A method and system for road planning in wind and solar power plant areas based on terrain analysis

Abstract

This application provides a method and system for road planning in wind and solar power plant areas based on terrain analysis. The method processes digital elevation model data and vehicle turning radius parameters of the wind and solar power plant area, performs hierarchical reconstruction and optimization, generates optimized sampling data, and constructs a road network using this data. Based on the road network data, further path optimization is performed to obtain a path candidate set. Finally, through electric field layout and equipment adaptation algorithms, construction adaptability processing, and safety smoothing correction methods, final road path data suitable for the wind and solar power plant area is generated. This improves road planning accuracy, enhances construction adaptability, meets the actual needs under different terrain conditions, avoids the problem of traditional path planning failing to reasonably integrate local environmental characteristics and operational safety requirements, and effectively improves the feasibility of wind and solar power plant road planning.

Description

Technical Field

[0001] This invention relates to the field of energy power plant road design and planning, specifically to a method and system for planning roads in wind and solar power plant areas based on terrain analysis. Background Technology

[0002] As an important form of renewable energy generation, wind and solar power plants are expanding in scale, making road planning within the power plant site a crucial aspect of project implementation. Roads not only need to provide access for equipment transportation but also need to support subsequent construction, equipment maintenance, and daily safe operation. Wind and solar power plant sites typically have complex terrain conditions, such as hills, mountains, and uneven terrain, which places higher demands on road design. Adaptable and precise road planning is essential for ensuring efficient power plant operation and reducing construction and maintenance costs. Therefore, a well-planned road network for wind and solar power plant sites can effectively support equipment transportation and construction, improve the convenience of equipment installation and maintenance, and ensure safety and smooth traffic flow during power plant operation.

[0003] Existing road planning methods for wind and solar power plant sites are mostly based on traditional terrain data analysis and path optimization algorithms, typically employing path planning techniques based on digital elevation models (DEMs). These methods generally achieve some success under relatively simple terrain conditions, but in complex terrain and variable environmental conditions, they fail to fully consider the specific needs of wind and solar power plant sites. For example, existing technologies often neglect the actual needs during construction and maintenance, failing to consider equipment access restrictions such as turning radius, slope, and road width during the planning stage, and failing to effectively integrate local environmental characteristics with the safety requirements of electric field operation. Traditional path optimization methods are often limited to single terrain features or simple constraints, failing to comprehensively consider the actual construction environment and equipment access requirements.

[0004] Therefore, existing path planning schemes fail to meet the complex terrain features, construction requirements, and equipment access and safety needs of wind and solar power plant sites. The accuracy and adaptability of path planning are poor, making it difficult to effectively address the actual needs under different terrain conditions. Furthermore, existing technologies fail to reasonably integrate local environmental characteristics with the safety requirements of electric field operation during path optimization, resulting in paths that not only have adaptability issues during construction but also pose significant challenges in subsequent equipment installation and routine maintenance. Summary of the Invention

[0005] In view of the above-mentioned actual situation, this application proposes a road planning method and system for wind and solar power plant sites based on terrain analysis, in order to solve the problem that existing road planning methods fail to fully consider the complex terrain features, construction and maintenance requirements, and equipment access and safety needs of wind and solar power plant sites, resulting in insufficient accuracy of route planning, poor construction adaptability, difficulty in effectively adapting to the actual needs under different terrain conditions, and failure to reasonably integrate local environmental features and operational safety requirements in the route optimization process.

[0006] A method for road planning in wind and solar power plant areas based on terrain analysis, the method comprising the following steps:

[0007] S1, acquire data to be processed, the data to be processed includes digital elevation model data of the wind and solar power station area, vehicle turning radius parameters and facility distribution data, the digital elevation model data includes elevation data and DEM pixel size, and the facility distribution data is the location and distribution information of all relevant equipment and infrastructure in the power station;

[0008] S2, performs hierarchical reconstruction and optimization on the data to be processed, and generates optimized sampling data;

[0009] S3, construct a road network based on the optimized sampling data to obtain road network data;

[0010] S4. Based on road network data, path selection and constraint adjustment are performed to obtain a path candidate set;

[0011] S5. Based on the path candidate set and the facility distribution data, generate power station road path data suitable for wind and solar power plants.

[0012] Furthermore, step S2 includes the following sub-steps:

[0013] S201, perform multi-scale decomposition on the elevation data based on gradient decomposition, extract terrain features at different scales, and set an adaptive sampling step size according to each terrain feature. Compare the sampling step size with the DEM cell size. When the sampling step size is greater than the cell size, perform sampling operation to generate terrain scale data.

[0014] S202 defines local path-sensitive areas based on terrain-scale data, vehicle turning radius, and slope information, and generates optimized sampling data by optimizing the sampling density of the sensitive areas through weighted convolution.

[0015] Furthermore, step S3 includes the following sub-steps:

[0016] S301, based on optimized sampling data, uses a weighted Voronoi diagram partitioning algorithm to spatially segment the terrain of the field area, obtaining Voronoi units for each terrain;

[0017] S302, combining the gradient information of each Voronoi unit with the terrain resistance coefficient, a multi-level weighted connection method is used to connect adjacent units to generate a preliminary road network candidate map;

[0018] S303, based on the topology of the preliminary road network candidate graph, combines the minimum spanning tree algorithm in graph theory to optimize and generate the road network, thus obtaining the final road network data.

[0019] Furthermore, step S4 includes the following sub-steps:

[0020] S401, based on road network data, uses heuristic search to calculate the shortest path from the starting point to the destination, generating a preliminary set of candidate paths;

[0021] S402, adjust the edge weights based on local environmental features, optimize the initial path candidate set using the edge weight adjustment model, and generate the adjusted path weight matrix;

[0022] S403 optimizes the path based on the path weight matrix and the Lagrange multiplier method, eliminating paths that do not meet the constraints and generating a path candidate set.

[0023] Furthermore, step S5 includes the following sub-steps:

[0024] S501, based on the path candidate set, uses electric field layout and equipment adaptation algorithms to modify the path structure, ensuring that the path avoids key equipment and infrastructure, and optimizes the path layout to generate the modified path structure.

[0025] S502, taking into account the construction and maintenance needs within the electric field, performs construction adaptability processing on the modified path to generate construction adaptability path data;

[0026] S503, based on construction adaptability path data and combined with electric field operation and safety requirements, performs path smoothing and correction processing to generate the final electric field road path data.

[0027] Furthermore, step S303 includes the following sub-steps:

[0028] The minimum spanning tree algorithm is performed through the following steps:

[0029] S3031, Sort all edges in the preliminary road network candidate graph to generate edge weight sequence data;

[0030] S3032, Based on the edge weight sequence data, the Kruskal algorithm or Prim algorithm is used to select the edges one by one, select the edge with the smallest weight, and connect the Voronoi units step by step until the spanning tree contains all units, generating the minimum spanning tree data;

[0031] S3033, based on the minimum spanning tree data, avoid loop generation, ensure that the selected path can connect all units and form a tree structure to generate road network data, the road network data containing the set of connected paths with the smallest weight among all Voronoi units.

