A method and system for partitioning power generation units based on mountain photovoltaics

Abstract

This application provides a method and system for zoning photovoltaic power generation units based on mountainous areas. This method involves gridding photovoltaic panel strings using a set aggregation distance to generate aggregation areas of varying sizes, forming a set of photovoltaic panel aggregation units. Based on this, the aggregation unit set is optimized according to the installed capacity requirements of the power generation units to obtain a pre-processed capacity unit set. The pre-processed capacity unit set is further segmented by spatial location and quantity to generate a power generation unit set. This method effectively addresses the problems of low segmentation accuracy, high computational complexity, and uneven zoning in the prior art.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic power generation, and in particular to a method and system for zoning power generation units based on mountain photovoltaics. Background Art

[0002] For mountain photovoltaic power generation, the rational division of power generation units is crucial, especially for pre-assembled photovoltaic panels. By dividing power generation units according to designated installed capacity, solar resources can be fully utilized, shadows and obstructions can be minimized, and overall power generation efficiency can be improved. Furthermore, rational unit division simplifies system maintenance and management, enabling faster fault location and repair, and reducing downtime. Furthermore, balancing the load across units prevents overloads and underloads, improving system stability and reliability. The unitized design also allows for more flexible system expansion and reduces construction and maintenance costs by reducing the use of cables and connecting equipment.

[0003] Existing power generation unit division mainly relies on planar segmentation methods. This involves vertically or horizontally dividing photovoltaic panels into specified numbers, either from top to bottom or from left to right. Panels that meet certain criteria are grouped together, and the results are iterated repeatedly. However, existing planar segmentation techniques suffer from low segmentation accuracy, often grouping multiple photovoltaic panels that are too far apart and highly discrete within the same unit. This results in complex photovoltaic panel routing, wasteful wiring, and extreme uneconomical results. Furthermore, local segmentation methods are prone to leaving irregularly shaped photovoltaic panel areas when dividing power generation units, increasing computational complexity. Summary of the Invention

[0004] In view of the above actual situation, this application proposes a power generation unit partitioning system based on mountain photovoltaics and a system thereof to solve the problems of low partitioning accuracy, high computational complexity and unbalanced partitioning in the existing technology.

[0005] A method for zoning power generation units based on mountain photovoltaics, characterized in that the method comprises the following steps:

[0006] S1, obtaining an aggregation unit set of photovoltaic panels, wherein the aggregation unit set is obtained by performing grid aggregation on all photovoltaic panel strings within a set aggregation distance, and aggregating them into areas of different sizes according to the distribution of photovoltaic panels in the grid;

[0007] S2, performing installed capacity optimization processing based on the installed capacity requirements of the power generation units of the aggregated unit set to obtain a pre-processed capacity unit set;

[0008] S3, partitioning the power generation units based on the pre-processing capacity unit set, wherein the partitioning is performed by dividing the pre-processing capacity units according to the spatial position and quantity conditions, thereby obtaining a power generation unit set.

[0009] Furthermore, the step S1 includes the following sub-steps:

[0010] S11, based on the terrain characteristics and site boundaries of the photovoltaic panel area, the boundaries of the photovoltaic panel are calculated using a dynamic boundary recognition algorithm to generate boundary data for grid division;

[0011] S12, based on the set maximum aggregation distance, the boundary data is divided and processed by the adaptive grid division model to generate a grid;

[0012] S13, traversing each of the grids using a Hilbert R-tree, locating and recording the photovoltaic panel positions within each grid based on the four points of the grid, and generating a photovoltaic panel position set;

[0013] S14, constructing a preliminary aggregation unit based on the photovoltaic panel location set, and expanding the aggregation area according to the neighboring grid relationship;

[0014] S15, the eight-neighborhood search algorithm is used to further expand the aggregation unit and finally form an aggregation unit set.

[0015] Furthermore, the specific calculation formula of the adaptive grid division model in S12 is as follows: ,in is the adaptive side length of the grid unit, D is the maximum aggregation distance, and K is the adjustment coefficient with a value range of , is the basic expansion scale factor, is the sensitivity coefficient, The average slope of the grid cells, is the terrain relief of the grid cell.

[0016] Furthermore, the S13 includes the following sub-steps:

[0017] S131, based on the grid, defining the boundary points of each grid, wherein the boundary points constitute the spatial query conditions of the Hilbert R-tree search;

[0018] S132, obtaining the two-dimensional coordinate data of the photovoltaic panel, and mapping it to the Hilbert space to generate a corresponding Hilbert index value;

[0019] S133, constructing a Hilbert R-tree structure based on the Hilbert index value of the photovoltaic panel;

[0020] S134, by traversing the boundary points of each grid, using the Hilbert R-tree to perform a range query, and finding and locating all photovoltaic panels within the current grid cell;

[0021] S135 , recording the found photovoltaic panel positions as a photovoltaic panel position set.

[0022] Furthermore, the step S14 includes the following sub-steps:

[0023] S141, constructing a preliminary aggregation unit, performing preliminary aggregation based on the set of photovoltaic panel positions in each grid according to spatial proximity relationships;

[0024] S142, performing aggregation expansion based on the spatial proximity between grids and the distribution density of photovoltaic panel locations. If a photovoltaic panel location set also exists in an adjacent grid of a certain grid, it is incorporated into the current aggregation unit and the expansion continues to its surrounding adjacent grids until the photovoltaic panel location set of the adjacent grid is empty or does not meet the expansion conditions.

[0025] S143: Determine the number of grids and layout structure of each aggregation unit based on the aggregation expansion.

[0026] Furthermore, the step S2 includes the following sub-steps:

[0027] S21, for each aggregation unit, calculating the geometric centroid position of the aggregation unit;

[0028] S22, based on the geometric centroid position, calculating the distance from each photovoltaic panel to the centroid and sorting them to determine the photovoltaic panel farthest from the centroid;

[0029] S23, based on the installed capacity requirements of the power generation unit, discard the photovoltaic panels farthest from the centroid in sequence until the number of photovoltaic panels meets the capacity requirements;

[0030] Furthermore, the S21 calculates the geometric centroid position of the aggregation unit based on the position coordinates of all photovoltaic panels in the aggregation unit. shared within photovoltaic panels, the position coordinates of each photovoltaic panel are ,in , the centroid The calculation formula is as follows: .

[0031] Furthermore, the S3 step includes the following sub-steps:

[0032] S31, based on the pre-processed capacity unit set, calculating the boundaries of the photovoltaic panel cluster and generating a boundary parameter set, combining the aspect ratio information and the requirement of an integer multiple of the installed capacity of the power generation unit, segmenting the pre-processed capacity unit set to form a candidate set of power generation unit segmentation;

[0033] S32, based on the candidate set of power generation unit partitioning, each unit after partitioning is verified using a connectivity capacity verification algorithm. If the conditions are not met, a partition adjustment unit set is generated;

[0034] S33, based on the partition adjustment unit set, borrowing photovoltaic panels from the opposing partition units according to the distance relationship between adjacent units to form a corrected power generation unit partition set;

[0035] S34 , based on the corrected power generation unit partition set, gradually check whether the installed capacity of all partition units meets the requirements, and finally output the power generation unit set.

