Complex mountain photovoltaic subunit division method and system based on adaptive optimization clustering

Abstract

The invention relates to the technical field of photovoltaic power station design, in particular to a complex mountain photovoltaic subunit division method and system based on adaptive optimization clustering. Through global dynamic adjustment, accurate matching of the number of each subunit string and a target value is achieved, the division balance is greatly improved, meanwhile, full-process automatic operation is achieved, the data processing efficiency and division consistency of a large-scale mountain photovoltaic project are remarkably improved, and through combination of density clustering and an optimization adjustment strategy, the division efficiency of the large-scale mountain photovoltaic project is improved. The method can flexibly adapt to different mountain terrains and string distribution modes, the division result meets the actual requirements of electrical design, cable laying and operation and maintenance management, line loss can be reduced, the system efficiency can be improved, the engineering cost can be reduced, and in the face of the influence of complex geographical environment factors in a complex mountain scene, through parameter self-adaptive setting, the system efficiency can be improved. Algorithm parameters can be adjusted according to specific requirements of projects, and the method is suitable for various complex mountain scenes.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power plant design technology, and in particular to a method and system for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering. Background Technology

[0002] With the advancement of "dual carbon" goals and the rapid development of the photovoltaic industry, complex mountain photovoltaic power stations have become an important development direction due to their efficient use of land resources. As flat terrain resources become increasingly scarce, mountain photovoltaic power stations, with their advantages of high land use efficiency and strong adaptability, are gradually becoming a development focus. However, the string distribution of mountain photovoltaic power stations is affected by topographical undulations, slope variations, and geological conditions, exhibiting significant spatial unevenness and morphological diversity: strings are densely packed in some areas, while the overall distribution is scattered and irregular in boundaries, which places string constraints on sub-unit division. As the basic building block of a photovoltaic power station, the quality of sub-unit division directly affects string wiring, electrical efficiency, operation and maintenance costs, and system reliability.

[0003] Currently, photovoltaic sub-unit partitioning mainly relies on manual experience or methods based on simple rules such as horizontal and vertical orientation. Manual partitioning methods are inefficient, inconsistent, and unsuitable for large-scale mountainous projects. Existing software tools are mostly applicable to flat terrain and lack adaptability to complex terrain. Clustering algorithms such as DBSCAN and K-means are widely used in data partitioning, but their direct application in real-world mountainous scenarios has significant shortcomings: for example, they cannot simultaneously guarantee the balance of string counts and spatial compactness within sub-units, are sensitive to initial parameters, and lack dynamic optimization mechanisms for boundary strings.

[0004] Therefore, there is an urgent need in this field for a method that can automate, efficiently and accurately divide mountain photovoltaic sub-units to improve the scientific, economic and reliability of photovoltaic power plant design. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering, characterized by the following steps: S1. Obtain the spatial coordinate data of the photovoltaic strings of the target photovoltaic power station, perform preliminary clustering through density clustering algorithm to obtain the regional outline of the photovoltaic strings of the target photovoltaic power station, and divide multiple initial regions based on density accessibility and connectivity; S2. Based on the obtained multiple initial regions and the preset target number of sub-unit strings, the center point optimization iterative algorithm is used to pre-divide each initial region into sub-units to generate multiple pre-divided sub-units. S3. Based on the obtained multiple pre-divided sub-units, calculate the deviation between the actual number of strings in each pre-divided sub-unit and the target number of strings in the sub-unit, construct an adjacency relation matrix reflecting the spatial topological relationship of each pre-divided sub-unit, and based on the adjacency relation matrix, perform global dynamic allocation and iterative adjustment of strings in the pre-divided sub-units through a comprehensive weight model until the number of strings in all pre-divided sub-units is the same as the target number of strings in the sub-unit, and obtain the photovoltaic sub-unit division result of the target photovoltaic power station.