[0032] Furthermore, the heuristic search computation in S401 includes Dijkstra's algorithm for shortest path search and A algorithm for heuristic estimation, and the heuristic search computation is performed according to the following steps:

[0033] S4011, Initialize the path cost of the node, set the path cost of the starting point to 0, and the path cost of other nodes to infinity.

[0034] S4012, calculate the total cost of each node to be processed, including the actual cost and heuristic estimate, and select the node with the lowest cost for expansion;

[0035] S4013, update the path cost of adjacent nodes based on edge weights and heuristic estimates, and continue to expand nodes until the endpoint node is expanded or all nodes are processed.

[0036] S4014, generate the shortest path from the starting point to the ending point, and add the path to the preliminary path candidate set.

[0037] Furthermore, the constraints in S403 are minimum road width and maximum slope limits.

[0038] Furthermore, the electric field layout and equipment adaptation algorithm in S501 includes path obstacle avoidance processing and path optimization processing; the construction adaptability processing in S502 includes path width adjustment, slope adjustment, turning radius optimization, and path smoothness improvement processing; the path smoothing and correction processing in S503 includes path curve smoothing, slope equalization, turning radius correction, and safe area avoidance processing.

[0039] Furthermore, this application also discloses a road planning system for wind and solar power plant sites based on terrain analysis, characterized in that the system includes:

[0040] The acquisition unit is used to acquire data to be processed, which includes digital elevation model data of the wind and solar power plant area, vehicle turning radius parameters, and facility distribution data. The digital elevation model data includes elevation data and DEM pixel size, and the facility distribution data is the location and distribution information of all relevant equipment and infrastructure within the power plant.

[0041] The optimized sampling unit is used to perform hierarchical reconstruction and optimization of the data to be processed, generating optimized sampling data.

[0042] A road network construction unit is used to construct a road network based on the optimized sampling data, thereby obtaining road network data;

[0043] The route filtering unit is used to filter routes and adjust constraints based on road network data, thereby obtaining a set of candidate routes;

[0044] The power plant path generation unit is used to generate power plant road path data suitable for wind and solar power plants based on the path candidate set and the facility distribution data.

[0045] The proposed method and system for road planning in wind and solar power plant areas based on terrain analysis achieves multi-scale terrain analysis and optimization, combined with local path-sensitive areas and environmental characteristics, to optimize paths for complex terrain and actual construction needs in wind and solar power plant areas. This ensures that the generated road paths can adapt to changes in different terrains, equipment access requirements, and actual construction and maintenance conditions, ultimately providing a high-precision, feasible, and safe road planning scheme. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the method flow for a wind and solar power plant site road planning method based on terrain analysis proposed in this application;

[0047] Figure 2 This is a flowchart illustrating the process of optimizing sampling data in a wind and solar power plant site road planning method based on terrain analysis proposed in this application.

[0048] Figure 3 A schematic diagram illustrating the process of obtaining road network data in a terrain-analysis-based road planning method for wind and solar power plant sites proposed in this application.

[0049] Figure 4 This is a schematic diagram illustrating the process of generating a path candidate set in a wind and solar power plant site road planning method based on terrain analysis proposed in this application.

[0050] Figure 5 This is a schematic diagram illustrating the process of generating the final electric field road path data in a wind and solar power plant site road planning method based on terrain analysis proposed in this application.

[0051] Figure 6 A schematic diagram of a road planning system for a wind and solar power plant site based on terrain analysis is provided for an embodiment of this application. Detailed Implementation

[0052] The simulation technology route in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] The features and performance of the present invention will be further described in detail below with reference to embodiments. Please refer to the appendix. Figure 1 As shown, a road planning method for wind and solar power plant sites based on terrain analysis is proposed. The method includes the following steps:

[0055] S1, acquire data to be processed, the data to be processed includes digital elevation model data of the wind and solar power station area, vehicle turning radius parameters and facility distribution data, the digital elevation model data includes elevation data and DEM pixel size, and the facility distribution data is the location and distribution information of all relevant equipment and infrastructure in the power station;

[0056] In some implementations, acquiring the data to be processed is the first step in road planning, involving the collection and integration of multiple data sources. Digital elevation model (DEM) data is fundamental for acquiring site topographic information, providing necessary topographic data support for subsequent route planning. DEM data includes ground elevation information and pixel size; the former reflects the ground elevation values ​​at different points within the site, while the latter defines the spatial resolution of the data, i.e., the actual area size of each pixel in reality.

[0057] In this embodiment, the acquired digital elevation model (DEM) data is obtained through remote sensing or laser scanning technologies (such as LiDAR). These technologies can accurately record the topographic relief of the site and generate highly accurate elevation data. Specifically, the elevation data of the DEM represents the height information of different locations on the ground. It not only helps in analyzing the slope and aspect of the terrain but also provides detailed terrain features for path selection in subsequent steps. This elevation data is represented in the form of a grid, with each grid point corresponding to a specific spatial coordinate and altitude, forming an elevation map covering the entire site.

[0058] Furthermore, the cell size in DEM data defines the actual geographical extent represented by each elevation point within the field, measured in meters. For example, a cell size of 1 meter means that each elevation point represents an area of ​​1 square meter. The choice of cell size directly affects the spatial resolution of the data and the accuracy of subsequent calculations. In some implementations, a smaller cell size means obtaining higher-precision terrain information, but it may also lead to an increase in data volume. Therefore, a trade-off needs to be struck based on specific requirements when acquiring the data.

[0059] Vehicle turning radius parameters, as another input data point in road planning, are primarily used to describe the turning radius limitations of roads within wind and solar power plant sites. The turning radius refers to the radius of the trajectory traced by the wheels when a vehicle turns. In some implementations, considering the size and driving requirements of different types of vehicles within the site, the turning radius needs to be set according to the specific vehicle type. A smaller turning radius requires roads with larger curve radii to prevent vehicles from derailing or being unable to pass smoothly during turns. Turning radius data affects the construction of the road network during subsequent route planning, especially when curve design and gradient limitations need to be considered.

[0060] By integrating digital elevation model data and vehicle turning radius parameters, this step provides accurate input information for subsequent data processing and path optimization. The acquisition and processing of this data provides a solid foundation for the implementation of road planning in wind and solar power plant areas, ensuring the scientific validity and feasibility of road planning under complex terrain and various driving conditions. In this embodiment, relying on high-precision remote sensing technology and reasonable vehicle parameter settings, necessary terrain support and driving constraints can be provided for path planning, ensuring the achievement of a reasonable road layout.

[0061] S2, performs hierarchical reconstruction and optimization on the data to be processed, and generates optimized sampling data;

[0062] For details, please refer to the appendix. Figure 2 As shown, this step includes the following sub-steps:

[0063] S201, perform multi-scale decomposition on the elevation data based on gradient decomposition, extract terrain features at different scales, and set an adaptive sampling step size according to each terrain feature. Compare the sampling step size with the DEM cell size. When the sampling step size is greater than the cell size, perform sampling operation to generate terrain scale data.

