A photovoltaic array topology generation method and system based on digital terrain analysis

Abstract

The present application relates to photovoltaic power station planning and design technical field, it discloses a kind of photovoltaic array topological generation method and system based on digital terrain analysis, solve the problem that traditional scheme exists under complex terrain terrain adaptability is poor, low efficiency, insufficient illumination utilization and multiple engineering constraints are difficult to coordinate.This application carries out curvature-guided adaptive gridding to digital elevation model, generate the micro-slope unit grid that conforms to real terrain, combined with annual solar view position time series data, carry out radiation receiving amount space-time integration and light receiving efficiency calibration to each micro-slope unit;Adopt efficiency priority level driving sequential layout synthesis, first construct compliance main array in high-efficiency area, then fill gap by variable efficiency unit to form initial topological layout;Finally rely on geometric projection principle to carry out shelter simulation analysis, combined with slope, site boundary constraint carries out iterative optimization adjustment, output optimal topological layout satisfying each item engineering limit.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power plant planning and design, and specifically to a method and system for generating photovoltaic array topology based on digital terrain analysis. Background Technology

[0002] With the widespread application of photovoltaic (PV) power generation technology, the construction scenarios for PV power plants are expanding from flat areas to complex terrains such as mountains and hills. Complex terrain areas offer greater potential for development, but also place higher demands on the layout and design of PV arrays. To fully utilize the solar resources in these areas while ensuring the economic viability and safety of the power plants, developing intelligent and automated PV array design methods that can accurately adapt to complex terrains has become an urgent need for the industry.

[0003] In existing technologies, the design of photovoltaic arrays for complex terrain generally adopts an auxiliary design approach based on two-dimensional planar maps or simplified digital elevation models. Designers mainly rely on topographic contour maps and engineering experience to manually plan the placement and orientation of photovoltaic modules. This process is usually carried out with the help of general computer-aided design software, and basic spacing and slope requirements are attempted to be met through repeated manual adjustments. In this process, the assessment of solar radiation resources often relies on regional meteorological data or simplified illumination models, making it difficult to perform refined coupling analysis with the micro-geometric characteristics of the specific terrain.

[0004] The defects of the aforementioned existing technology are:

[0005] (1) The human-driven design process is highly dependent on experience, resulting in low design efficiency and difficulty in reproduction.

[0006] (2) Insufficient analysis of three-dimensional terrain results in poor fit between the layout scheme and the real terrain, which may lead to local shadow occlusion, structural instability or increased workload.

[0007] (3) The lack of accurate calculation of the annual radiation reception potential of each location on the terrain can easily lead to a waste of light resources.

[0008] (4) It is difficult to coordinate multiple engineering constraints such as slope, spacing, and boundary manually. It is easy to overlook one aspect and thus affect the engineering feasibility and power generation revenue of the scheme.

[0009] In summary, existing photovoltaic array topology designs face challenges such as poor terrain adaptability, low design efficiency, insufficient utilization of solar resources, and difficulty in coordinating and satisfying multiple engineering constraints under complex terrain conditions. Summary of the Invention

[0010] The technical problem to be solved by the present invention is to provide a photovoltaic array topology generation method and system based on digital terrain analysis, so as to solve the problems of poor terrain adaptability, low efficiency, insufficient light utilization and difficulty in coordinating multiple engineering constraints in existing photovoltaic array design schemes under complex terrain.

[0011] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0012] On the one hand, the present invention provides a photovoltaic array topology generation method based on digital terrain analysis, comprising the following steps:

[0013] S1. Acquire digital terrain elevation data, solar apparent position time series data, photovoltaic unit reference geometric data, micro-slope unit subdivision threshold parameters, and array layout constraint data;

[0014] S2. Based on the micro-sloping unit subdivision threshold parameter, the digital terrain elevation data is subjected to curvature-guided meshing processing to generate a micro-sloping unit mesh;

[0015] S3. Based on the micro-sloping surface unit grid and the solar apparent position time series data, the light-receiving efficiency of each micro-sloping surface unit is calibrated to obtain the light-receiving efficiency calibration results of each micro-sloping surface unit;

[0016] S4. Based on the light-receiving efficiency calibration results and the photovoltaic unit reference geometric data, sequential layout synthesis is performed according to efficiency priority to generate a synthesized topology layout;

[0017] S5. Based on the principle of geometric projection, perform occlusion analysis on the synthesized topology layout, and combine slope constraints and boundary constraints for iterative adjustment to obtain the corrected synthesized topology layout.