[0006] Based on the above method, the present invention also provides a complex mountain photovoltaic sub-unit partitioning system based on adaptive optimization clustering. The system is implemented based on any one of the above-mentioned complex mountain photovoltaic sub-unit partitioning methods based on adaptive optimization clustering, including a data acquisition and region partitioning module, used to acquire the spatial coordinate data of the photovoltaic strings of the target photovoltaic power station, perform preliminary clustering through density clustering algorithm to obtain the regional outline of the photovoltaic strings of the target photovoltaic power station, and divide multiple initial regions based on density accessibility and connectivity. The sub-unit pre-division module, based on the obtained multiple initial regions and the preset target number of sub-unit strings, pre-divides each initial region into sub-units using a center point optimization iterative algorithm, generating multiple pre-divided sub-units. The adjacency matrix construction module is used to calculate the deviation between the actual number of strings in each pre-divided sub-unit and the target number of strings in the sub-unit based on the obtained multiple pre-divided sub-units, and construct an adjacency matrix that reflects the spatial topological relationship of each pre-divided sub-unit. The photovoltaic string allocation optimization module is used to perform global dynamic allocation and iterative adjustment of strings in the pre-divided sub-units based on the adjacency relationship matrix and through a comprehensive weight model, until the number of strings in all pre-divided sub-units is the same as the target number of strings in the sub-unit, thus obtaining the photovoltaic sub-unit division result of the target photovoltaic power station.

[0007] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the aforementioned method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering.

[0008] A storage medium storing a computer program thereon, which, when executed by a processor, implements the method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering as described above.

[0009] This invention achieves precise matching between the number of strings in each sub-unit and the target value through global dynamic adjustment, greatly improving the partitioning balance. At the same time, the fully automated operation significantly improves the data processing efficiency and partitioning consistency of large-scale mountain photovoltaic projects. Combined with density clustering and optimization adjustment strategies, it can flexibly adapt to different mountain terrains and string distribution patterns. The partitioning results meet the actual needs of electrical design, cable laying, and operation and maintenance management, which helps to reduce line losses, improve system efficiency, and reduce engineering costs. In the face of complex mountain scenarios and the influence of complex geographical environmental factors, the algorithm parameters can be adjusted according to the specific needs of the project through adaptive parameter settings, making it suitable for a variety of complex mountain scenarios. Attached Figure Description

[0010] Figure 1 This is a flowchart of a method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering, according to an embodiment of the present invention.

[0011] Figure 2 This is a schematic diagram of a complex mountain photovoltaic sub-unit partitioning system based on adaptive optimization clustering, according to an embodiment of the present invention.

[0012] Figure 3 This is a schematic diagram of the clustering results in an embodiment of the present invention.

[0013] Figure 4 This is a schematic diagram of sub-unit division according to an embodiment of the present invention.

[0014] Figure 5 This is a schematic diagram of the adjacency relation matrix construction result in an embodiment of the present invention.

[0015] Figure 6 This is a schematic diagram of the complex mountain photovoltaic sub-unit division result according to an embodiment of the present invention.

[0016] Figure 7 This is a schematic diagram of the internal structure of a computer device according to an embodiment of the present invention.

[0017] Figure 8 This is a schematic diagram of the storage medium according to an embodiment of the present invention.

[0018] In the diagram, 200 is the terminal device, 210 is the memory, 211 is the RAM, 212 is the cache memory, 213 is the ROM, 214 is the program / utility, 215 is the program module, 220 is the processor, 230 is the bus, 240 is the external device, 250 is the I / O interface, 260 is the network adapter, and 300 is the program product. Detailed Implementation

[0019] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention; that is, the described embodiments are only a part of the embodiments of the invention, and not all of them. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0022] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or machine that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or machine. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or machine that includes said element.

[0023] Example 1: This embodiment takes a complex mountainous photovoltaic project in a certain region as an example. The project contains approximately several thousand photovoltaic strings, distributed across areas with significant terrain undulations and high spatial heterogeneity. The goal is to divide all strings into multiple sub-units, with each sub-unit containing 230 strings.