[0064] In some implementations, gradient decomposition is used to extract terrain features at different scales from digital elevation model (DEM) data. During gradient decomposition, the elevation data is first processed, and by decoupling information at different scales, local and global terrain features can be identified and extracted. Gradient decomposition is based on the rate of change of the terrain surface, also known as the gradient, which spatially reflects the magnitude and direction of elevation changes and is a core indicator in terrain analysis.

[0065] Gradient describes the rate of change of surface relief. In this embodiment, given elevation data... Where x and y represent the horizontal and vertical coordinates in the geographic coordinate system, respectively, and the gradient... Represented as: ,in and These are the partial derivatives of the elevation data in the x and y directions, respectively, reflecting the slope and aspect of the terrain.

[0066] When performing multi-scale decomposition, wavelet transform or scale-space methods are used to decompose the elevation data layer by layer. In some implementations, by performing multi-level scale-space processing on the gradient data, terrain features at different scales can be extracted separately. By processing the terrain gradient at each scale, different levels of terrain information can be obtained, including local terrain undulations and global terrain morphology.

[0067] In this embodiment, different scale factor selections extract terrain features suitable for different analytical needs through gradient decomposition. For example, a smaller scale factor can analyze small-scale local changes in detail, such as excessive slope or obstacle detection; a larger scale factor focuses on the overall terrain trend over a large area, assisting in macroscopic terrain identification during path planning. This multi-scale decomposition allows for adaptive selection of appropriate scales according to different scenario requirements, accurately describing and modeling terrain data. Let be the terrain data at scale s, where s represents the scale factor. By performing scale decomposition on the gradient data, the gradient values ​​at different scales can be used for subsequent path planning and terrain analysis. The formula for decomposing the gradient data is expressed as: ,in It is the wavelet function corresponding to the scaling factor s. This represents the elevation data offset at position t with x as the horizontal coordinate. By convolving the terrain data using this wavelet function, multi-scale terrain data can be obtained.

[0068] Next, an adaptive sampling step size is set based on the terrain features extracted at different scales. The choice of sampling step size directly affects the accuracy of the data and computational efficiency. In some implementations, the sampling step size is adjusted according to the degree of change in terrain features. In areas with significant terrain changes, the sampling step size is relatively small to ensure that subtle terrain changes can be captured; in areas with gentler terrain changes, the sampling step size is larger to reduce data redundancy and computational burden.

[0069] set up Let be the sampling step size at scale s. The choice depends on the changes in terrain gradient. Specifically, when the terrain changes drastically, the gradient... When the gradient is large, a smaller sampling step size should be chosen; conversely, in areas with slow terrain change and smaller gradients, a larger sampling step size is appropriate. The comparison between the sampling step size and the DEM cell size determines the execution conditions of the sampling operation. Let... The size of the DEM pixels, if the sampling step size Larger than the pixel size Then a sampling operation is performed. Specifically, when: At this time, a sampling operation is performed, and the sampled points are added to the generated topographic-scale data. This operation ensures sufficient sampling in areas with large topographic features, while reducing unnecessary computation in flat areas, thereby optimizing the data distribution.

[0070] The generated topographic scale data contains topographic feature information at different scales. This data not only reflects the detailed changes in the terrain but also helps in subsequent steps such as path planning and road network construction. In this embodiment, through this multi-scale decomposition and adaptive sampling process, the utilization efficiency of topographic data can be effectively improved, while ensuring the accuracy and rationality of sampling under different topographic features.

[0071] S202 defines local path-sensitive areas based on terrain-scale data, vehicle turning radius, and slope information, and optimizes the sampling density of the sensitive areas through weighted convolution to generate optimized sampling data;

[0072] In some implementations, this step identifies local areas that significantly impact route planning by integrating topographical data, vehicle turning radii, and slope information; these areas are referred to as route-sensitive areas. Route-sensitive areas primarily refer to regions with significant impact on road planning, such as steep slopes and sharp turns, where terrain is complex or vehicle movement is restricted. The sampling density of these areas needs to be appropriately adjusted based on their impact on route planning to ensure the rationality and feasibility of the route.

[0073] The definition of a local path-sensitive area is based on a comprehensive consideration of terrain information, vehicle turning radius, and slope factors. In this embodiment, it is set as follows: Let be the gradient data at scale s, representing the rate of change of the terrain; let be... This refers to slope information, indicating the magnitude of the ground slope. Calculated using the following formula: ,in and These represent the terrain gradients in the x and y directions, respectively. The steeper the slope, the more drastic the terrain changes, and the higher the requirements for the path when vehicles are driving.

[0074] Furthermore, the vehicle's turning radius *r* plays a decisive role in path planning. In this embodiment, the influence of the turning radius is reflected in the turning restrictions of the terrain area. A smaller turning radius means that the vehicle can turn in a smaller space, requiring a more compact road curve design. Within the site, it is necessary to combine the turning radius information with slope and terrain features to determine which areas are path-sensitive areas. The delineation of path-sensitive areas is achieved through the following model: ,in Indicates at point Path sensitivity at the location, These are weighting coefficients, which control the gradient data respectively. Slope information and turning radius Impact on path sensitivity. In some implementations, weighting coefficients are dynamically adjusted according to specific needs, ensuring higher sampling density in areas with dramatic terrain changes and lower sampling density in relatively flat areas. Areas with higher path sensitivity mean a greater impact on road planning, thus requiring optimization of sampling density to address this.

[0075] Once the path-sensitive region is defined, the process moves to the stage of weighted convolution to optimize the sampling density. Weighted convolution is a spatially weighted averaging technique whose goal is to adjust the density of sampling points based on path sensitivity. Let... For The sampling density at a location, when using a weighted convolution operation, is determined by the following formula: ,in The convolution kernel function controls the influence of the surrounding region on the current sampling point. The path sensitivity value reflects the sensitivity of a given location. Weighted convolution adjusts the sampling density at each location by weighting the sensitivity values ​​of its neighborhood in the spatial domain. This operation optimizes the sampling density in sensitive areas, ensuring more sampling points in highly sensitive regions (such as steep slopes and sharp turns) and fewer sampling points in gentler areas, thus reducing computational cost. In this embodiment, the weighted convolution kernel function... Using a Gaussian function or other suitable space-weighted functions, the specific form is as follows: ,in This is a parameter used to control the width of the convolutional kernel. Smaller... The value indicates a smaller weighted region and a larger weighted region. A value indicates a larger weighted region. In some implementations, the shape and size of the convolution kernel can be adjusted according to terrain features and path planning requirements to achieve optimal sampling results.

[0076] After weighted convolution processing, the resulting sampling density reflects changes in terrain and the actual needs of vehicle travel. The generated optimized sampling data is the final set of sampling points. These sampling points will serve as the basis for subsequent road network construction and route planning, ensuring the accuracy and feasibility of route planning.