[0018] In this solution, a complete automated link from terrain analysis to final layout is constructed by sequentially executing data acquisition, curvature adaptive meshing, radiation performance calibration, performance-priority sequential layout, occlusion simulation and constraint iterative adjustment. It no longer relies on human experience and two-dimensional simplified terrain, but is based on three-dimensional micro-terrain and annual solar radiation, which fundamentally solves the problems of poor adaptability to complex terrain, insufficient light utilization and difficulty in coordinating multiple constraints.

[0019] Furthermore, in step S1, the digital terrain elevation data is a discrete or continuous dataset describing the elevation distribution of the terrain surface; the solar apparent position time series data is a time series describing the changes in solar azimuth and elevation angles throughout the year; the photovoltaic unit reference geometric data is a set of geometric parameters describing the size and shape of standard photovoltaic modules; the micro-slope unit subdivision threshold parameter is a scalar value controlling the subdivision accuracy of the terrain surface; and the array layout constraint data is a set of rules describing the slope, spacing, and boundary constraints in the actual site.

[0020] In this scheme, the solar apparent position time series data ensures the authenticity of the annual radiation calculation, the photovoltaic unit reference geometric data ensures the accuracy of the component placement and spacing judgment, the micro-slope unit subdivision threshold parameter controls the terrain analysis accuracy, and the array layout constraint data ensures that the final layout meets the engineering specifications. This serves as a unified and clear data foundation for subsequent steps.

[0021] Further, in step S2, based on the micro-slope unit subdivision threshold parameter, curvature-guided meshing processing is performed on the digital terrain elevation data to generate a micro-slope unit mesh, including:

[0022] S21. Perform Gaussian curvature field calculation on the digital terrain elevation data to determine the principal curvature of each sampling point on the terrain surface, and obtain the Gaussian curvature value of each sampling point through differential geometric operations, thereby obtaining the terrain curvature distribution field;

[0023] S22. Based on the terrain curvature distribution field, control the grid density and use the micro-sloping element subdivision threshold parameter as the grid subdivision stopping criterion. Use the leading edge advancement method to generate anisotropic triangular meshes to obtain micro-sloping element meshes composed of triangular facets.

[0024] In this scheme, the severity of terrain undulation is quantified by Gaussian curvature field, so that the mesh density is adaptively matched with the micro-geometric features of the terrain: the mesh is automatically densified in high curvature areas to improve the terrain fit; the mesh is automatically sparsed in low curvature areas to reduce the amount of computation; at the same time, a subdivision threshold is used as a stopping condition to balance accuracy and efficiency; and an anisotropic triangular mesh is generated by the leading edge method to make the micro-sloping elements fit the real terrain surface better, providing a geometric basis for subsequent accurate radiation calculation and reliable layout installation.

[0025] Furthermore, in step S3, based on the micro-sloping surface unit grid and the solar apparent position time series data, the light-receiving efficiency of each micro-sloping surface unit is calibrated to obtain the light-receiving efficiency calibration results for each micro-sloping surface unit, including:

[0026] S31. For each cell in the micro-sloping cell grid, perform spatiotemporal integration processing in conjunction with the solar apparent position time series data to obtain the normal direct radiation component of the cell, and use it as the radiation receiving reference value.

[0027] S32. Based on the normal direct radiation component and the preset energy flux density threshold, the performance is calibrated to obtain the light-receiving performance calibration results for each micro-sloping unit.