[0024] like Figure 1 As shown, this embodiment of the invention provides a method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering, including the following steps: S1. Obtain the spatial coordinate data of the photovoltaic strings of the target photovoltaic power station, perform preliminary clustering through density clustering algorithm to obtain the regional outline of the photovoltaic strings of the target photovoltaic power station, and divide multiple initial regions based on density accessibility and connectivity; S2. Based on the obtained multiple initial regions and the preset target number of sub-unit strings, the center point optimization iterative algorithm is used to pre-divide each initial region into sub-units to generate multiple pre-divided sub-units. S3. Based on the obtained multiple pre-divided sub-units, calculate the deviation between the actual number of strings in each pre-divided sub-unit and the target number of strings in the sub-unit, construct an adjacency relation matrix reflecting the spatial topological relationship of each pre-divided sub-unit, and based on the adjacency relation matrix, perform global dynamic allocation and iterative adjustment of strings in the pre-divided sub-units through a comprehensive weight model until the number of strings in all pre-divided sub-units is the same as the target number of strings in the sub-unit, and obtain the photovoltaic sub-unit division result of the target photovoltaic power station.

[0025] Furthermore, step S1 includes the following sub-steps: S101. Obtain the spatial coordinate data of all photovoltaic strings in the target photovoltaic power station, perform standardized preprocessing, and extract the spatial density features of the photovoltaic strings; S102. Based on the spatial density characteristics of photovoltaic strings and the terrain data of the target photovoltaic power station, the density clustering algorithm is adaptively initialized with parameters including the neighborhood radius parameter and the minimum number of points contained in the core point threshold parameter. S103. Based on the density clustering algorithm after adaptive parameter initialization, traverse all photovoltaic string spatial coordinate data, identify core points, boundary points and noise points, and connect core points and boundary points respectively based on density accessibility and density reachability relationship to obtain initial region clusters. The noise points include isolated points that do not belong to any initial region cluster. S104. Remove noise points from the initial region clusters to obtain multiple initial regions for the photovoltaic strings of the target photovoltaic power station.

[0026] Specifically, the coordinate data of photovoltaic strings in complex mountain photovoltaic power stations are preprocessed, and preliminary clustering is performed based on the spatial distribution density characteristics of photovoltaic strings to identify and divide multiple initial regions. During the preliminary clustering process, clustering parameters are adaptively set according to the mountainous terrain distribution and string distribution density to identify and remove noise points, forming an initial region division that is adaptable to the terrain distribution.

[0027] Specifically, based on the spatial coordinate data of photovoltaic strings, a density clustering algorithm is used to identify the initial regional outline in the mountainous environment. Through adaptive parameter settings, core points, boundary points, and noise points are effectively identified. The region is divided according to density accessibility and connectivity, and noise points are removed to improve the efficiency and accuracy of subsequent processing.

[0028] Specifically, the implementation principle of each sub-step in the above embodiments is as follows: First, extract the coordinate data of all photovoltaic strings. Then, based on the project design parameters, set the clustering parameter to 32 and the minimum cluster point to 23. Execute the density clustering algorithm based on these parameters, successfully identifying 10 initial regions and removing noise points. The clustering results are as follows: Figure 3 As shown, the visible area has a clear outline and reflects the terrain features well.

[0029] Furthermore, step S2 includes the following sub-steps: S201. Based on the multiple initial regions obtained in step S1, count the total number of photovoltaic strings in each initial region, and calculate the number of sub-unit divisions in each initial region based on the preset target number of sub-unit strings. S202. Based on the number of sub-units, randomly select the same number of points in each initial region as initial cluster centers, calculate the Euclidean distance from each string to each cluster center, and assign the string to the sub-unit to which the nearest center point belongs. S203. Using the central store optimization iterative algorithm, the geometric center coordinates of each sub-unit are updated sequentially. The allocation and update steps are repeated until the cluster center positions converge, resulting in multiple pre-divided sub-units.