[0077] S3, construct a road network based on the optimized sampling data to obtain road network data;

[0078] For details, please refer to the appendix. Figure 3 As shown, this step includes the following sub-steps:

[0079] S301, based on optimized sampling data, uses a weighted Voronoi diagram partitioning algorithm to spatially segment the terrain of the field area, obtaining Voronoi units for each terrain;

[0080] In some implementations, this step utilizes optimized sampled data to spatially segment the terrain of the field area using a weighted Voronoi diagram partitioning algorithm. The weighted Voronoi diagram partitioning algorithm is a mathematical method based on point set partitioning. It generates polygonal regions around a given point set (sampled data points) such that all points within each polygon are closer to the point that generated the polygon than to other generated points. In this embodiment, the optimized sampled data provides a series of sampling points related to terrain features, which serve as seed points in the weighted Voronoi diagram for spatial partitioning to obtain different terrain Voronoi units.

[0081] The construction of the weighted Voronoi diagram not only considers the distance relationships between points but also introduces additional weight information. In this embodiment, the complexity of the terrain is represented by the sensitivity of the sampling points, and the sensitivity value is the optimized sampling density from the previous step. Specifically, at a given sampling point... and their corresponding weights In this case, the spatial segmentation of the weighted Voronoi diagram is calculated according to the following formula: ,in Represents any point in space. These are sampling points. Indicates spatial distance. Indicates sampling point The formula shows that the distance from each spatial point to the sampling point is not only determined by the Euclidean distance, but also affected by the sampling point weight, thus making areas with larger weights (such as areas with large terrain changes or path-sensitive areas) more attractive.

[0082] The partitioning of a weighted Voronoi diagram is achieved through the following steps. First, a set of sampling points is defined. Each sampling point has a weight. Then, according to the principle of weighted distance, the space is divided into several Voronoi units. For any spatial point... The Voronoi element to which it belongs is determined by the following conditions: The sampling points that minimize the weighted distance The corresponding Voronoi element is the element to which the spatial point P belongs.

[0083] In this embodiment, the terrain of the field is divided into multiple Voronoi units based on the distribution and weights of the sampling points using the aforementioned weighted Voronoi graph partitioning algorithm. Each Voronoi unit corresponds to a terrain region, where the weighted distance between each point in the region and the sampling points within that unit is the shortest. In this way, by using the distribution and weight information of the sampling points, the terrain of the field can be divided into multiple regions with certain inherent characteristics, providing basic data for subsequent path connections and road network construction.

[0084] In some implementations, the distribution and weighting of sampling points are adjusted according to different needs. The magnitude of the weight values ​​directly affects the boundary positions of Voronoi cells, thus significantly impacting the accuracy of terrain segmentation. When optimizing the sampling data, sampling points in sensitive areas have higher weights, enabling the weighted Voronoi map to generate smaller Voronoi cells in these areas, refining the terrain subdivision. By constructing the weighted Voronoi map, the terrain of the field area is divided into multiple Voronoi cells, which have good spatial segmentation, facilitating subsequent path connection and optimization based on terrain features.

[0085] S302, combining the gradient information of each Voronoi unit with the terrain resistance coefficient, a multi-level weighted connection method is used to connect adjacent units to generate a preliminary road network candidate map;

[0086] In some implementations, this step combines the gradient information of each Voronoi cell with the terrain resistance coefficient, and uses a multi-level weighted connection method to connect adjacent cells, thereby generating a preliminary road network candidate map. This step establishes path connections between adjacent Voronoi cells by considering terrain variations and resistance characteristics, thus initially constructing the road network structure of the site.

[0087] In this embodiment, the path connection is primarily determined by two factors: the gradient information of each Voronoi unit and the terrain resistance coefficient. Gradient information represents the rate of terrain change; areas with larger gradients typically indicate greater terrain undulation and higher driving difficulty. The terrain resistance coefficient indicates the driving resistance in a specific area, which is related to slope, soil type, and vegetation cover. Combining this information helps optimize the path selection of the road network, enabling path connections to avoid high-gradient and high-resistance areas, thereby reducing construction and traffic costs.

[0088] Specifically, gradient information is calculated using terrain elevation data. Given any Voronoi cell... Elevation data, gradient This represents the rate of change of elevation at any point within the unit. The formula for calculating the gradient is: ,in The elevation is represented by x and y, which are the spatial coordinates within the cell. The slope information for each point within the cell is obtained by calculating the gradient, which is used for subsequent path selection.

[0089] Terrain drag coefficient The terrain resistance coefficient is set based on specific terrain attributes. In this embodiment, the terrain resistance coefficient is calculated based on slope and other influencing factors (such as surface type and vegetation cover). The specific resistance coefficient is represented by the following model: , where f is a function related to terrain attributes, the specific form of which depends on the terrain data and environmental model used. In some implementations, the terrain drag coefficient is obtained through polynomial fitting, interpolation, or a preset empirical value.

[0090] The path connection process uses a multi-level weighting method. Specifically, for each pair of adjacent Voronoi units... and The connection cost between them is calculated using a weighted method. It is determined by both gradient information and terrain drag coefficient. The calculation formula is: ,in and These are the weights for gradient information and terrain drag coefficient, respectively. This represents the difference in gradient between adjacent units. and These are the terrain drag coefficients of two Voronoi units. This weighted approach prioritizes areas with smaller gradient changes and lower drag when connecting routes, ensuring the accessibility and economy of the road network.

[0091] In this embodiment, the result of path connection is a graph structure containing multiple Voronoi units and their connecting paths, with the weight of each edge calculated using a weighted average. This graph structure serves as a preliminary candidate road network, providing foundational data for subsequent road network optimization and path selection.

[0092] S303, based on the topology of the preliminary road network candidate graph, combines the minimum spanning tree algorithm in graph theory to optimize and generate the road network, thus obtaining the final road network data.

[0093] In some implementations, this step utilizes the topology of the preliminary road network candidate graph and combines it with the Minimum Spanning Tree (MST) algorithm from graph theory to optimize and generate the road network, thus obtaining the final road network data. The MST algorithm optimizes the path connections in the preliminary road network candidate graph, reducing redundant path connections and generating a minimum-cost, highly connected road network structure.

[0094] In this embodiment, the topology of the preliminary road network candidate graph consists of multiple Voronoi cells and the connecting paths between them. This graph structure represents all potential path connection methods within the field area. The weight of each edge is usually obtained by a weighted connection method, representing the connection cost between two cells. Based on this, the minimum spanning tree algorithm is applied to optimize path connections, ensuring that the generated road network has the lowest overall construction cost while meeting the road network connectivity requirements.

[0095] The minimum spanning tree algorithm, given a connected graph, generates a subgraph containing all vertices and no cycles, with the minimum total weight of the edges, by selecting a certain number of edges. In this step, each node in the graph corresponds to a Voronoi cell, and each edge corresponds to a path connecting two adjacent cells. The edge weight is calculated by weighting the gradient difference and the terrain resistance coefficient. According to the definition in graph theory, the goal of the minimum spanning tree is to find a tree that contains all nodes in the graph and has the minimum sum of path weights. Specifically, given a graph G=(V,E), where V is the set of vertices representing Voronoi cells, and E is the set of edges representing the paths connecting cells. Each edge... weight The minimum spanning tree problem is calculated using the aforementioned weighted connection formula. The minimum spanning tree problem is described by the following mathematical model: Where T represents the set of edges in the minimum spanning tree. For the edge The goal is to select several edges from a candidate graph G such that these edges form a tree, all nodes in the tree are connected, and the total weight of the edges is minimized.