[0028] In this scheme, by first performing a solar radiation spatiotemporal integral on each micro-sloping unit, the normal direct radiation component that can truly reflect the relationship between the terrain orientation and solar incidence is obtained, thus achieving accurate quantification of the light-receiving capacity of the micro-terrain unit; then, based on a preset threshold, the radiation component is calibrated for performance classification, converting continuous radiation values ​​into discrete performance levels that can be directly used for layout decisions, providing objective and reliable data for subsequent photovoltaic array arrangement according to performance priority.

[0029] Furthermore, in step S31, the spatiotemporal integration process is to integrate the product of the cosine of the angle between the unit surface normal and the solar rays and the theoretical value of the normal solar irradiance outside the atmosphere over the continuous time domain of the solar apparent trajectory.

[0030] In step S32, the performance calibration includes: comparing the normal direct radiation component of each unit with a preset threshold range, classifying the units into high-performance unit type, variable-performance unit type or low-performance unit type according to the comparison results, and assigning a unique calibration identifier to each category.

[0031] In this scheme, the annual solar motion is geometrically coupled with the normal of the micro-sloping unit through spatiotemporal integration, and the total amount of direct radiation that each unit can receive throughout the year is accurately calculated, replacing the traditional regional and simplified radiation estimation, and realizing the accurate quantification of the radiation potential of micro-viewpoints; then, the continuous radiation value is converted into discrete performance levels through threshold classification, so that the layout strategy can be directly executed based on the performance level.

[0032] Furthermore, in step S4, based on the light-receiving efficiency calibration results and the photovoltaic unit reference geometric data, a sequential layout synthesis is performed according to efficiency priority to generate a synthesized topology layout, including:

[0033] S41. Construct a set of main skeleton units based on the high-efficiency unit class in the unit light-receiving efficiency calibration and the photovoltaic unit reference geometric data;

[0034] S42. Based on the main skeleton unit set, the variable efficiency unit class in the unit light-receiving efficiency calibration, and the photovoltaic unit reference geometric data, a synthetic topology layout is generated by filling the gap regions between the main skeleton unit sets.

[0035] In this scheme, sequential layout synthesis is performed according to efficiency priority. First, a set of main frame units is constructed with high-efficiency units to ensure that the main body of the photovoltaic array is placed in the area with the strongest radiation reception capacity, thus guaranteeing the core power generation revenue. Then, variable efficiency units are used to reasonably fill the gaps in the main frame, making full use of the site space and improving land utilization without violating the component space constraints and spacing rules. Through the layered layout method of main frame construction and gap filling, the synthesized topology layout takes into account power generation efficiency, spatial compliance and layout integrity, providing a structurally reasonable initial layout for subsequent shading analysis and constraint adjustment.

[0036] Furthermore, in step S41, the construction of the main skeleton unit set involves selecting a unit from the high-efficiency unit class as the starting unit, and iteratively selecting and connecting neighboring high-efficiency units that meet the repulsion relationship based on the spatial repulsion relationship between the component envelope shapes defined by the photovoltaic unit reference geometric data, until no unit that meets the condition can be connected.

[0037] In step S42, the generation of the synthetic topology layout is achieved by identifying the continuous unoccupied area defined by adjacent unit pairs in the main skeleton unit set, selecting units from the variable performance unit class whose component envelope shape is located in the area and does not overlap with the component envelope shape of any main skeleton unit, and adding the selected units to the filling unit set in descending order of light reception performance.

[0038] The synthetic topology layout is formed by merging the main skeleton unit set and the filling unit set.

[0039] In this scheme, by selecting the best starting point from the high-efficiency units and iteratively expanding according to the spatial repulsion relationship of the component envelope shape, a main frame layout with structural stability and optimal radiation benefit can be automatically formed under the premise of ensuring that the components do not overlap and meet the minimum spacing constraint, thus ensuring the reliability and efficiency of the array core power generation area. At the same time, by accurately identifying the gaps in the main frame, selecting suitable variable efficiency units and filling them in order of light reception efficiency from high to low, the use of idle space can be maximized without damaging the main frame or causing interference or shading.

[0040] Furthermore, in step S5, occlusion analysis is performed on the synthesized topology layout based on the principle of geometric projection, and iterative adjustments are made in conjunction with slope constraints and boundary constraints to obtain the corrected synthesized topology layout, including:

[0041] S51. Based on the synthetic topology layout and the spacing rules defined in the array layout constraint data, perform geometric projection occlusion analysis to generate a set of cell-level occlusion coefficients.