[0030] Specifically, based on the initial region and the preset target value of the photovoltaic string in the sub-unit, each initial region is initially divided into sub-units to generate sub-units with a string number close to the target value. During the initial division of the sub-units, the target string number is set based on constraints such as transformer capacity, cable load and electrical loss in actual engineering, and the number of each string is initially close to the target value through the center point optimization and redistribution mechanism.

[0031] Specifically, within each initial region, the number of sub-units to be divided is calculated based on the project's preset target number of photovoltaic strings (an example value is 230 photovoltaic strings per sub-unit, which can be adjusted according to project requirements). A center point optimization iterative method is then used for the initial sub-unit division. This process continuously optimizes the center point position and string allocation to bring the number of strings within each sub-unit close to the target value, laying the foundation for subsequent adjustments.

[0032] Specifically, the implementation principle and flow of each sub-step in the above embodiments are as follows: First, for each initial region, the number of sub-units k is calculated based on the total number of strings it contains (e.g., if a region contains 2300 strings, then k=10). Then, a center point optimization iterative method is used for initial partitioning, and the sub-unit pre-partitioning is as follows: Figure 4 As shown in Table 1, the number of sub-units ranges from 132 to 351. The pre-division results of the sub-units are shown in Table 1. It can be seen that some sub-units are close to the target value.

[0033] Table 1 Sub-unit pre-partitioning results

[0034] Furthermore, step S3, which involves constructing an adjacency relation matrix reflecting the spatial topological relationships of each pre-divided sub-unit, includes: constructing the convex hull boundaries of all strings surrounding each sub-unit using computational geometry algorithms, and calculating the minimum Euclidean distance between the convex hull boundaries of any two sub-units; based on a dynamic distance threshold, if the minimum distance between two sub-units is less than the dynamic distance threshold, then they are determined to be adjacent to each other, and the corresponding element in the adjacency relation matrix is ​​marked as connected; otherwise, it is marked as disconnected.

[0035] Furthermore, the dynamic distance threshold is determined by the actual string distance of each photovoltaic string within the target photovoltaic power station under operating conditions.

[0036] Furthermore, step S3, which describes the global dynamic allocation and iterative adjustment of strings within the pre-divided sub-units using a comprehensive weight model, includes: counting the actual number of strings in each pre-divided sub-unit, calculating the deviation between the actual number of strings and the target number of strings in the sub-unit, marking sub-units with an actual number of strings greater than the target number of strings as multi-string sub-units, marking sub-units with an actual number of strings less than the target number of strings as few-string sub-units, and the rest as qualified sub-units; based on the adjacency matrix, calculating the adjustment path for transferring strings from multi-string sub-units to few-string sub-units using a shortest path algorithm, and identifying the set of strings to be divided in the source sub-units that are close to the boundary of the target sub-unit for adjacent source and target sub-units on the adjustment path, calculating the allocation priority score of the strings to be divided using the comprehensive weight model, moving the strings from the source sub-units to the target sub-units in order of score, and updating the adjacency matrix and the string number deviation in real time until the string number deviation of all sub-units is zero, thus obtaining the photovoltaic sub-unit division result of the target photovoltaic power station.

[0037] Specifically, an adjacency matrix for sub-units is constructed. Based on a comprehensive weighting model, the boundary strings of the sub-units are dynamically adjusted globally. Through multiple iterations, the number of strings in all sub-units is made to precisely match the target value. The adjacency matrix is ​​calculated based on the minimum distance between units according to the convex hull boundary of the sub-units. A dynamic distance threshold is set, and a matrix structure identifying the adjacency relationships of sub-units is generated accordingly.

[0038] Specifically, during the global dynamic adjustment process, a comprehensive weight model is used to assign priority to the boundary strings. This model integrates the geometric distance from the string to the target unit and its impact on the change of the clustering quality profile coefficient. Through path planning and iterative adjustment, it achieves accurate matching of the number of strings and optimization of the overall clustering structure.