[0096] In this embodiment, the minimum spanning tree algorithm can be implemented using Kruskal's algorithm or Prim's algorithm. In some implementations, Kruskal's algorithm sorts all edges by weight and then sequentially selects the edge with the smallest weight and adds it to the spanning tree until the spanning tree contains all nodes. When selecting an edge, if the two nodes connected by the edge are already in the same spanning tree, the edge is skipped; otherwise, it is added to the spanning tree. This algorithm can efficiently generate a minimum spanning tree.

[0097] Another common implementation is using Prim's algorithm. This algorithm starts from an initial node and progressively expands the spanning tree, selecting the edge with the smallest weight at each step and connecting it to the tree, until the spanning tree contains all nodes. The advantage of Prim's algorithm is that it only needs to consider the edges of the currently generated portion when expanding the spanning tree, resulting in relatively low computational cost.

[0098] In practice, the minimum spanning tree algorithm proceeds through the following steps:

[0099] S3031, Sort all edges in the preliminary road network candidate graph to generate edge weight sequence data;

[0100] S3032, Based on the edge weight sequence data, the Kruskal algorithm or Prim algorithm is used to select the edges one by one, select the edge with the smallest weight, and connect the Voronoi units step by step until the spanning tree contains all units, generating the minimum spanning tree data;

[0101] S3033, based on the minimum spanning tree data, avoid loop generation, ensure that the selected path can connect all units and form a tree structure to generate road network data, the road network data containing the set of connected paths with the smallest weight among all Voronoi units;

[0102] By applying the minimum spanning tree algorithm, this step effectively optimizes the road network topology, reduces unnecessary path connections, and thus ensures that the road network maintains connectivity while achieving minimum construction cost and maximum efficiency. This process is a crucial step in road network generation, providing a reasonable preliminary framework for subsequent path optimization and construction.

[0103] S4. Based on road network data, path selection and constraint adjustment are performed to obtain a path candidate set;

[0104] For details, please refer to the appendix. Figure 4 As shown, this step includes the following sub-steps:

[0105] S401, based on road network data, uses heuristic search to calculate the shortest path from the starting point to the destination, generating a preliminary set of candidate paths;

[0106] In some implementations, this step combines road network data and heuristic search algorithms. In this embodiment, the heuristic search computation includes Dijkstra's algorithm for shortest path search and A* algorithm for heuristic estimation.

[0107] In this embodiment, the road network data consists of the aforementioned minimum spanning tree data and the connection paths between Voronoi units. Nodes represent key points in the road network, and edges represent connection paths between nodes. A heuristic search algorithm is used to calculate the shortest path from the starting node to the ending node, and path optimization is performed based on the cumulative cost of the current path and the estimated cost of the target node.

[0108] Dijkstra's algorithm selects a path by calculating the shortest distance among all possible paths, making it suitable for graphs with no negative edge weights. Specifically, the mathematical formula for Dijkstra's algorithm is: ,in Indicates the distance from the starting point to the node. The shortest path cost, Indicates from node To the node edge weights, It is the set of nodes in the graph. Dijkstra's algorithm continuously updates the shortest path value of each node until the shortest path from the start point to the end point is calculated.

[0109] In the A* algorithm, heuristic estimation is introduced into path search, making the search process depend not only on the actual cost from the current node to the starting point, but also on the estimated cost from the current node to the destination. The cost function of the A* algorithm is defined as: Here, g(n) represents the actual cost from the starting point to the current node n, h(n) is the heuristically estimated cost from node n to the destination, and f(n) is the total cost of node n. During the heuristic search process, the A* algorithm expands upon the node with the smallest f(n) value, thus finding the shortest path more efficiently.

[0110] The Dijkstra-A* hybrid algorithm combines Dijkstra's shortest path search with the heuristic estimation of the A* algorithm. It maintains Dijkstra's global optimization of the path while incorporating A*'s guidance on the target direction, thus improving the efficiency of path search. In implementation, the Dijkstra-A* hybrid algorithm follows these steps:

[0111] S4011, Initialize the path cost of the node, set the path cost of the starting point to 0, and the path cost of other nodes to infinity.

[0112] S4012, calculate the total cost of each node to be processed, including the actual cost and heuristic estimate, and select the node with the lowest cost for expansion;

[0113] S4013, update the path cost of adjacent nodes based on edge weights and heuristic estimates, and continue to expand nodes until the endpoint node is expanded or all nodes are processed.

[0114] S4014, generate the shortest path from the starting point to the ending point, and add the path to the preliminary path candidate set.

[0115] In some implementations, by combining heuristic estimation with shortest path search, the Dijkstra-A* hybrid algorithm can find the optimal path in a shorter time while avoiding blind search. This algorithm provides an efficient and accurate method for generating a candidate path set, offering preliminary candidate path data for subsequent path optimization.

[0116] S402, adjust the edge weights based on local environmental features, optimize the initial path candidate set using the edge weight adjustment model, and generate the adjusted path weight matrix;

[0117] In some implementations, this step incorporates local environmental features, including slope and turning radius, to further adjust the edge weights of the path, and then optimizes the initial path candidate set based on the adjusted weights to generate an adjusted path weight matrix. By considering local terrain and vehicle driving characteristics, the quality of the path is refined, providing more reasonable edge weights for subsequent path selection and optimization.

[0118] In this embodiment, the edge weights are adjusted dynamically by introducing a series of environmental factors to the edges in the initial path candidate set. Specifically, the formula for adjusting the edge weights is: ,in This indicates the adjusted edge weights. These are the original edge weights in the initial path candidate set. It is an adjustment factor related to slope. It is an adjustment factor related to the turning radius. and This is a coefficient that influences the weighting of slope and turning radius, controlling the degree of influence of various environmental factors in path weight adjustment. Slope Adjustment Factor This indicates the impact of the slope of each edge on the path on driving. Paths with steeper slopes increase the difficulty of driving; therefore, the adjustment factor is [not specified]. Calculated using the following formula: ,in It is a coefficient that controls the impact of slope on path weight. It is the slope angle of side e.

[0119] Turning radius adjustment factor This reflects the impact of path curves on driving difficulty. Paths with smaller turning radii have sharper curves, resulting in greater resistance for the vehicle when navigating them; therefore, their adjustment factors are... Calculated using the following formula: ,in It is a coefficient that controls the impact of the turning radius on the path weight. It is the turning radius of edge e.

[0120] By introducing adjustment factors for slope and turning radius, the weights of each edge on the path can be dynamically adjusted, ensuring that path selection considers not only the influence of terrain but also the adaptability of vehicle driving. The adjusted edge weights reflect the overall traffic capacity of the path and can more accurately express the relative advantages and disadvantages of each path.

[0121] In some implementations, the adjusted edge weight matrix is ​​used to optimize the initial path candidate set. By applying the edge weight adjustment model, more difficult paths (such as those with steep slopes or sharp turns) are assigned larger weights, thus reducing the probability of these paths being selected. Conversely, relatively flat paths with large turning radii have lower weight values, increasing their likelihood of being selected. The generated adjusted path weight matrix contains edge weight information for all paths and provides a refined path selection basis based on terrain features for subsequent path optimization. Through this step, path optimization is not limited to shortest path calculation but also fully considers road traffic performance and actual vehicle driving characteristics.

