the corresponding weight.

[0076] Constructing a new error equation

[0077] (1) Calculate the elevation difference and error term of adjacent conventional routes

[0078] First, calculate the elevation difference between two adjacent conventional routes i and i+1:

[0079] ΔZT i,i+1 =ZT i+1 -ZT i

[0080] The corresponding error term difference (assuming the difference after weight adjustment) is:

[0081]

[0082] Let δZΔzT i,i+1 For zT i,i+1According to the error propagation law and other related knowledge, we can get the correction number of ΔzT i,i+1 The new error equation is:

[0083]

[0084] in is the theoretical average value of the elevation difference between adjacent conventional routes (similar to the average elevation value assumed above), qΔzT i,i+1 is ΔzT i,i+1 The corresponding weight is related to q i ,q i+1 There is a certain relationship, which can be determined based on the specific measurement adjustment theory and actual conditions.

[0085] (2) Calculate the elevation difference and error term between the framework route and the adjacent conventional route

[0086] Calculate the elevation difference between the frame route and the adjacent conventional route:

[0087] ΔZT j,i =ZT j -ZT i

[0088] The corresponding error term difference is:

[0089]

[0090] Let δZΔzT j,i is ΔzT j,i The correction number of ΔzT is obtained based on the error propagation law and other related knowledge. j,i The new error equation is:

[0091]

[0092] in It is the average value of the elevation difference between the theoretical framework route and the adjacent conventional route.

[0093] qΔZT j,i is ΔZT j,i The corresponding weight.

[0094] It can be seen that the layout of the route structure will match more image connection points. More error equations can be established based on these feature points, which can achieve the following effects:

[0095] Increased number: For each tie point, an error equation can be established based on factors such as its coordinate difference and elevation difference in different images to measure the measurement error. More tie points means more geometric constraints, so more equations can be established to describe the relationship between images.

[0096] Improved association: These error equations can more accurately describe the spatial relationship between images. By using these error equations through methods such as least squares adjustment, the relative position and attitude of the images can be adjusted to improve the association between the images.

[0097] Enhanced network stability: Due to the existence of the framework route, the image connection points and the error equation work together to build a more complex and stable geometric network. It is like building a denser frame structure in a three-dimensional space, with each image being the node of the frame, and the connection points and error equations being the ropes and constraints connecting the nodes. This stable network can better resist the uncertainty caused by measurement errors, terrain undulations and other factors.

[0098] Improve the anti-interference ability of the grid: When facing complex field environments and changeable measurement conditions, such as the impact of wind on the flight attitude of the drone and image deformation caused by terrain undulations, a stable grid can effectively reduce the interference of these factors on the overall measurement results. It can correct the relative position deviation between images in a timely manner according to the adjustment mechanism of the error equation, maintain the relative stability of the entire grid, and thus ensure measurement accuracy and data quality.

[0099] The S3 drone flies along the route to take aerial photos, obtain image data of the photovoltaic station area, and obtain the drone's POS data;

[0100] The S31 drone carrying the equipment is lifted to a preset altitude and flies along the route;

[0101] During the flight of S32, the drone photographs the photovoltaic station, collects high-resolution image data, and records the drone's POS data in real time when each photo is taken.

[0102] The drone automatically flies along the predetermined route and captures high-resolution image data of the photovoltaic station. During the flight, the GPS or RTK equipment on the drone will record the precise position and attitude information of each photo taken in real time, namely POS data. The image data and POS data obtained by the drone are quality checked. According to the requirements, the route benchmark design resolution is 3cm, the relative line height is 314m, the heading overlap is 80%, and the lateral overlap is 65%. If the overlap is not enough and a gap occurs, a supplementary flight is required.

[0103] S4 processes the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the surface;

[0104] S41 uses the image data and POS data acquired by the drone to perform aerial triangulation and generate initial point cloud data;

[0105] S42 performs dense matching on the initial point cloud data to generate high-precision three-dimensional point cloud data of the surface;

[0106] S43 performs accuracy detection on the generated high-precision surface three-dimensional point cloud data.

[0107] In this embodiment, the image data and POS data acquired by the drone are subjected to the steps of data preprocessing, feature point matching, control point puncture and bundle adjustment to complete the aerial triangulation; data preprocessing is to perform radiation correction, geometric correction, de-distortion and other preprocessing operations on the acquired image data to ensure that the data quality is suitable for subsequent processing; feature point matching is to use feature detection algorithms (such as SIFT, SURF, ORB, etc.) to identify and match the same-name points between multi-view images. This step is the basis for establishing a three-dimensional model, and the matching accuracy directly affects the quality of subsequent results; control point puncture is to mark the control points with known ground coordinates on the image, and these points are used to correct The geometric error of the image can be corrected and the measurement accuracy can be improved. The bundle adjustment adopts the bundle adjustment technology and optimizes the coordinates of the exterior orientation elements and encrypted points of all images through the principle of least squares method to obtain consistency and high precision of the entire area. The POS data is used to assist the orientation of the image, reduce the amount of matching calculations, and improve the processing speed. The POS data is combined with the image matching points to generate the three-dimensional coordinates of the ground points to form point cloud data. The initial point cloud data generated by aerial triangulation is densely matched to generate high-precision surface three-dimensional point cloud data. Finally, the high-precision surface three-dimensional point cloud data is precision tested to ensure the accuracy and reliability of the data.