[0039] Furthermore, the comprehensive weight model is expressed as: ; in, Indicates the string to be divided The allocation priority score, Indicates the string to be divided The Euclidean distance to the boundary string of the target group. Represents a string The change in the group profile coefficient after being assigned to the target group. This represents the distance weight value, used to control the geometric compactness of the distance from the target group to the boundary of the target group. The magnitude of the weight value reflects the importance of considering the distance from the target group during the partitioning process. The contour coefficient weight value is used to balance the impact of string allocation on the clustering effect of strings within a group. The distance weight value and the pre-contour coefficient weight value balance the impact of the strings to be divided on the geometric compactness and clustering quality of the target sub-units during the partitioning process, thereby improving the partitioning accuracy.

[0040] Specifically, the implementation principle and flow of each sub-step in the above embodiments are as follows: First, for each pre-divided sub-unit in the initial region, calculate the deviation between the number of pre-divided sub-unit strings and the target number of strings to determine the number of sub-units with more strings (number of strings > target value) and fewer strings (number of strings < target value). For example, sub-unit 1 contains 317 strings (87 strings with more strings), and sub-unit 2 contains 197 strings (33 strings with fewer strings). Then, construct an adjacency matrix and calculate the minimum distance between the convex hull boundaries of each sub-unit. When the distance is less than a threshold (50 for example, determined based on the distance between adjacent strings), they are considered adjacent, and the matrix element is marked as 1. The result of constructing the adjacency matrix is ​​as follows: Figure 5 As shown; Next, the adjustment path is planned. Based on the adjacency matrix, the shortest path method is used to plan the group assignment path from multiple sub-units to fewer sub-units, such as finding the shortest transition path from sub-unit 1 to sub-unit 2; then, the boundary groups are assigned based on the comprehensive weight model: according to... The process involves sorting and prioritizing strings that are close together and have minimal impact on the profile coefficient. The 87 boundary strings of subunit 1 are then distributed among multiple subunits with fewer strings (33 of which are assigned to subunit 2). Finally, iterative adjustments are made, recalculating the adjacency matrix and deviation after each allocation until the number of strings in all subunits equals the target value. After iteration, all subunits have 230 strings, and the average profile coefficient increases from 0.58 to 0.72, significantly improving compactness and independence, meeting the requirements for construction in complex mountainous terrain. The final result of the complex mountainous photovoltaic subunit division is as follows: Figure 6 As shown, the boundaries between sub-units are clear and the space is compact, meeting the design requirements of complex mountain photovoltaic power stations.

[0041] Example 2

[0042] As a preferred embodiment of the above embodiments, a complex mountain photovoltaic sub-unit partitioning system based on adaptive optimization clustering is proposed. This system is implemented based on the complex mountain photovoltaic sub-unit partitioning method based on adaptive optimization clustering described in any of the above embodiments, such as... Figure 2 As shown, it includes: The data acquisition and region division module is used to acquire the spatial coordinate data of the photovoltaic strings of the target photovoltaic power station, perform preliminary clustering through density clustering algorithm to obtain the regional outline of the photovoltaic strings of the target photovoltaic power station, and divide multiple initial regions based on density accessibility and connectivity. The sub-unit pre-division module, based on the obtained multiple initial regions and the preset target number of sub-unit strings, pre-divides each initial region into sub-units using a center point optimization iterative algorithm, generating multiple pre-divided sub-units. The adjacency matrix construction module is used to calculate the deviation between the actual number of strings in each pre-divided sub-unit and the target number of strings in the sub-unit based on the obtained multiple pre-divided sub-units, and construct an adjacency matrix that reflects the spatial topological relationship of each pre-divided sub-unit. The photovoltaic string allocation optimization module is used to perform global dynamic allocation and iterative adjustment of strings in the pre-divided sub-units based on the adjacency relationship matrix and through a comprehensive weight model, until the number of strings in all pre-divided sub-units is the same as the target number of strings in the sub-unit, thus obtaining the photovoltaic sub-unit division result of the target photovoltaic power station.