[0122] S403 optimizes the path based on the path weight matrix and the Lagrange multiplier method, eliminating paths that do not meet the constraints (minimum road width and maximum slope limits) and generating a path candidate set.

[0123] In some implementations, the constraints are minimum road width and maximum slope limits;

[0124] In this embodiment, the Lagrange multiplier method is applied to the path optimization process, aiming to minimize the path weights while satisfying a series of constraints. Assume the path weight matrix is ​​represented as... ,in Let be the weight of the i-th edge. The constraints include:

[0125] Minimum road width constraint: The road width of each path should be greater than or equal to a certain threshold. ,Right now: , ,in Let P represent the road width of path e, and let P be the set of paths.

[0126] Maximum slope limit: The slope of each path should be less than or equal to the maximum slope threshold. ,Right now: , ,in It is the slope of path e.

[0127] In some implementations, the optimization objective of the path is to minimize the sum of the path weights while satisfying the constraints mentioned above. To achieve this objective, a Lagrangian function can be constructed, incorporating the constraints into the objective function. The Lagrangian function takes the form: , in It is a Lagrange function. The adjusted edge weights obtained above. and These are Lagrange multipliers, related to the minimum road width constraint and the maximum gradient constraint, respectively. By solving the extremum problem of the Lagrange function, the set of optimal paths that satisfy the constraints can be obtained.

[0128] In some implementations, the balance between path weights and constraints is obtained by solving the partial derivatives of the Lagrange equation. In this embodiment, by taking the partial derivatives of the Lagrange function and setting them to zero, the following equation is obtained: , , The solutions to these equations form the set of optimal paths. This refers to the path that satisfies the minimum road width and maximum gradient constraints. The optimal path obtained will be the set of candidate paths with the minimum path weight that meet the constraints.

[0129] In this embodiment, the path candidate set generated based on this optimization process is constrained and optimized to eliminate all paths that do not meet the conditions, ensuring that the finally selected path has the lowest driving difficulty and the best driving efficiency while meeting the actual constraints.

[0130] S5. Based on the path candidate set and the facility distribution data, generate a road planning scheme suitable for wind and solar power plants, and output the final power plant road path data.

[0131] For details, please refer to the appendix. Figure 5 As shown, this step includes the following sub-steps:

[0132] S501, based on the path candidate set, uses electric field layout and equipment adaptation algorithms to modify the path structure, ensuring that the path avoids key equipment and infrastructure, optimizes the path layout, and generates the modified path structure.

[0133] In some implementations, the electric field layout and equipment adaptation algorithm includes path obstacle avoidance processing and path optimization processing. The aim is to refine the path candidate set, ensuring that paths avoid critical equipment and infrastructure within the electric field, while simultaneously optimizing the path layout. The algorithm comprises two main parts: path obstacle avoidance processing and path optimization processing. Path obstacle avoidance processing primarily analyzes the paths in the path candidate set, eliminating paths that conflict with equipment or infrastructure, and replans paths based on terrain and equipment distribution. Path optimization processing, based on the electric field's requirements and equipment distribution, optimizes path selection, resulting in a more rational final path layout that avoids obstacles affecting path safety or operational efficiency, while ensuring path continuity and optimality.

[0134] In this embodiment, the core of obstacle avoidance processing lies in modifying the shape or direction of the path to ensure that the path does not pass through these devices or facilities, based on the location constraints of the equipment and infrastructure. Assume there are several critical devices or infrastructures within the electric field. Each device Each has an area it occupies. The paths in the path candidate set are represented as The goal of path obstacle avoidance is to adjust the path. This ensures that the path does not interact with any device. occupied area overlapping.

[0135] In some implementations, path correction can be achieved by adding constraints; for example, the turning points of the path can be adjusted to prevent the path from approaching equipment areas. This can be handled by introducing an "obstacle avoidance function." For path p and device The goal is to maximize the shortest distance between the path and the device. ,Right now: In this embodiment, the path optimization process employs an adaptation algorithm based on electric field requirements. The goal of this algorithm is to optimize each path in the candidate path set, reducing path length or minimizing energy loss along the path. The objective of path optimization is represented by the following optimization function: ,in Representing a path The weights on the scale (are a weighted sum of factors such as slope and curvature). This represents the objective function for path optimization.

[0136] To consider the practical requirements of the electric field layout, path optimization also needs to take into account the adaptability requirements of the equipment, ensuring that the paths can reasonably connect various devices and serve the functional requirements of the electric field. For example, some paths need to pass through equipment areas to connect different devices; these paths should be designed in detail to ensure mutual adaptability with the equipment. For example, in this embodiment, the adaptability between devices and paths is optimized through the following constraints: ,in For path p and device The device adaptation function is given, where α is the device adaptation threshold. The optimization objective is to minimize the total weight of the path while satisfying device adaptation requirements. This optimization problem is solved by finding the minimization function f(P) and the constraints. The optimal solution is obtained by combining the following methods.

[0137] In summary, through the above steps, the candidate path set is structurally modified using the electric field layout and equipment adaptation algorithm to generate a revised path structure. This path structure avoids critical equipment and infrastructure, optimizes the path layout, meets the actual needs of the electric field, and ensures the safety, feasibility, and efficiency of the path.

[0138] S502, taking into account the construction and maintenance needs within the electric field, performs construction adaptability processing on the modified path to generate construction adaptability path data;

[0139] In some implementations, construction adaptability measures include path width adjustment, slope adjustment, turning radius optimization, and path smoothness improvement. The main purpose of these measures is to ensure that the path meets the spatial requirements for equipment installation and maintenance operations during construction, and to ensure good accessibility during construction and ease of maintenance later. Specifically, path width adjustment takes into account the passage of heavy machinery, equipment transportation, and construction tools; slope adjustment ensures that construction machinery and maintenance equipment can safely and smoothly traverse changes in terrain; turning radius optimization improves the efficiency of equipment transportation and construction vehicle passage; and path smoothness improvement facilitates precise equipment installation during construction and smooth maintenance operations later.

[0140] In this embodiment, the core of the construction adaptability treatment is based on the modified path structure. Adjustments are made to the path width, slope, turning radius, and ground flatness to meet the needs of construction and maintenance. The goal of path width adjustment is to ensure that the path can accommodate the necessary construction equipment, materials, and transportation vehicles. Let the width of path p be w(p), requiring that the path width at least meets the width required for equipment passage. ,Right now: If the original path width does not meet this condition, the path will be expanded, and the adjusted path width will be... Represented as: Furthermore, slope adjustment applies slope constraints to paths within the candidate path set to ensure that changes in path slope do not impede the passage of construction machinery. In some implementations, the path slope... It is calculated from the elevation differences and horizontal distances between various points on the path. If the slope of the path exceeds the preset maximum slope limit... If so, the path needs to be adjusted. Let z(p) be the height of a point on path p, and the height of two adjacent points on the path... The slope between them is expressed as: ,in This represents the horizontal distance between two adjacent points. The slope is calculated as follows. Greater than the maximum slope limit At this time, the slope of the path will be adjusted to meet the requirements by either smoothing the slope or selecting a more suitable path. The adjusted slope is expressed as follows: Optimizing the turning radius is primarily to ensure that the turning angle of the path is not too large, thus enabling large construction equipment to turn smoothly. Let there be two consecutive points on path p. The turning radius between them is The turning radius must be greater than or equal to the preset minimum turning radius.