[0108] S5 processes the three-dimensional point cloud data of the surface to generate a digital elevation model;

[0109] S51 performs point cloud filtering on the three-dimensional point cloud data of the surface to remove noise and outliers;

[0110] S52 constructs a point cloud network for the three-dimensional point cloud data of the ground surface, and connects the hashed point cloud data into a triangular irregular grid;

[0111] S53 performs interpolation calculation on the triangular irregular grid and converts it into a digital elevation model in the form of a regular grid.

[0112] In this embodiment, the main purpose of point cloud filtering is to remove noise and outliers, and to downsample the three-dimensional point cloud data of the surface to reduce the amount of data and improve processing efficiency; point cloud networking, also known as triangulation, is the process of connecting hashed point cloud data into a triangular irregular grid. The commonly used algorithm is Delaunay triangulation. Finally, the triangular irregular grid is interpolated and converted into a digital elevation model in the form of a regular grid.

[0113] S6, repeating steps S3 to S5 at every preset time period to obtain digital elevation models at different times;

[0114] A fixed time period is preset, and steps S3 to S5 are repeated periodically to obtain digital elevation models at different times.

[0115] S7 compares the digital elevation models at different times and analyzes the settlement of the photovoltaic station area.

[0116] S71 compares the digital elevation models at different times, calculates the elevation differences between different times, and generates differential data;

[0117] S72 calculates the settlement amount of each point in the photovoltaic station area based on the differential data;

[0118] S73 identifies the settlement center and the area with a large settlement rate based on the settlement amount of each point in the photovoltaic station area, and analyzes the spatial distribution of the settlement.

[0119] In this embodiment, professional analysis tools are used to compare digital elevation models at different times, calculate the elevation differences between different times, generate differential data, and generate differential DEM, which can reveal the surface settlement. At the same time, based on the differential data, the settlement amount of each point in the photovoltaic station area is calculated. Finally, the spatial distribution characteristics of the settlement are analyzed to obtain the settlement situation of the photovoltaic station area.

[0120] The method for rapid detection of geological subsidence risks in photovoltaic stations using drones described in this embodiment adopts drone photography technology to achieve full-area monitoring of photovoltaic stations, can clearly and intuitively reflect the actual situation of the landforms in the subsidence area, can ensure the comprehensiveness of ground subsidence monitoring, and greatly improves the aerial photography efficiency, modeling speed and modeling accuracy, thus solving the problems of high cost, low efficiency and low accuracy of traditional subsidence monitoring methods.

[0121] The second embodiment of the present application is:

[0122] See also Figure 8 ,in, Figure 8 It is a structural schematic diagram of a rapid detection system for geological subsidence hazards in a UAV photovoltaic station of the present invention.

[0123] The invention provides a UAV photovoltaic station geological subsidence danger rapid detection system, comprising a deployment module 1, a route planning module 2, a UAV acquisition module 3, a first processing module 4, a second processing module 5 and an analysis module 6; the deployment module 1, the route planning module 2, the UAV acquisition module 3, the first processing module 4, the second processing module 5 and the analysis module 6 are connected in sequence;

[0124] The layout module 1 is used to survey the photovoltaic station area, select locations suitable for arranging image control points, arrange image control point markers, and measure and record the precise coordinates of each image control point;

[0125] The route planning module 2 is used to plan and design the route according to the layout of the photovoltaic station area and the layout positions of the image control points;

[0126] The drone acquisition module 3 is used to configure the drone to fly along the route to take aerial photos, obtain image data of the photovoltaic station area, and obtain POS data of the drone;

[0127] The first processing module 4 is used to process the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the surface;

[0128] The second processing module 5 is used to process the three-dimensional point cloud data of the surface to generate a digital elevation model;

[0129] The analysis module 6 is used to compare the digital elevation models at different times and analyze the settlement of the photovoltaic station area.