[0043] Specifically, the implementation principle and process of the above system are as follows: First, the system initiates the data acquisition and region division module. This module reads the original spatial coordinate data of all photovoltaic strings in the target photovoltaic power station through an interface or database, and performs noise reduction and standardization preprocessing. Based on the spatial distribution density characteristics of the preprocessed coordinate data, the neighborhood radius and core point threshold parameters of the clustering algorithm are adaptively set, and the density clustering algorithm is executed to identify photovoltaic string clusters with density connectivity, thereby eliminating discrete noise points and identifying the natural regional outline of the photovoltaic strings. Based on the principles of density accessibility and connectivity, the entire power station is divided into multiple non-interfering initial regions.

[0044] Subsequently, the sub-unit pre-division module receives multiple initial region data obtained from the above division. For each initial region, the system counts the total number of photovoltaic strings contained within it and, combined with the project's preset target number of strings for a single sub-unit (such as the rated capacity of an inverter or combiner box), calculates the number of sub-units that should be divided in the initial region. Then, using a centroid optimization iterative algorithm (such as K-means and its variants), cluster centers are randomly initialized in the initial region and iteratively allocated and updated, prioritizing the spatial geometric compactness of the sub-units. This generates multiple pre-division sub-units for each initial region with string numbers that approximate but are not necessarily strictly equal to the target value.

[0045] Next, the adjacency matrix construction module performs topological analysis and deviation calculation on the pre-divided sub-units generated above. On the one hand, the module traverses all pre-divided sub-units, counts the actual number of strings in each unit, compares it with the preset target number of strings in the sub-unit, and calculates the string number deviation value (i.e., the number of missing or excess strings) for each sub-unit. On the other hand, the module constructs the convex hull boundary of each pre-divided sub-unit through computational geometry algorithm, and calculates the minimum Euclidean distance between the convex hulls of any two sub-units. When the distance is less than the set neighborhood threshold, the two are determined to be adjacent. Based on this, an adjacency matrix reflecting the spatial adjacency relationship between all pre-divided sub-units is constructed.

[0046] Finally, the photovoltaic string allocation optimization module intervenes based on the generated adjacency matrix and string number deviation information. This module uses a graph search algorithm to plan the adjustment path for transferring strings from excess sub-units to deficient sub-units, and uses a comprehensive weight model to calculate the allocation priority score of the boundary strings to be adjusted. This model comprehensively considers the geometric distance of the string to the target sub-unit and the impact of the movement on the overall clustering contour coefficient. According to the priority, the strings are moved from the source sub-unit to the target adjacent sub-unit one by one. After each movement, the adjacency matrix and deviation status are updated in real time. Through multiple rounds of global dynamic allocation and iterative adjustment, until the string number deviation of all pre-divided sub-units is zero, that is, completely the same as the target string number of the sub-unit, the final output is the photovoltaic sub-unit division result of the target photovoltaic power station that meets the quantity constraints and has the optimal spatial shape.

[0047] Example 3

[0048] Furthermore, as a preferred embodiment of the present invention, a terminal device based on an adaptive optimization clustering method for partitioning complex mountain photovoltaic sub-units is proposed, such as... Figure 7 As shown, the terminal device 200 includes at least one memory 210, at least one processor 220, and a bus 230 connecting different platform systems.

[0049] The memory 210 may include a readable medium in the form of volatile memory, such as random access memory (RAM) 211 and / or cache memory 212, and may further include read-only memory (ROM) 213.

[0050] The memory 210 also stores a computer program that can be executed by the processor 220, causing the processor 220 to execute any of the above-described adaptive optimization clustering-based methods for partitioning complex mountain photovoltaic sub-units in this application. The specific implementation and technical effects are consistent with those described in the above embodiments, and some details will not be repeated here. The memory 210 may also include a program / utility 214 having a set (at least one) of program modules 215. Such program modules include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0051] Accordingly, processor 220 can execute the aforementioned computer program, as well as executable program / utility 214.