[0042] S52. Based on the set of unit-level occlusion coefficients and the slope and boundary constraints in the array layout constraint data, perform constraint satisfaction adjustment processing to obtain the corrected synthetic topology layout.

[0043] In this solution, occlusion analysis of the synthetic topology layout is performed using the principle of geometric projection. This transforms the occlusion relationships between components into quantifiable unit-level occlusion coefficients, enabling accurate identification of occlusion risks and avoiding subjectivity and bias caused by manual judgment. Furthermore, constraint adjustments are performed based on the occlusion coefficients, slope limits, and boundary ranges, which can simultaneously resolve engineering issues such as excessive occlusion, excessive slope, and layout exceeding boundaries. Through iterative optimization, the final layout fully meets the engineering construction requirements.

[0044] Furthermore, in step S51, the geometric projection occlusion analysis process calculates and superimposes the projection areas of each unit in the typical solar altitude angle domain to determine the occlusion time ratio of each component unit and generate a set of unit-level occlusion coefficients.

[0045] In step S52, the constraint satisfaction adjustment process involves identifying units whose occlusion coefficient exceeds a preset threshold as units to be adjusted, performing position translation or axial rotation on the units to be adjusted based on slope constraints, and trimming layout portions that exceed the permissible range based on boundary constraints, thereby generating a corrected topology layout that satisfies all given constraints.

[0046] In this scheme, by performing point-by-point calculations and time-series overlays on the projection areas of each unit within a typical solar altitude angle domain, the shading relationships between components at different times throughout the year can be accurately restored. This transforms the abstract shading problem into a quantifiable proportion of shading time, forming a set of unit-level shading coefficients, enabling an objective and precise assessment of the degree of shading. Based on this, units with excessive shading coefficients are automatically identified and adjusted. Combined with slope constraints, these units are translated or rotated axially to reduce the impact of shading without violating terrain slope limitations. At the same time, out-of-bounds layouts are trimmed according to site boundary constraints, ensuring that all units meet multiple constraints of shading, slope, and boundaries. Ultimately, a corrected topology layout that is safe, compliant, generates stable power, and can be directly implemented in engineering is generated.

[0047] On the other hand, the present invention also provides a photovoltaic array topology generation system based on digital terrain analysis for implementing the above method, the system comprising:

[0048] The acquisition unit is used to acquire digital terrain elevation data, solar apparent position time series data, photovoltaic unit reference geometric data, micro-slope unit subdivision threshold parameters, and array layout constraint data;

[0049] A meshing processing unit is used to perform curvature-guided meshing processing on digital terrain elevation data based on the micro-slope unit subdivision threshold parameters to generate a micro-slope unit mesh.

[0050] The efficiency calibration unit is used to calibrate the light-receiving efficiency of each micro-sloping unit based on the micro-sloping unit grid and the solar apparent position time series data, and obtain the light-receiving efficiency calibration results of each micro-sloping unit.

[0051] Topology synthesis unit is used to generate a synthesized topology layout by sequentially arranging and synthesizing the photovoltaic unit based on the light-receiving efficiency calibration results and the reference geometric data of the photovoltaic unit according to the efficiency priority.

[0052] The compliance adjustment unit is used to perform occlusion analysis on the synthetic topology layout based on the principle of geometric projection, and to perform iterative adjustment in combination with slope constraints and boundary constraints to obtain the corrected synthetic topology layout.

[0053] In this solution, a modular architecture and methodological steps are used in a one-to-one correspondence, with each unit working together to form an automated processing closed loop: the acquisition unit obtains unified input, the grid processing unit completes terrain adaptive analysis, the efficiency calibration unit completes accurate radiation assessment, the topology synthesis unit completes intelligent layout, and the compliance adjustment unit completes engineering constraint verification and correction; the entire system replaces the manual design process, realizes the automated, intelligent, and standardized generation of photovoltaic array topologies in complex terrains, and improves design efficiency, scheme consistency, and reproducibility.