[0130] What is disclosed above is only one or more preferred embodiments of the present application, and cannot be used to limit the scope of rights of the present application. Ordinary technicians in this field can understand that all or part of the processes of implementing the above embodiments and equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A method for rapid detection of geological subsidence hazards in a photovoltaic station using an unmanned aerial vehicle, characterized in that: include: S1 Survey the photovoltaic station area, select suitable locations for setting up image control points, set up image control point markers, and measure and record the precise coordinates of each image control point; S2 plans and designs the flight path based on the layout of the photovoltaic station area and the location of the image control points; The S3 drone flies along the route to take aerial photos, obtain image data of the photovoltaic station area, and obtain the drone's POS data; S4 processes the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the surface; S5 processes the three-dimensional point cloud data of the surface to generate a digital elevation model; S6, repeating steps S3 to S5 at every preset time period to obtain digital elevation models at different times; S7 compares the digital elevation models at different times and analyzes the settlement of the photovoltaic station area.

2. A method for rapid detection of geological subsidence danger in a UAV photovoltaic station as claimed in claim 1, characterized in that: Survey the photovoltaic station area, select suitable locations for setting up image control points, set up image control point markers, and measure and record the precise coordinates of each image control point, including: Combined with high-resolution image data, the photovoltaic station area is surveyed; Select easily identifiable locations in the photovoltaic station area to lay out image control point markers; RTK technology is used to measure and record the precise coordinates of each image control point.

3. A method for rapid detection of geological subsidence danger of a UAV photovoltaic station as claimed in claim 2, characterized in that: Survey the photovoltaic station area, select suitable locations for setting up image control points, set up image control point markers, and measure and record the precise coordinates of each image control point, including: Image control point markers use highly recognizable markers that contrast clearly with the surrounding environment.

4. A method for rapid detection of geological subsidence danger in a UAV photovoltaic station as claimed in claim 3, characterized in that: The route is planned and designed according to the layout of the photovoltaic station area and the location of the image control points, specifically including: The route design adopts the layout of 5n conventional routes and 2n framework routes perpendicular to the conventional routes.

5. A method for rapid detection of geological subsidence danger in a UAV photovoltaic station as claimed in claim 4, characterized in that: The drone flies along the route to take aerial photos, obtain image data of the photovoltaic station area, and obtain the drone's POS data, including: The drone carrying the equipment is lifted to a preset altitude and flies along the route; During the flight, the drone photographs the photovoltaic station, collects high-resolution image data, and records the drone's POS data in real time when each photo is taken.

6. A method for rapid detection of geological subsidence danger in a UAV photovoltaic station as claimed in claim 5, characterized in that: The processing of the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the surface specifically includes: Use the image data and POS data obtained by the drone to perform aerial triangulation and generate initial point cloud data; The initial point cloud data is densely matched to generate high-precision three-dimensional surface point cloud data.

7. A method for rapid detection of geological subsidence danger in a UAV photovoltaic station as claimed in claim 6, characterized in that: The processing of the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the ground surface also includes: Perform accuracy check on the generated high-precision surface three-dimensional point cloud data.

8. A method for rapid detection of geological subsidence danger in a UAV photovoltaic station as claimed in claim 7, characterized in that: The processing of the three-dimensional point cloud data of the ground surface to generate a digital elevation model specifically includes: Perform point cloud filtering on the three-dimensional point cloud data of the surface to remove noise and outliers; Construct point cloud mesh for the three-dimensional point cloud data of the ground surface, and connect the hashed point cloud data into a triangular irregular grid; Interpolation calculation is performed on the triangular irregular grid and converted into a digital elevation model in the form of a regular grid.

9. A method for rapid detection of geological subsidence danger in a UAV photovoltaic station as claimed in claim 8, characterized in that: The digital elevation models at different times are compared to analyze the settlement of the photovoltaic station area, specifically including: Compare digital elevation models at different times, calculate the elevation differences between different times, and generate differential data; Based on the differential data, calculate the settlement of each point in the photovoltaic station area; Based on the settlement amount at each point in the PV station area, the settlement center and the area with large settlement rate are identified, and the spatial distribution of settlement is analyzed.

10. A UAV photovoltaic station geological subsidence danger rapid detection system, applied to a UAV photovoltaic station geological subsidence danger rapid detection method as claimed in any one of claims 1 to 9, characterized in that: It includes a deployment module, a route planning module, a drone collection module, a first processing module, a second processing module and an analysis module; the deployment module, the route planning module, the drone collection module, the first processing module, the second processing module and the analysis module are connected in sequence; The layout module is used to survey the photovoltaic station area, select locations suitable for arranging image control points, arrange image control point markers, and measure and record the precise coordinates of each image control point; The route planning module is used to plan and design the route according to the layout of the photovoltaic station area and the layout positions of the image control points; The drone acquisition module is used to configure the drone to fly along the route to take aerial photos, obtain image data of the photovoltaic station area, and obtain POS data of the drone; The first processing module is used to process the image data and POS data acquired by the drone to generate three-dimensional point cloud data of the surface; The second processing module is used to process the three-dimensional point cloud data of the surface to generate a digital elevation model; The analysis module is used to compare the digital elevation models at different times and analyze the settlement of the photovoltaic station area.

Images