[0036] Furthermore, the east-west boundaries of the four-boundary range of the photovoltaic panel cluster are w,e and the north-south boundaries are s,n. The boundary parameter set is the length L and width W of the four-boundary range, where L=ew, W=ns. The pre-processing capacity unit set is divided according to the following conditions: when L>D, it is divided longitudinally, and when L≤D, it is divided horizontally, where D is the maximum aggregation distance.

[0037] In addition, the present application also discloses a power generation unit zoning system based on mountain photovoltaics, characterized in that the system includes:

[0038] An aggregation unit generation unit is used to obtain an aggregation unit set of photovoltaic panels. The aggregation unit set is obtained by gridding all photovoltaic panel strings within a set aggregation distance and aggregating them into areas of different sizes according to the distribution of photovoltaic panels in the grid;

[0039] an installed capacity optimization unit, configured to perform installed capacity optimization processing on the aggregated unit set according to the installed capacity requirements of the power generation unit, to obtain a pre-processed capacity unit set;

[0040] The power generation unit partitioning unit is used to partition the pre-processing capacity unit set into power generation units, wherein the partitioning is performed by dividing the pre-processing capacity units according to the spatial position and quantity conditions, thereby obtaining the power generation unit set.

[0041] The present application proposes a power generation unit partitioning system based on mountain photovoltaics and its system, which realizes the preliminary classification of photovoltaic panel positions through grid clustering, avoids the calculation of the position relationship between photovoltaic panels one by one, greatly reduces the computational complexity and improves efficiency. At the same time, the plane segmentation method is combined to ensure the global consistency of the partitioning, avoiding the problem of insufficient number or irregular shape of the remaining photovoltaic panels due to the completion of the early partitioning, thereby improving the integrity and rationality of the partitioning. In addition, the fast search mechanism of the Hilbert R tree can be used to correct unreasonable partitions during each partitioning process. By adjusting and merging adjacent photovoltaic panels, the results of each partitioning meet the set distance and constraint requirements, thereby optimizing the partitioning effect and further improving the stability and scalability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1This is a schematic diagram comparing the effects of a mountain photovoltaic power generation unit partitioning method and a plane segmentation method proposed in this application;

[0043] Figure 2 This is a schematic diagram of example data in a method for zoning power generation units based on mountain photovoltaics proposed in this application;

[0044] Figure 3 A flow chart of a method for zoning power generation units based on mountain photovoltaics proposed in this application;

[0045] Figure 4 This is a schematic diagram of the process of forming an aggregated unit set in a method for partitioning power generation units based on mountain photovoltaics proposed in this application;

[0046] Figure 5 This is a schematic diagram of the results of rapid grid aggregation of example data in a method of partitioning power generation units based on mountain photovoltaics proposed in this application;

[0047] Figure 6 This is a schematic diagram of the result of clustering and segmenting into different numbers of parts on both sides in a mountain photovoltaic power generation unit partitioning method proposed in this application;

[0048] Figure 7 This is a schematic diagram of a partition adjustment unit set that needs to be adjusted in a power generation unit partitioning method based on mountain photovoltaics proposed in this application;

[0049] Figure 8 This is a schematic diagram of the processing flow of the example data secondment process proposed in this application, in which lower cluster 1 is seconded by upper cluster 1 and meets the quantity requirements of upper cluster 1;

[0050] Figure 9 This is a schematic diagram of the processing flow of the example data secondment process proposed in this application, in which upper cluster 2 is seconded by lower cluster 2 to meet the number of lower cluster 2;

[0051] Figure 10 This is a schematic diagram of a processing flow chart in which upper cluster 2 is seconded by lower cluster 1 and meets the quantity of lower cluster 2 in the example data secondment process proposed in this application;

[0052] Figure 11 This is a schematic diagram showing the results of a power generation unit set using example data in a power generation unit partitioning method based on mountain photovoltaics proposed in this application;

[0053] Figure 12 A schematic diagram of a process for partitioning a pre-processed capacity unit set into power generation units in a power generation unit partitioning method based on mountain photovoltaics proposed in this application to obtain a power generation unit set;

[0054] Figure 13A schematic diagram of the structure of a mountain photovoltaic power generation unit partitioning system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] The following will clearly and completely describe the simulation technology route in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0056] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0057] Existing methods for dividing power generation units mainly rely on plane segmentation technology, which divides photovoltaic panels vertically or horizontally according to a specified number from top to bottom or from left to right until the conditions are met and they are classified into one category. However, this method often results in a large number of disconnected power generation units in the partitioning result, increasing the complexity of cable routing and wasting materials. Please refer to the attached Figure 1 ,This traditional partitioning method has a significant difference in effect compared to the ,partitioning method of this application.,In traditional plane partitioning, local partitioning easily ignores the ,overall layout, resulting in unreasonable partitioning.

[0058] For example, for a small area of ​​photovoltaic panels facing south, the traditional plane segmentation method is prone to uneven partitioning or overly dispersed shapes. In contrast, the partitioning method of the present invention shows a more efficient partitioning effect. Figure 2 As shown in the figure, the power generation unit division is optimized by further processing the original data graph. The power generation unit division method based on mountain photovoltaic proposed in this application is a reasonable division optimization of the example graph. Please refer to the attached Figure 3 As shown, the application method includes the following steps:

[0059] S1, obtaining an aggregation unit set of photovoltaic panels, wherein the aggregation unit set is obtained by performing grid aggregation on all photovoltaic panel strings within a set aggregation distance, and aggregating them into areas of different sizes according to the distribution of photovoltaic panels in the grid;

[0060] Specifically, this step includes the following sub-steps:

[0061] S11, based on the terrain characteristics and site boundaries of the photovoltaic panel area, the boundaries of the photovoltaic panel are calculated using a dynamic boundary recognition algorithm to generate boundary data for grid division;

[0062] This step uses a dynamic boundary recognition algorithm to calculate the four boundaries of the PV panel distribution area, based on the topographical characteristics and site boundaries of the PV panel area. This calculation then generates boundary data for subsequent gridding. This step ensures that the boundary data accurately reflects the actual site conditions and the spatial distribution of the PV panels, providing a reliable foundation for subsequent gridding and aggregation.

[0063] Specifically, in this embodiment, the terrain features are three-dimensional terrain information obtained from a digital elevation model or lidar scanning, including slope, altitude, and terrain relief parameters. The site boundary is determined by using a geographic information system or existing site planning data to define the site's external boundaries. The site boundary includes natural obstacles such as rivers and mountains, or artificial boundaries such as fences and roads.