[0141] Path smoothness improvement involves smoothing out uneven sections of the path to ensure smooth operation of construction equipment and prevent steep slopes or sharp turns from affecting work. Smoothness improvement is achieved through the following methods: for each segment of path p... and Calculate path curvature Furthermore, the curvature is controlled within a suitable range to avoid overly steep changes in the path. The formula for calculating the path curvature is: When the curvature exceeds the predetermined range, the path will reduce the curvature through smoothing. The adjusted curvature will be smoothed through methods such as path interpolation or curve fitting.

[0142] In some implementations, the result of construction adaptability processing is the generation of construction adaptability path data that adapts to the needs of equipment installation and maintenance. This path data meets the requirements of construction machinery, equipment transportation and installation, and subsequent maintenance, and can effectively avoid obstacles on the path during actual construction, ensuring the safety and efficiency of construction. The width, slope, turning radius, and flatness of the path all meet the requirements of construction operations inside the electric field, and the final generated construction adaptability path data provides effective support for subsequent construction stages.

[0143] S503, based on construction adaptability path data and combined with electric field operation and safety requirements, performs path smoothing and correction processing to generate the final electric field road path data.

[0144] In some implementations, path smoothing and correction processing includes path curve smoothing, slope reduction, turning radius correction, and safety zone avoidance processing. This further improves the path's traffic capacity and ensures its stability to meet the safety requirements of equipment passage and personnel operation within the electric field. In this embodiment, the final path smoothing and correction processing involves the following aspects: path curve optimization, slope correction and smoothing, turning radius optimization, and avoidance of environmental features that meet the safety requirements within the electric field.

[0145] Path curve optimization aims to improve the smoothness of a path and avoid traffic obstructions caused by sharp bends or discontinuous curves. In some implementations, each segment of the path curve is smoothed. Let the curvature of the path curve be... Therefore, curve optimization can be achieved by reducing the rate of curvature change. The amount of curvature change between each segment of the path. Represented as: To smooth the path curve, the goal of path adjustment is to correct the path using curve fitting or spline interpolation techniques (such as B-spline or cubic spline interpolation) to make the curvature change as gradual as possible. After curve optimization, the new curvature... The curve will meet certain smoothness requirements to avoid drastic changes in curvature. The optimized path can be controlled by the following constraints: ,in This represents the maximum permissible curvature of the path curve.

[0146] Slope correction and leveling are primarily to ensure that the path slope meets the requirements for electric field operation. The slope within the electric field should accommodate the equipment's passage, avoiding excessively steep slopes that could lead to construction and maintenance difficulties. In some implementations, the path slope... Based on the set maximum slope limit Corrections are made. When the slope of the path exceeds the limit, the path will be smoothed through interpolation or reconstruction to ensure that the slope of each segment does not exceed the maximum slope limit. The corrected slope is expressed as: .

[0147] Turning radius correction is used to ensure that the turning angle of the path accommodates the passage requirements of mechanical equipment within the electric field. In some implementations, the turning radius of the path... The minimum turning radius has been revised to meet electric field safety requirements. The corrected turning radius is ensured to meet this requirement through local adjustments to the path: This formula represents the turning radius of the path. With minimum turning radius The radius with the larger value is used as the corrected turning radius. The path correction also needs to avoid conflicts with critical areas, equipment, and safety zones within the electric field. Therefore, during path optimization, safety zone avoidance is performed based on the functional zone division and environmental characteristics within the electric field. Let the safety zone within the electric field be... Path p should avoid being associated with Overlap. The path correction process adjusts the path's position according to the area's avoidance requirements, ensuring the corrected path does not intersect with the safe zone. The safety of the corrected path can be controlled by the following constraints: the relationship between path p and the safe zone. Non-overlapping, i.e. This indicates the path P and the safe zone. There is no overlap; otherwise, the value is 0, indicating overlap and intersection.

[0148] In this embodiment, the final smoothing and correction processing generates path data that meets the safety requirements for electric field operation. This path data not only ensures smooth equipment passage during construction and transportation but also takes into account the environmental and safety requirements within the electric field, avoiding interference with the functional areas, equipment, and safety zones of the electric field. Through the above series of smoothing and correction processes, the final path not only possesses construction adaptability but also meets the actual operation and safety requirements of the electric field, generating the final electric field road path data, providing effective support for subsequent construction, operation, and maintenance.

[0149] Based on the description of the above embodiments of the wind and solar power plant site road planning method based on terrain analysis, this application also discloses a wind and solar power plant site road planning system based on terrain analysis. This system can be a computer program (including program code) that runs the aforementioned wind and solar power plant site road planning method based on terrain analysis. Please see the appendix. Figure 6 As shown, the wind and solar power plant site road planning system based on terrain analysis can operate the following units:

[0150] The acquisition unit 110 is used to acquire data to be processed, which includes digital elevation model data of the wind and solar power plant area, vehicle turning radius parameters, and facility distribution data. The digital elevation model data includes elevation data and DEM pixel size, and the facility distribution data is the location and distribution information of all relevant equipment and infrastructure in the power plant.

[0151] The optimized sampling unit 120 is used to perform hierarchical reconstruction and optimization processing on the data to be processed, and generate optimized sampling data.

[0152] The road network construction unit 130 is used to construct a road network based on the optimized sampling data, thereby obtaining road network data;

[0153] The route filtering unit 140 is used to filter routes and adjust constraints based on road network data, thereby obtaining a set of candidate routes;

[0154] The power station path generation unit 150 is used to generate power station road path data suitable for wind and solar power plants based on the path candidate set and the facility distribution data.