[0052] Bus 230 can represent one or more of several types of bus structures, including a memory bus or memory controller, peripheral bus, graphics acceleration port, processor, or a local bus using any of the various bus structures.

[0053] Terminal device 200 can also communicate with one or more external devices 240, such as keyboards, pointing devices, Bluetooth devices, etc., and with one or more devices capable of interacting with it, and / or with any device that enables it to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication can be performed via input / output (I / O) interface 250. Furthermore, terminal device 200 can also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 260. Network adapter 260 can communicate with other modules of terminal device 200 via bus 230. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with terminal device 200, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0054] Example 4

[0055] As a preferred embodiment of Example 1, a computer-readable storage medium is proposed for a method of partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering. The computer-readable storage medium stores instructions that, when executed by a processor, implement any of the aforementioned methods for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering. Its specific implementation method and the achieved technical effects are consistent with those described in the above embodiments, and some details will not be repeated.

[0056] like Figure 8 As shown, the program product 300 provided in this embodiment for implementing the above-described method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering can be a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a personal computer. However, the program product 300 of the present invention is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The program product 300 can employ any combination of one or more readable media. A readable medium can be a readable signal medium or a readable storage medium. A readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0057] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof. Program code for performing operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user device, as a standalone software package, partially on a user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing devices can be connected to user computing devices via any type of network, including local area networks (LANs) or wide area networks (WANs), or they can be connected to external computing devices (e.g., via the Internet using an Internet service provider).

[0058] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering, characterized in that, Includes the following steps: S1. Obtain the spatial coordinate data of the photovoltaic strings of the target photovoltaic power station, perform preliminary clustering through density clustering algorithm to obtain the regional outline of the photovoltaic strings of the target photovoltaic power station, and divide multiple initial regions based on density accessibility and connectivity; S2. Based on the obtained multiple initial regions and the preset target number of sub-unit strings, the center point optimization iterative algorithm is used to pre-divide each initial region into sub-units to generate multiple pre-divided sub-units. S3. Based on the obtained multiple pre-divided sub-units, calculate the deviation between the actual number of strings in each pre-divided sub-unit and the target number of strings in the sub-unit, construct an adjacency relation matrix reflecting the spatial topological relationship of each pre-divided sub-unit, and based on the adjacency relation matrix, perform global dynamic allocation and iterative adjustment of strings in the pre-divided sub-units through a comprehensive weight model until the number of strings in all pre-divided sub-units is the same as the target number of strings in the sub-unit, and obtain the photovoltaic sub-unit division result of the target photovoltaic power station.

2. The method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering according to claim 1, characterized in that, Step S1 includes the following sub-steps: S101. Obtain the spatial coordinate data of all photovoltaic strings in the target photovoltaic power station, perform standardized preprocessing, and extract the spatial density features of the photovoltaic strings; S102. Based on the spatial density characteristics of photovoltaic strings and the terrain data of the target photovoltaic power station, the density clustering algorithm is adaptively initialized with parameters including the neighborhood radius parameter and the minimum number of points contained in the core point threshold parameter. S103. Based on the density clustering algorithm after adaptive parameter initialization, traverse all photovoltaic string spatial coordinate data, identify core points, boundary points and noise points, and connect core points and boundary points respectively based on density accessibility and density reachability relationship to obtain initial region clusters. The noise points include isolated points that do not belong to any initial region cluster. S104. Remove noise points from the initial region clusters to obtain multiple initial regions for the photovoltaic strings of the target photovoltaic power station.