[0064] In some embodiments, the dynamic boundary recognition algorithm is designed to calculate the boundaries that adapt to complex terrain based on the actual distribution of photovoltaic panels and the constraints of terrain boundaries. The dynamic boundary recognition algorithm includes preliminary boundary calculation, boundary correction and expansion, and determination of boundaries. In this embodiment, the preliminary boundary calculation is to calculate the preliminary boundaries of the photovoltaic panels using the convex hull algorithm, which finds the smallest convex polygon that contains all photovoltaic panels in two-dimensional space. Let the coordinate set of the photovoltaic panels be , each point Represents the geographical location of the photovoltaic panel. The calculation formula of the convex hull H is: Since the convex hull cannot fully reflect the actual boundary of complex terrain, the preliminary boundary needs to be corrected and expanded. In this embodiment, the boundary correction and expansion is to adjust the boundary shape according to the terrain characteristics through the boundary expansion model. Specifically, the boundary expansion model formula is: ,in is the corrected boundary. ⊕ represents the boundary extension operation. E is the set of extension vectors that define the extension distance of the boundary in each direction. The calculation of the extension vector E is based on terrain characteristic parameters, including slope, terrain relief, and site boundary restrictions. The specific calculation is as follows: ,in is the expansion distance of the i-th boundary point. To dynamically adjust the coefficient and control the expansion amplitude. is the terrain adaptation function, based on the slope of the boundary point and undulating terrain Calculate the expansion distance. Terrain adaptation function Defined as: , is the basic expansion scale factor, is the sensitivity coefficient, is the slope of the i-th boundary point, The function takes into account the slope and terrain relief to ensure that the boundary expansion is reasonable under complex terrain conditions. Next, the four-boundary range calculation is performed based on the corrected boundary. Calculate the four boundaries of the photovoltaic panel distribution area. Assume that the four boundary points are: the easternmost point westernmost point southernmost point northernmost point , the calculation formula of the four-range is as follows: , the bounds are defined as a rectangular area: bounds (boundary data) = It is worth mentioning that the dynamic adjustment mechanism adjusts the expansion coefficient Adapt to different terrain complexities and ensure flexibility and accuracy of boundary calculation. The adjustment process is automatically adjusted based on terrain characteristic parameters. Value. Adjustment mechanism formula: ,in: is the basic expansion scale factor. is the sensitivity coefficient, control Speed ​​of response to changes in terrain characteristics. is the average slope. is the terrain relief. This formula ensures that when the slope and terrain relief are large, the expansion distance E is appropriately increased to adapt to the boundary changes of complex terrain; in flat areas, the expansion distance is kept small to avoid over-expansion.

[0065] S12, based on the set maximum aggregation distance, the boundary data is divided and processed by the adaptive grid division model to generate a grid;

[0066] The purpose of this step is to generate a grid structure for subsequent steps based on the boundaries of the PV panel distribution area and the set maximum aggregation distance D. This grid structure not only facilitates PV panel management and calculations, but also ensures that the spacing between PV panels within each grid cell meets the specified aggregation requirements.

[0067] In this embodiment, the maximum convergence distance D refers to the maximum allowable distance between any two photovoltaic panels to ensure effective convergence. To achieve this, an adaptive meshing model is used to process boundary data and generate a qualified mesh. Based on the maximum convergence distance D, this model calculates the side length and diagonal length of each grid cell to ensure that the relative distance between all photovoltaic panels within the grid is less than D.

[0068] Specifically, in some embodiments, the side length L of the grid is determined by the maximum aggregation distance D, and the calculation formula is as follows: In a two-dimensional plane, the diagonal length of a square grid unit is equal to the square root of the sum of the squares of its side lengths. Since the distance between any two photovoltaic panels in the grid is required to not exceed D, the diagonal length of the grid must be equal to or less than D, that is, the condition is met: , so the side length L of the grid is calculated as the above formula. Through this calculation, the distribution area of ​​the entire photovoltaic panel will be divided into a series of grid units of uniform size. Assume that the distribution area of ​​the entire photovoltaic panel is a rectangular area, and define its four boundaries as , the area will be divided into multiple grid cells , where i and j represent row and column coordinates respectively. The total number of grid cells is determined by the four-dimensional range and the grid side length. The formula for the number of grid divisions is: , , where row and col are the number of rows and columns after the grid is divided, and L is the side length of the grid. In this embodiment, the generated grid unit set is expressed as: , each grid cell Represents a square area in the photovoltaic panel distribution area, with a side length of L. Since the diagonal length of the grid division is equal to D, there is no need to further determine the distance between photovoltaic panels within the grid unit, because the mutual distance of all photovoltaic panels in the grid must be less than the diagonal length D.

[0069] In some embodiments, the adaptive meshing model not only divides the grid based on the maximum aggregation distance D, but also considers the terrain conditions and the distribution density of photovoltaic panels, dynamically adjusting the grid size to better adapt to complex terrain and optimize the spatial distribution of photovoltaic panels. Specifically, the adaptive adjustment formula for the grid side length L is as follows: , For grid cells k is the adjustment coefficient, which is used to control the amplitude of the adaptive adjustment. is the terrain adaptation function , based on the grid cells The average slope within and terrain relief Calculate the adjustment factor. Through the above formula, the grid division model can dynamically adjust the grid side length according to the terrain characteristics of different grid cells. In areas with steep slopes or significant terrain undulations, the side length of the grid cells is appropriately reduced to more accurately capture the distribution of photovoltaic panels; while in areas with flat terrain or low photovoltaic panel distribution density, the size of the grid cells is appropriately increased to reduce the amount of calculation and storage requirements.

[0070] In addition, in order to ensure the aggregation effect of photovoltaic panels in each grid unit, this embodiment also introduces an aggregation verification mechanism. This mechanism verifies the actual distribution of photovoltaic panels in each grid unit to confirm whether their spacing is indeed less than the maximum aggregation distance D. The specific steps are as follows: If , then adjust the grid boundary or re-divide the grid, where For grid cells The actual distance between any two photovoltaic panels within the grid is D. If the spacing between some photovoltaic panels exceeds D, the grid boundaries are dynamically adjusted or the grid cells are re-divided to ensure that the final grid meets the aggregation requirements. Dynamic adjustments include boundary expansion or contraction and local grid re-division. In summary, the mathematical expression of the adaptive grid division model is as follows: .in: is a grid cell The adaptive side length. D is the maximum aggregation distance set. K is the adjustment coefficient with a value range of 0 <k≤1。 is the basic expansion scale factor, is the sensitivity coefficient. Grid Cell The average slope within. For grid cells The terrain relief within.