[0155] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A road planning method for wind and solar power plant sites based on terrain analysis, characterized in that, The method includes the following steps: S1, acquire data to be processed, the data to be processed includes digital elevation model data of the wind and solar power station area, vehicle turning radius parameters and facility distribution data, the digital elevation model data includes elevation data and DEM pixel size, and the facility distribution data is the location and distribution information of all relevant equipment and infrastructure in the power station; S2, performs hierarchical reconstruction and optimization on the data to be processed, and generates optimized sampling data; S3, construct a road network based on the optimized sampling data to obtain road network data; S4. Based on road network data, path selection and constraint adjustment are performed to obtain a path candidate set; S5, Based on the path candidate set and the facility distribution data, generate power station road path data suitable for wind and solar power plants; Step S2 includes the following sub-steps: S201, perform multi-scale decomposition on the elevation data based on gradient decomposition, extract terrain features at different scales, and set an adaptive sampling step size according to each terrain feature. Compare the sampling step size with the DEM cell size. When the sampling step size is greater than the cell size, perform sampling operation to generate terrain scale data. S202 defines local path-sensitive areas based on terrain-scale data, vehicle turning radius, and slope information, and optimizes the sampling density of the sensitive areas through weighted convolution to generate optimized sampling data; Step S4 includes the following sub-steps: S401: Based on road network data, the shortest path from the starting point to the destination is calculated through heuristic search, generating a preliminary set of candidate paths; S402, adjust the edge weights based on local environmental features, optimize the initial path candidate set using the edge weight adjustment model, and generate the adjusted path weight matrix; S403, based on the path weight matrix and combined with the Lagrange multiplier method, performs constraint optimization on the paths, eliminates paths that do not meet the constraints, and generates a path candidate set. The edge weight adjustment model is as follows: ,in This indicates the adjusted edge weights. These are the original edge weights in the initial path candidate set. It is an adjustment factor related to slope. It is an adjustment factor related to the turning radius. and It is a coefficient that influences the weight of slope and turning radius; Step S5 includes the following sub-steps: S501 uses electric field layout and equipment adaptation algorithms based on the path candidate set to modify the path structure, ensuring that the path avoids key equipment and infrastructure, and optimizes the path layout to generate the modified path structure. S502, combining the construction and maintenance needs within the electric field, performs construction adaptability processing on the modified path structure, thereby generating construction adaptability path data; S503, based on construction adaptability path data and combined with electric field operation and safety requirements, performs path smoothing and correction processing to generate the final electric field road path data.

2. The method for road planning in wind and solar power plant areas based on terrain analysis according to claim 1, characterized in that, Step S3 includes the following sub-steps: S301, based on optimized sampling data, uses a weighted Voronoi diagram partitioning algorithm to spatially segment the terrain of the field area, obtaining Voronoi units for each terrain; S302, combining the gradient information of each Voronoi unit with the terrain resistance coefficient, a multi-level weighted connection method is used to connect adjacent units to generate a preliminary road network candidate map; S303, based on the topology of the preliminary road network candidate graph, combined with the minimum spanning tree algorithm in graph theory, optimizes and generates the road network to obtain road network data; The multi-level weighted connection method is for each pair of adjacent Voronoi units V i and V j The connection cost between them is calculated using a weighted method. The calculation formula is as follows: ,in and These are the weights for gradient information and terrain drag coefficient, respectively. This represents the difference in gradient between adjacent units. and These are the terrain drag coefficients of two Voronoi units, respectively.

3. The method for road planning in wind and solar power plant areas based on terrain analysis according to claim 2, characterized in that, Step S303 includes the following sub-steps: The minimum spanning tree algorithm is performed through the following steps: S3031, Sort all edges in the preliminary road network candidate graph to generate edge weight sequence data; S3032, Based on the edge weight sequence data, the Kruskal algorithm or Prim algorithm is used to select the edges one by one, select the edge with the smallest weight, and connect the Voronoi units step by step until the spanning tree contains all units, generating the minimum spanning tree data; S3033, based on the minimum spanning tree data, avoid loop generation, ensure that the selected path can connect all units and form a tree structure to generate road network data, the road network data containing the set of connected paths with the smallest weight among all Voronoi units.

4. The method for road planning in wind and solar power plant areas based on terrain analysis according to claim 1, characterized in that, The heuristic search computation described in S401 includes Dijkstra's algorithm for shortest path search and... A* Algorithm The heuristic estimation is performed according to the following steps: S4011, Initialize the path cost of the node, set the path cost of the starting point to 0, and the path cost of other nodes to infinity. S4012, calculate the total cost of each node to be processed, including the actual cost and heuristic estimate, and select the node with the lowest cost for expansion; S4013, update the path cost of adjacent nodes based on edge weights and heuristic estimates, and continue to expand nodes until the endpoint node is expanded or all nodes are processed. S4014, generate the shortest path from the starting point to the ending point, and add the path to the preliminary path candidate set.

5. The method for road planning in wind and solar power plant areas based on terrain analysis according to claim 1, characterized in that, The constraints in S403 are minimum road width and maximum slope limits.

6. The method for road planning in wind and solar power plant areas based on terrain analysis according to claim 1, characterized in that, The electric field layout and equipment adaptation algorithm in S501 includes path obstacle avoidance processing and path optimization processing; the construction adaptability processing in S502 includes path width adjustment, slope adjustment, turning radius optimization, and path smoothness improvement processing; the path smoothing and correction processing in S503 includes path curve smoothing, slope equalization, turning radius correction, and safe area avoidance processing.

7. A road planning system for wind and solar power plant sites based on terrain analysis, characterized in that, The system includes: The acquisition unit is used to acquire data to be processed, which includes digital elevation model data of the wind and solar power plant area, vehicle turning radius parameters, and facility distribution data. The digital elevation model data includes elevation data and DEM pixel size, and the facility distribution data is the location and distribution information of all relevant equipment and infrastructure within the power plant. The optimized sampling unit is used to perform hierarchical reconstruction and optimization of the data to be processed, generating optimized sampling data. A road network construction unit is used to construct a road network based on the optimized sampling data, thereby obtaining road network data; The route filtering unit is used to filter routes and adjust constraints based on road network data, thereby obtaining a set of candidate routes; The power plant path generation unit is used to generate power plant road path data suitable for wind and solar power plants based on the path candidate set and the facility distribution data. The optimized sampling unit performs the following sub-steps: S201, perform multi-scale decomposition on the elevation data based on gradient decomposition, extract terrain features at different scales, and set an adaptive sampling step size according to each terrain feature. Compare the sampling step size with the DEM cell size. When the sampling step size is greater than the cell size, perform sampling operation to generate terrain scale data. S202 defines local path-sensitive areas based on terrain-scale data, vehicle turning radius, and slope information, and optimizes the sampling density of the sensitive areas through weighted convolution to generate optimized sampling data; The path filtering unit performs the following sub-steps: S401: Based on road network data, the shortest path from the starting point to the destination is calculated through heuristic search, generating a preliminary set of candidate paths; S402, adjust the edge weights based on local environmental features, optimize the initial path candidate set using the edge weight adjustment model, and generate the adjusted path weight matrix; S403, based on the path weight matrix and combined with the Lagrange multiplier method, performs constraint optimization on the paths, eliminates paths that do not meet the constraints, and generates a path candidate set. The edge weight adjustment model is as follows: ,in This indicates the adjusted edge weights. These are the original edge weights in the initial path candidate set. It is an adjustment factor related to slope. It is an adjustment factor related to the turning radius. and It is a coefficient that influences the weight of slope and turning radius; The power plant path generation unit performs the following sub-steps: S501 uses electric field layout and equipment adaptation algorithms based on the path candidate set to modify the path structure, ensuring that the path avoids key equipment and infrastructure, and optimizes the path layout to generate the modified path structure. S502, combining the construction and maintenance needs within the electric field, performs construction adaptability processing on the modified path structure, thereby generating construction adaptability path data; S503, based on construction adaptability path data and combined with electric field operation and safety requirements, performs path smoothing and correction processing to generate the final electric field road path data.

Images