3. The method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering according to claim 1, characterized in that, Step S2 includes the following sub-steps: S201. Based on the multiple initial regions obtained in step S1, count the total number of photovoltaic strings in each initial region, and calculate the number of sub-unit divisions in each initial region based on the preset target number of sub-unit strings. S202. Based on the number of sub-units, randomly select the same number of points in each initial region as initial cluster centers, calculate the Euclidean distance from each string to each cluster center, and assign the string to the sub-unit to which the nearest center point belongs. S203. Using the central store optimization iterative algorithm, the geometric center coordinates of each sub-unit are updated sequentially. The allocation and update steps are repeated until the cluster center positions converge, resulting in multiple pre-divided sub-units.

4. The method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering according to claim 1, characterized in that, Step S3, which involves constructing an adjacency matrix reflecting the spatial topological relationships of each pre-divided sub-unit, includes: constructing the convex hull boundaries of all strings surrounding each sub-unit using computational geometry algorithms, and calculating the minimum Euclidean distance between the convex hull boundaries of any two sub-units; based on a dynamic distance threshold, if the minimum distance between two sub-units is less than the dynamic distance threshold, then the two are determined to be adjacent, and the corresponding element in the adjacency matrix is ​​marked as connected; otherwise, it is marked as disconnected.

5. The method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering according to claim 4, characterized in that, The dynamic distance threshold is determined by the actual string distance of each photovoltaic string within the target photovoltaic power station under operating conditions.

6. The method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering according to claim 1, characterized in that, Step S3, which describes the global dynamic allocation and iterative adjustment of strings within the pre-divided sub-units using a comprehensive weighting model, includes: counting the actual number of strings in each pre-divided sub-unit; calculating the deviation between the actual number of strings and the target number of strings in the sub-unit; marking sub-units with an actual number of strings greater than the target number as multi-string sub-units; marking sub-units with an actual number of strings less than the target number as few-string sub-units; and marking the rest as qualified sub-units. Based on the adjacency matrix, the adjustment path for transferring strings from multi-string sub-units to few-string sub-units is calculated using a shortest path algorithm. For adjacent source and target sub-units on the adjustment path, the set of strings to be divided in the source sub-units that are close to the boundary of the target sub-unit is identified. The allocation priority score of the strings to be divided is calculated using the comprehensive weighting model. The strings are moved from the source sub-units to the target sub-units in order of their scores, and the adjacency matrix and the string number deviation are updated in real time until the string number deviation of all sub-units is zero, thus obtaining the photovoltaic sub-unit division result of the target photovoltaic power station.

7. The method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering according to claim 6, characterized in that, The comprehensive weight model is expressed as follows: ； in, Indicates the string to be divided The allocation priority score, Indicates the string to be divided The Euclidean distance to the boundary string of the target group. Represents a string The change in the group profile coefficient after being assigned to the target group. This represents the distance weight value. This represents the contour coefficient weight value.

8. A system for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering, the system being implemented based on the method for partitioning complex mountain photovoltaic sub-units based on adaptive optimization clustering as described in any one of claims 1-7, characterized in that, include: The data acquisition and region division module is used to acquire the spatial coordinate data of the photovoltaic strings of the target photovoltaic power station, perform preliminary clustering through density clustering algorithm to obtain the regional outline of the photovoltaic strings of the target photovoltaic power station, and divide multiple initial regions based on density accessibility and connectivity. The sub-unit pre-division module, based on the obtained multiple initial regions and the preset target number of sub-unit strings, pre-divides each initial region into sub-units using a center point optimization iterative algorithm, generating multiple pre-divided sub-units. The adjacency matrix construction module is used to calculate the deviation between the actual number of strings in each pre-divided sub-unit and the target number of strings in the sub-unit based on the obtained multiple pre-divided sub-units, and construct an adjacency matrix that reflects the spatial topological relationship of each pre-divided sub-unit. The photovoltaic string allocation optimization module is used to perform global dynamic allocation and iterative adjustment of strings in the pre-divided sub-units based on the adjacency relationship matrix and through a comprehensive weight model, until the number of strings in all pre-divided sub-units is the same as the target number of strings in the sub-unit, thus obtaining the photovoltaic sub-unit division result of the target photovoltaic power station.

Images