[0071] S13, traversing each of the grids using a Hilbert R-tree, locating and recording the photovoltaic panel positions within each grid based on the four points of the grid, and generating a photovoltaic panel position set;

[0072] In this example, to efficiently and accurately determine the distribution of photovoltaic panels within each grid cell, a Hilbert R-tree algorithm is used to traverse the grid. Based on the generated grid structure, the photovoltaic panel positions within each grid cell are quickly located and recorded, generating a set of photovoltaic panel positions, providing an accurate data foundation for subsequent calculations and optimization.

[0073] The Hilbert R-tree, as a spatial index structure, efficiently handles the retrieval and query of multidimensional spatial data. It combines the hierarchical structure of the R-tree with the spatial filling properties of the Hilbert curve, making the distribution of spatial data more compact and enabling rapid retrieval of photovoltaic panel locations within a limited spatial range. The Hilbert curve maintains data locality by mapping multidimensional spatial data into a one-dimensional curve, ensuring that adjacent spatial points remain close together in the one-dimensional mapping. This enables the Hilbert R-tree to quickly locate the boundaries of each grid cell when processing photovoltaic panel location data and efficiently retrieve the panel location within these boundaries.

[0074] Specifically, the principle of traversal search is to use the four points of each grid cell as the boundary conditions for retrieval. The four points of the grid cell are defined as , the point set consists of the minimum and maximum boundaries of the grid cell, representing the spatial range of the grid cell. The entire photovoltaic panel area is indexed by the Hilbert R-tree to locate all photovoltaic panels in the current grid cell. In order to improve the efficiency of data processing, this step is done by converting the two-dimensional coordinate information of the photovoltaic panel into The one-dimensional index value is mapped to the Hilbert space and a fast lookup is performed on the index to determine the location of the photovoltaic panel within the grid cell in a short time.

[0075] In some embodiments, the specific steps are as follows:

[0076] S131, based on the grid, defining the boundary points of each grid, wherein the boundary points constitute the spatial query conditions of the Hilbert R-tree search;

[0077] Specifically, based on the grid cells generated in step S12, define the boundary points of each grid cell. ,These boundary points constitute the spatial query conditions of Hilbert R-tree retrieval.

[0078] S132, obtaining the two-dimensional coordinate data of the photovoltaic panel, and mapping it to the Hilbert space to generate a corresponding Hilbert index value;

[0079] Specifically, the two-dimensional coordinates of the photovoltaic panel Mapping to one-dimensional index value through Hilbert curve, the mapping formula is: , where h is the one-dimensional index value corresponding to the Hilbert curve, and x and y are the two-dimensional coordinates of the photovoltaic panel. The space-filling property of the Hilbert curve ensures that similar photovoltaic panels have small differences in their one-dimensional index values.

[0080] S133, constructing a Hilbert R-tree structure based on the Hilbert index value of the photovoltaic panel;

[0081] Specifically, an R-tree index structure is constructed based on the Hilbert curve index values ​​of the photovoltaic panels. Each node stores the location data of the photovoltaic panels within the corresponding area and their Hilbert index interval. The R-tree structure is dynamically updated and can automatically adjust the node structure according to the actual location of the photovoltaic panels.

[0082] S134, by traversing the boundary points of each grid, using the Hilbert R-tree to perform a range query, and finding and locating all photovoltaic panels within the current grid cell;

[0083] Specifically, by traversing the boundaries of each grid cell , using the Hilbert R-tree to perform range query, find and locate all photovoltaic panels within the current grid cell boundary. The search logic is as follows: , then the position of photovoltaic panel i is recorded. All photovoltaic panels that meet the conditions will be stored as part of the photovoltaic panel position set.

[0084] S135 , recording the found photovoltaic panel positions as a photovoltaic panel position set.

[0085] Specifically, for each grid cell , record the photovoltaic panel location set in the form of: ,in, Represents a grid cell The set of photovoltaic panel locations within is the specific location coordinate of the photovoltaic panel. Ultimately, the set of photovoltaic panel locations of all grid cells will form a complete set of the entire photovoltaic panel area, providing input data for subsequent calculations.

[0086] In some embodiments, to further improve query accuracy and efficiency, a density adjustment mechanism is introduced in the steps. By calculating the density coefficient of the photovoltaic panels in the grid unit , the node division criteria of the Hilbert R-tree can be dynamically adjusted to improve the search efficiency in densely populated areas of photovoltaic panels. The calculation formula for the density coefficient is: ,in, For grid cells The number of photovoltaic panels in is the area of ​​the grid cell. The density-based adaptive adjustment mechanism further optimizes the search efficiency of the R-tree when processing areas with densely distributed photovoltaic panels.

[0087] S14, constructing a preliminary aggregation unit based on the photovoltaic panel location set, and expanding the aggregation area according to the neighboring grid relationship;

[0088] In this embodiment, a clustering algorithm is used to construct preliminary clustering units based on the set of PV panel locations. The clustering area is then gradually expanded based on the proximity between grid cells. The core of this step is to cluster PV panels within multiple grid cells based on the spatial proximity principle and the distribution characteristics of the PV panel locations within adjacent grid cells to form preliminary clustering units. The clustering range is then further expanded to construct a clustered structure covering a larger area.

[0089] In some embodiments, the construction of preliminary aggregated units is based on the relative density of photovoltaic panel locations within a grid and the proximity between grids. ,in Represents an aggregate unit consisting of several adjacent grid cells. Each aggregate unit At least one grid cell and its adjacent grids. In the initial state, if there is no grid in a certain aggregation unit, the first grid unit with a photovoltaic panel position set is used by default. As the initial polymerization unit .

[0090] Specifically, the aggregation algorithm is based on the following steps:

[0091] S141, constructing a preliminary aggregation unit, performing preliminary aggregation based on the set of photovoltaic panel positions in each grid according to spatial proximity relationships;

[0092] Specifically, for each grid cell , if the photovoltaic panel location set If it is not empty, it will start to build the preliminary aggregation unit. The construction of the preliminary aggregation unit is based on the aggregation center grid. and its neighboring grid relationship. The neighboring grid is represented as , that is, the adjacent grids in the up, down, left, right, and diagonal directions. The formula for each preliminary aggregation unit is: ,in is an aggregation unit, including the central grid and its surrounding adjacent grids, and the photovoltaic panel location sets in these grid cells are not empty.

[0093] S142, performing aggregation expansion based on the spatial proximity between grids and the distribution density of photovoltaic panel locations. If a photovoltaic panel location set also exists in an adjacent grid of a certain grid, it is incorporated into the current aggregation unit and the expansion continues to its surrounding adjacent grids until the photovoltaic panel location set of the adjacent grid is empty or does not meet the expansion conditions.

[0094] Specifically, if adjacent grids The photovoltaic panel location collection is not empty, the adjacent grid will be included in the current aggregation unit The aggregation unit is further expanded by checking the adjacent grids around the grid. This process is repeated recursively until no new adjacent grids are found or the set of photovoltaic panel positions of the adjacent grids is empty. As for the stopping condition, the expansion of the aggregation unit is based on the spatial proximity between grids and stops when the set of photovoltaic panel positions of the adjacent grids is empty. , the grid set in its final state will be expressed as: , all grids in the aggregation unit contain a set of photovoltaic panel locations, and the grids have a spatial proximity relationship with each other.

[0095] S143: Determine the number of grids and layout structure of each aggregation unit based on the aggregation expansion.

[0096] Specifically, through the above expansion rules, the initial aggregation unit set C will consist of multiple aggregation units, each aggregation unit It covers several adjacent grid cells and the locations of the PV panels within them. As the expansion progresses, the number and structure of the initial aggregation cells will be dynamically adjusted according to the distribution of PV panels, ensuring that the distribution of PV panels between grid cells can be accurately aggregated and represented.

[0097] In some embodiments, the construction of the aggregation unit also takes into account the density of the photovoltaic panel distribution and the distance between the grids, and defines the density coefficient of the aggregation unit. for: ,in Represents an aggregation unit The number of all photovoltaic panels in Indicates the total area covered by the aggregation unit. If the value exceeds the preset threshold, the aggregation unit will be further split to construct a smaller aggregation area to improve the accuracy and effectiveness of the aggregation results.

[0098] This step constructs preliminary clustering units based on the spatial proximity between grids and the distribution of photovoltaic panels. The clustering area can then be dynamically expanded based on the actual distribution of photovoltaic panels. It's important to note that this expansion is primarily a simple extension of the relationship between adjacent grids. This means that grids' proximity determines whether they can be added to the same clustering unit. This essentially preliminarily determines the spatial boundaries of each cluster, allowing the photovoltaic panels to be initially clustered together. This is a relatively rough initial clustering process. The goal is to establish a preliminary structure for the clustering unit.

[0099] S15, the eight-neighborhood search algorithm is used to further expand the aggregation unit and finally form an aggregation unit set.

[0100] In this embodiment, based on the preliminary aggregation unit generated in the above steps, the eight-neighborhood search algorithm is used to further expand the aggregation unit, and finally form a complete set of aggregation units. This step expands the grid layer by layer through the spatial neighborhood relationship to ensure that the photovoltaic panels in each aggregation unit have high spatial continuity and reasonable aggregation scale. It is worth mentioning that the expansion method of this step no longer relies solely on the proximity relationship of the grid, but is based on the distance relationship of the specific photovoltaic panel position, that is, based on the actual distribution of the photovoltaic panels. Specifically, it relies on the spatial distance condition between the photovoltaic panels, that is, taking the grid in each aggregation as the center, based on the aggregation distance D to determine whether the neighboring grid can be added to the aggregation unit, and further accurately optimizes the boundary and internal structure of the aggregation unit through the algorithm to ensure the rationality and accuracy of the aggregation unit.

[0101] In some embodiments, the eight-neighborhood search algorithm uses each aggregation unit Center grid within As the starting point, it gradually expands to the eight neighboring grids around it. The eight neighborhoods are defined as each grid The eight surrounding adjacent grids are: The core principle of the eight-neighborhood search is to determine whether the photovoltaic panels between the central grid and the neighboring grids should belong to the same aggregation unit based on the spatial distance relationship of the photovoltaic panel location set. The expansion process includes the following steps:

[0102] Neighborhood distance determination: for each aggregation unit , with any internal grid As the center, detect its eight neighborhood grids in turn The location of the photovoltaic panels. If the center grid and its neighboring grids The minimum distance between the photovoltaic panel locations within a set The following conditions are met: , where D is the maximum aggregation distance, then the neighborhood grid Add to current aggregation unit This condition ensures that the photovoltaic panels in each aggregation unit are closely connected in space, and the maximum distance does not exceed the preset maximum aggregation distance.

[0103] Recursive expansion process: Once a neighboring grid is added to the current aggregation unit, the grid is used as the new center grid, and the same distance determination and expansion operations are continued for the eight surrounding grids. This process is repeated recursively until no new grids can be added to the eight neighborhoods of all grids in the current aggregation unit, indicating that the expansion of the aggregation unit is complete.

[0104] Aggregation termination condition: when the eight neighboring grids of all grids in a certain aggregation unit cannot meet the minimum distance judgment condition When , the aggregation extension ends. At this point, the aggregation unit has included all the photovoltaic panel positions that are close to each other. The final aggregation unit set is In the , all grids in each aggregation unit meet the distance judgment conditions of neighborhood expansion.

[0105] Photovoltaic panel clustering result output: For each aggregation unit, output the set of all photovoltaic panel locations it contains ,in Represents an aggregation unit The photovoltaic panels in the final aggregation unit constitute the complete photovoltaic panel clustering result, which is the so-called aggregation unit set: , , i≤n, the photovoltaic panel collection in each aggregation unit not only reflects the spatial distribution of photovoltaic panels, but also provides important basic data for subsequent photovoltaic panel layout optimization and regional planning. Figure 5 As shown in the figure, according to the above steps, the sample data finally obtains two clusters.

[0106] In some implementations, the maximum aggregation distance can be adjusted based on specific scenarios or PV panel density to accommodate aggregation requirements of varying scale and accuracy. The efficiency and scalability of the eight-neighborhood search algorithm ensures computational efficiency and accuracy when processing large-scale PV panel location data.

[0107] S2, performing installed capacity optimization processing based on the installed capacity requirements of the power generation units of the aggregated unit set to obtain a pre-processed capacity unit set;

[0108] In this embodiment, based on the above generated aggregation unit set ,(i≤n), the number of photovoltaic panels in each aggregation unit is optimized for installed capacity to ensure that the number of photovoltaic panels in each aggregation unit meets the requirements of an integer power generation unit, and finally a pre-processing capacity unit set is generated. This optimization process reasonably adjusts the spatial distribution and number of photovoltaic panels in the aggregation unit, so that each unit set has an executable power generation scale.

[0109] In some embodiments, for each polymeric unit First, count the number of photovoltaic panels contained in it to determine whether it meets the number requirement of integer installed units. Suppose each power generation unit contains photovoltaic panels, if aggregated units Number of photovoltaic panels in The following conditions are met: ,in The deviation range is allowed. If it is within the range, the aggregation unit does not need further processing and is directly used as part of the pre-processing capacity unit. However, if the aggregation unit Number of photovoltaic panels in Exceeding the permitted range, i.e. , then the photovoltaic panels in the unit need to be reduced to meet the power generation unit capacity requirements. The reduction process is carried out through the following steps:

[0110] S21, for each aggregation unit, calculating the geometric centroid position of the aggregation unit;

[0111] In this step, the calculation of the centroid position is based on the position coordinates of all photovoltaic panels in the aggregation unit. shared within photovoltaic panels, the position coordinates of each photovoltaic panel are ,in , centroid The calculation formula is as follows: ,in and The centroid represents the spatial center of the photovoltaic panel in the aggregation unit.

[0112] S22, based on the geometric centroid position, calculating the distance from each photovoltaic panel to the centroid and sorting them to determine the photovoltaic panel farthest from the centroid;

[0113] The spatial distribution of each photovoltaic panel in the aggregation unit is further analyzed by calculating the Euclidean distance between it and the centroid. distance The calculation formula is: ,After calculating the distance from each photovoltaic panel to the centroid, the photovoltaic panels are sorted in descending order of distance, so as to give priority to reducing photovoltaic panels that are farther away, ensuring a more compact spatial distribution of the remaining photovoltaic panels.

[0114] S23, based on the installed capacity requirements of the power generation unit, discard the photovoltaic panels farthest from the centroid in sequence until the number of photovoltaic panels meets the capacity requirements;

[0115] In this step, the number of photovoltaic panels in the aggregate unit will be adjusted according to the installed capacity requirements. The installed capacity of each power generation unit is photovoltaic panels, if the number of photovoltaic panels in the aggregation unit Exceeding this requirement, i.e. , then the photovoltaic panels farthest from the centroid need to be gradually discarded. The order of discarding is based on the distance calculated previously, until the number of photovoltaic panels in the aggregation unit meets the following conditions: ,in After this reduction process, the number of photovoltaic panels in each aggregate unit is guaranteed to meet the installed capacity standard of the power generation unit while maintaining a reasonable spatial layout.

[0116] After the above steps, the pre-processing capacity unit set is finally formed. , (m≤n), where each unit meets the installed capacity requirements of integer power generation units. This result provides basic data for subsequent photovoltaic field planning and installation optimization.

[0117] In some implementations, the allowable deviation range can be adjusted based on the actual project situation. , to adapt to the installation needs of photovoltaic panels of different sizes while maintaining efficient installed capacity configuration.

[0118] S3, partitioning the power generation units based on the pre-processing capacity unit set, wherein the partitioning is performed by dividing the pre-processing capacity units according to the spatial position and quantity conditions, thereby obtaining a power generation unit set.

[0119] S31, based on the boundary data of the pre-processed capacity unit set, calculate the boundaries of the photovoltaic panel cluster and generate a boundary parameter set, combine the aspect ratio information and the integer multiple requirement of the installed capacity of the power generation unit, sort and split the pre-processed capacity unit set to form a power generation unit segmentation candidate set.

[0120] In this example, based on the boundary data of the pre-processed capacity unit set, the bounding box of each cluster unit is calculated to determine its physical distribution aspect ratio and spatial location, providing a reference for the subsequent division of power generation units. The core of this process lies in the rational planning of the spatial layout of photovoltaic panels, ensuring that the power generation unit capacity matches the number and location of pre-processed capacity units.

[0121] In some embodiments, for each cluster unit Calculate its four boundary parameters, that is, determine its east-west boundary w,e and north-south boundary s,n. The boundary calculation formula is: , , , , for ,in Represents clustering units The location coordinates of the photovoltaic panels inside. Next, by calculating the length L and width W of the four-dimensional range, the spatial distribution of the cluster units is determined based on the aspect ratio. The formula is as follows: L = ew, W = ns. When L > D, it indicates that the cluster unit is more extended in the vertical direction and is divided along the vertical direction. Conversely, when L ≤ D, it indicates that the cluster unit is more widely distributed in the horizontal direction and is divided horizontally.

[0122] Combined with the requirement of integer multiples of the installed capacity of the power generation units in the pre-processing capacity unit concentration, the power generation units are divided into integer multiples of the number of photovoltaic panels in each power generation unit.

[0123] In some embodiments, if the clustering unit If the length L of the cluster is greater than the threshold (aggregation distance) D, the vertical sorting and segmentation are prioritized. At this time, all photovoltaic panels in the cluster are first sorted in ascending order according to their y coordinates, and the photovoltaic panels are divided into two parts from top to bottom based on the installed capacity of the power generation unit. For example, please refer to the attached Figure 6 As shown, the data of the second cluster is used as a subsequent example, the cluster unit Contains 274 photovoltaic panels, the power generation unit standard capacity is 55 photovoltaic panels, it can be divided into the upper part (including three power generation units, a total of 165 photovoltaic panels), and the lower part (Contains two power generation units, a total of 109 photovoltaic panels).

[0124] If the clustering unit If the length L is less than or equal to D, it is split horizontally. At this time, all photovoltaic panels are sorted according to the x coordinate and divided into left and right parts in the same way. Finally, each cluster unit can form a segmentation candidate set, which is recorded as , where each subset meets the requirement of an integer multiple of the power generation unit capacity.

[0125] The candidate set of power generation unit segmentation generated through this process provides basic data for subsequent power plant planning and ensures that each segmented unit can meet the spatial layout requirements and ensure the rationality of installed capacity.

[0126] S32, based on the candidate set of power generation unit partitioning, each unit after partitioning is verified using a connectivity capacity verification algorithm. If the conditions are not met, a partition adjustment unit set is generated;

[0127] In this embodiment, connectivity and capacity verification is performed on each split unit based on the candidate set of power generation unit splits. This ensures that the split units not only meet the power generation unit installed capacity requirements but also maintain spatially aggregated connectivity. Specifically, a connectivity capacity verification algorithm is used to reaggregate the photovoltaic panels in each split unit to determine whether the connectivity and maximum aggregation distance of the photovoltaic panels within it meet the requirements.

[0128] In some embodiments, for each unit in the candidate set of power generation unit segmentation and , according to the aggregation operation in the previous step S1, the photovoltaic panels inside it are aggregated again. The aggregation operation is performed according to the set maximum aggregation distance D, ensuring that the relative distance between all photovoltaic panels in physical space does not exceed the D value.

[0129] According to the aggregation results, if and If there is only one PV panel cluster in each of the clusters, it means that these PV panels still maintain connectivity after segmentation, and their distances are all within the maximum aggregation distance D, meeting the connectivity requirement. In this case, no further adjustment is required.

[0130] If the number of clusters in the split unit is more than one, please refer to the attached Figure 7 As shown, or If multiple photovoltaic panel clusters appear, it means that the split units have lost connectivity. All PV panels within the grid met the connectivity requirements before the split, but the split operation broke the connectivity between these PV panels. According to the connectivity capacity verification algorithm, any split unit that loses connectivity will not meet the integer multiple requirement of the installed capacity of the power generation unit, and therefore its capacity condition will not be further checked.

[0131] In the event that connectivity verification fails, a zoning adjustment unit set will be generated to record the units that need to be rezoned and provide a data basis for subsequent zoning adjustment steps.

[0132] S33, based on the partition adjustment unit set, borrowing photovoltaic panels from the opposing partition units according to the distance relationship between adjacent units to form a corrected power generation unit partition set;

[0133] In this embodiment, based on the partition adjustment unit set, photovoltaic panels from adjacent opposing partitions are used to ensure that the installed capacity of each power generation unit meets the integer multiple requirement. The core of this step is to optimize the number of photovoltaic panels after the division to ensure that the partition units maintain connectivity while meeting the installed capacity requirements. This operation first identifies the sub-cluster with the fewest photovoltaic panels in the partition adjustment unit set. Based on the distance relationship between the sub-cluster and the adjacent opposing partitions, the sub-cluster is supplemented from the nearest photovoltaic panel cluster until the requirement is met.

[0134] In some embodiments, the sub-cluster set in the partition adjustment unit set is Sort by the number of photovoltaic panels. The subcluster with the least number of photovoltaic panels As the first secondment target. When replenishing photovoltaic panels, select the sub-cluster that is closest to it , and give priority to secondment and The nearest photovoltaic panel is selected until the number of photovoltaic panels meets the requirement of an integer multiple of the installed capacity of the power generation unit. The order of photovoltaic panel borrowing in each cluster is based on the distance priority principle by calculating the distance matrix between photovoltaic panels. , select the nearest opposing partition for secondment, the number of secondment operations Determine as follows: ,in Indicates that from the opposite partition The number of photovoltaic panels on loan. After the loan is completed, the number of photovoltaic panels in the partition unit needs to be rechecked to ensure that it meets the requirements of the installed capacity of the power generation unit.

[0135] In some embodiments, see the attached Figure 8As shown in the figure, the number of PV panels in Upper Cluster 1 in the zone adjustment unit is 17, which does not meet the requirement of an integer multiple of the installed capacity of the power generation unit. To meet the installed capacity standard, PV panels are borrowed from the opposite and closest Lower Cluster 1. First, the PV panels in Lower Cluster 1 are sorted, and the PV panels closest to Upper Cluster 1 are selected to supplement the number of PV panels until the number of PV panels in Upper Cluster 1 reaches the required number.

[0136] After the first replenishment of photovoltaic panels, the upper cluster 1 meets the conditions, but the number of photovoltaic panels in the lower cluster 2 is insufficient. Please continue to refer to the attached Figure 9 As shown in the figure, in this case, lower cluster 2 is selected as the new loan target, and PV panels are selected from the corresponding upper clusters 2 and 3 for loan. Since the number of PV panels in upper cluster 3 already meets the requirement, the loan operation is prioritized from upper cluster 2 until the number of PV panels in lower cluster 2 meets the requirement.

[0137] After further inspection, please continue to refer to Figure 10 As shown in Figure 1, the number of PV panels in lower cluster 1 still does not meet the installed capacity requirements of the power generation unit. At this point, PV panels are borrowed from upper cluster 2, the closest opposing cluster to lower cluster 1, to complete the final partition adjustment. This process continues until the number of PV panels in all partitions meets the installed capacity requirements of the power generation unit.

[0138] Through this process, a corrected power generation unit partition set is generated, ensuring that each partition unit not only maintains good connectivity but also meets the standards of the power generation unit installed capacity, thereby providing accurate and reliable partition data for subsequent power generation unit configuration.

[0139] S34 , based on the corrected power generation unit partition set, gradually check whether the installed capacity of all partition units meets the requirements, and finally output the power generation unit set.

[0140] In this embodiment, based on the corrected power generation unit partition set, the installed capacity of each partition is checked one by one to ensure that it meets the requirement of an integer multiple of the set power generation unit installed capacity. This step verifies whether each partition has met the power generation unit classification criteria through a recursive check, and outputs the power generation unit set that meets the requirements based on the check results.

[0141] In some embodiments, an initial set of base partitions is defined , and each sub-cluster in the corrected power generation unit partition set Eliminate one by one For each sub-cluster set to be examined , by calculating the number of photovoltaic panels in each sub-cluster , to determine whether it meets the requirement of integer multiples of the installed capacity of the power generation unit. If the number of photovoltaic panels meets the requirements, it will be converted into a power generation unit partition object And output. The formula is as follows: , where M is an integer multiple of the installed capacity of the power generation unit. If this condition is met, then Converted into power generation unit Otherwise, Return to the base partition set , and continue to split and adjust.

[0142] In some embodiments, see the attached Figure 11 As shown, when the sub-cluster When the number of photovoltaic panels does not meet the requirements, add them back , return to the initial stage of S3, and perform segmentation and correction again. This cycle continues until all partition units meet the standard of power generation unit installed capacity. The final output power generation unit partition set , represents the set of all eligible generating units. This process ensures that each partition unit undergoes rigorous capacity verification and adjustment to form a complete set of generating unit partitions, and ensures that all partition units have good connectivity and capacity matching.

[0143] Through this step, a partition set that meets the installed capacity of the photovoltaic power station power generation unit is finally obtained, and the entire segmentation and correction process is completed.

[0144] Based on the description of the embodiment of the mountain photovoltaic-based power generation unit partitioning method, the present application also discloses a mountain photovoltaic-based power generation unit partitioning system. The mountain photovoltaic-based power generation unit partitioning system can be a computer program (including program code) that runs the mountain photovoltaic-based power generation unit partitioning method mentioned above. Figure 13 As shown, the power generation unit partitioning system based on mountain photovoltaics can operate the following units:

[0145] The aggregation unit generation unit 110 is used to obtain an aggregation unit set of photovoltaic panels. The aggregation unit set is obtained by gridding all photovoltaic panel strings within a set aggregation distance and aggregating them into areas of different sizes according to the distribution of photovoltaic panels in the grid;

[0146] an installed capacity optimization unit 120, configured to perform installed capacity optimization processing on the aggregated unit set according to the installed capacity requirements of the power generation unit to obtain a pre-processed capacity unit set;

[0147] The power generation unit partitioning unit 130 is configured to partition the pre-processing capacity unit set into power generation units, wherein the partitioning is performed by dividing the pre-processing capacity units according to their spatial positions and quantity conditions, thereby obtaining a power generation unit set.

[0148] The foregoing description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein and should not be construed as excluding other embodiments. Rather, the present invention can be used in various other combinations, modifications, and environments and can be modified within the scope of the concept described herein through the above teachings or techniques or knowledge in the relevant field. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention are intended to be protected by the appended claims.

Claims

1. A method for zoning power generation units based on mountain photovoltaics, characterized in that: The method comprises the following steps: S1, obtaining an aggregation unit set of photovoltaic panels, wherein the aggregation unit set is obtained by performing grid aggregation on all photovoltaic panel strings within a set aggregation distance, and aggregating them into areas of different sizes according to the distribution of photovoltaic panels in the grid; S2, performing installed capacity optimization processing based on the installed capacity requirements of the power generation units of the aggregated unit set to obtain a pre-processed capacity unit set; S3, partitioning the power generation units based on the pre-processing capacity unit set, wherein the partitioning is performed by dividing the pre-processing capacity units according to spatial position and quantity conditions, thereby obtaining a power generation unit set; The S1 step includes the following sub-steps: S11, based on the terrain characteristics and site boundaries of the photovoltaic panel area, the boundaries of the photovoltaic panel are calculated using a dynamic boundary recognition algorithm to generate boundary data for grid division; S12, based on the set maximum aggregation distance, the boundary data is divided and processed by the adaptive grid division model to generate a grid; S13, traversing each of the grids using a Hilbert R-tree, locating and recording the photovoltaic panel positions within each grid based on the four points of the grid, and generating a photovoltaic panel position set; S14, constructing a preliminary aggregation unit based on the photovoltaic panel location set, and expanding the aggregation area according to the neighboring grid relationship; S15, the eight-neighborhood search algorithm is used to further expand the aggregation unit and finally form an aggregation unit set.

2. A method for zoning power generation units based on mountain photovoltaics according to claim 1, characterized in that: The specific calculation formula of the adaptive grid division model in S12 is as follows: ,in is the adaptive side length of the grid unit, D is the maximum aggregation distance, and K is the adjustment coefficient with a value range of , is the basic expansion scale factor, is the sensitivity coefficient, The average slope of the grid cells, is the terrain relief of the grid cell.

3. The method for zoning power generation units based on mountain photovoltaics according to claim 1, characterized in that: The S13 includes the following sub-steps: S131, based on the grid, defining the boundary points of each grid, wherein the boundary points constitute the spatial query conditions of the Hilbert R-tree search; S132, obtaining the two-dimensional coordinate data of the photovoltaic panel, and mapping it to the Hilbert space to generate a corresponding Hilbert index value; S133, constructing a Hilbert R-tree structure based on the Hilbert index value of the photovoltaic panel; S134, by traversing the boundary points of each grid, using the Hilbert R-tree to perform a range query, and finding and locating all photovoltaic panels within the current grid cell; S135 , recording the found photovoltaic panel positions as a photovoltaic panel position set.

4. The method for zoning power generation units based on mountain photovoltaics according to claim 1, characterized in that: The step S14 includes the following sub-steps: S141, constructing a preliminary aggregation unit, performing preliminary aggregation based on the set of photovoltaic panel positions in each grid according to spatial proximity relationships; S142, performing aggregation expansion based on the spatial proximity between grids and the distribution density of photovoltaic panel locations. If a photovoltaic panel location set also exists in an adjacent grid of a certain grid, it is incorporated into the current aggregation unit and the expansion continues to its surrounding adjacent grids until the photovoltaic panel location set of the adjacent grid is empty or does not meet the expansion conditions. S143: Determine the number of grids and layout structure of each aggregation unit based on the aggregation expansion.

5. A method for zoning power generation units based on mountain photovoltaics according to any one of claims 1 to 4, characterized in that: The S2 step includes the following sub-steps: S21, for each aggregation unit, calculating the geometric centroid position of the aggregation unit; S22, based on the geometric centroid position, calculating the distance from each photovoltaic panel to the centroid and sorting them to determine the photovoltaic panel farthest from the centroid; S23, based on the installed capacity requirements of the power generation unit, discard the photovoltaic panels farthest from the centroid in sequence until the number of photovoltaic panels meets the capacity requirements.

6. A method for zoning power generation units based on mountain photovoltaics according to claim 5, characterized in that: The S21 calculates the geometric centroid position of the aggregation unit based on the position coordinates of all photovoltaic panels in the aggregation unit. shared within photovoltaic panels, the position coordinates of each photovoltaic panel are ,in , centroid The calculation formula is as follows: .

7. The method for zoning power generation units based on mountain photovoltaics according to claim 5, characterized in that: The S3 step includes the following sub-steps: S31, calculating the boundaries of the photovoltaic panel cluster based on the pre-processed capacity unit set and generating a boundary parameter set, combining the aspect ratio information and the requirement of an integer multiple of the installed capacity of the power generation unit, and segmenting the pre-processed capacity unit set to form a candidate set of power generation unit segmentation; S32, based on the candidate set of power generation unit partitioning, each unit after partitioning is verified using a connectivity capacity verification algorithm. If the conditions are not met, a partition adjustment unit set is generated; S33, based on the partition adjustment unit set, borrowing photovoltaic panels from the opposing partition units according to the distance relationship between adjacent units to form a corrected power generation unit partition set; S34 , based on the corrected power generation unit partition set, gradually check whether the installed capacity of all partition units meets the requirements, and finally output the power generation unit set.

8. A method for zoning power generation units based on mountain photovoltaics according to claim 7, characterized in that: The east-west boundaries of the four-boundary range of the photovoltaic panel cluster are w,e and the north-south boundaries are s,n. The boundary parameter set is the length L and width W of the four-boundary range, where L=ew, W=ns. The pre-processing capacity unit set is divided according to the following conditions: when L>D, it is divided longitudinally, and when L≤D, it is divided horizontally, where D is the maximum aggregation distance.

9. A power generation unit zoning system based on mountain photovoltaics, characterized in that: The system comprises: An aggregation unit generation unit is used to obtain an aggregation unit set of photovoltaic panels. The aggregation unit set is obtained by gridding all photovoltaic panel strings within a set aggregation distance and aggregating them into areas of different sizes according to the distribution of photovoltaic panels in the grid; an installed capacity optimization unit, configured to perform installed capacity optimization processing on the aggregated unit set according to the installed capacity requirements of the power generation unit, to obtain a pre-processed capacity unit set; A power generation unit partitioning unit, configured to partition the pre-processing capacity unit set into power generation units, wherein the partitioning is performed by dividing the pre-processing capacity units according to their spatial positions and quantity conditions, thereby obtaining a power generation unit set; The aggregation unit generation unit performs the following sub-steps: S11, based on the terrain characteristics and site boundaries of the photovoltaic panel area, the boundaries of the photovoltaic panel are calculated using a dynamic boundary recognition algorithm to generate boundary data for grid division; S12, based on the set maximum aggregation distance, the boundary data is divided and processed by the adaptive grid division model to generate a grid; S13, traversing each of the grids using a Hilbert R-tree, locating and recording the photovoltaic panel positions within each grid based on the four points of the grid, and generating a photovoltaic panel position set; S14, constructing a preliminary aggregation unit based on the photovoltaic panel location set, and expanding the aggregation area according to the neighboring grid relationship; S15, the eight-neighborhood search algorithm is used to further expand the aggregation unit and finally form an aggregation unit set.

Images