[0054] The beneficial effects of this invention are:

[0055] (1) Enhanced terrain adaptability and significantly improved adaptability to complex terrain:

[0056] This invention employs Gaussian curvature-guided adaptive meshing of digital terrain elevation data. It automatically adjusts mesh density based on the severity of terrain undulations, densifying the mesh in high-curvature areas such as ridges and valleys, and sparsening it in flat areas, generating micro-sloping units that closely resemble the actual terrain. Compared to traditional two-dimensional simplified terrain or rough manual fitting methods, this invention accurately reproduces the micro-geometric features of three-dimensional terrain, ensuring a high degree of match between the photovoltaic array layout and the terrain surface. This effectively avoids problems such as shading, unstable structural installation, and excessive earthwork caused by insufficient terrain fit, improving the layout adaptability and engineering feasibility in complex mountainous and hilly terrains.

[0057] (2) Improved accuracy in light efficiency assessment and maximized utilization of light resources:

[0058] This invention, based on annual solar apparent position time-series data, performs spatiotemporal radiation integral calculations on each micro-sloping surface unit to accurately determine the annual normal direct radiation component of each unit, achieving quantitative calibration of the unit-level solar radiation potential. Furthermore, through threshold classification, the units are categorized into high-efficiency, variable-efficiency, and low-efficiency levels, forming a clear basis for layout decisions. Compared to traditional methods relying on regional meteorological data or simplified illumination models, this invention achieves refined coupling between topography and solar radiation, enabling priority utilization of high-radiation areas, fundamentally avoiding waste of solar resources, and improving the power generation revenue per unit area of ​​photovoltaic power plants.

[0059] (3) Layout generation is automated and intelligent, greatly improving design efficiency and consistency of solutions:

[0060] This invention employs a performance-priority-driven sequential layout strategy. It first constructs a high-performance main frame array, then fills the gaps with variable-performance units, adhering to component spatial repulsion and spacing constraints throughout the process. This eliminates the need for repeated manual adjustments, line drawing, and arrangement. The entire process, from data input to topology output, is fully automated, reducing reliance on designer experience. The solution is reproducible and standardized, significantly shortening the design cycle for photovoltaic power plants in complex terrains and improving design efficiency and solution stability.

[0061] (4) Multiple engineering constraints are met in a coordinated manner, resulting in a more compliant and reliable layout scheme:

[0062] This invention quantifies the degree of shading into unit-level shading coefficients through geometric projection shading analysis, enabling precise identification of shading risks. Using shading thresholds, slope limitations, and site boundaries as rigid constraints, iterative adjustments are made to complete unit translation, rotation, and boundary pruning, ensuring the final layout simultaneously meets all engineering requirements for shading prevention, slope limitation, boundary protection, and spacing maintenance. Compared to the shortcomings of manual coordination of multiple constraints, which is prone to oversight, this invention systematically eliminates layout hazards, improving the long-term safety, stability, and power generation reliability of the power plant. Attached Figure Description

[0063] Figure 1 This is a flowchart of the photovoltaic array topology generation method based on digital terrain analysis in an embodiment of the present invention.

[0064] Figure 2 This is a structural diagram of a photovoltaic array topology generation system based on digital terrain analysis in an embodiment of the present invention.

[0065] Figure 3 This is a schematic diagram illustrating the relationship between the terrain curvature field and the adaptive mesh size in an embodiment of the present invention.

[0066] Figure 4 This is a schematic diagram of the light-receiving efficiency distribution and threshold calibration of the micro-sloping surface unit in an embodiment of the present invention. Detailed Implementation

[0067] This invention aims to provide a photovoltaic array topology generation method and system based on digital terrain analysis, solving the problems of poor terrain adaptability, low efficiency, insufficient light utilization, and difficulty in coordinating multiple engineering constraints in existing photovoltaic array designs under complex terrain. Its core idea is: based on high-precision digital terrain analysis, adaptive meshing guided by Gaussian curvature is used to perform micro-geometric fitting on complex terrain, transforming continuous terrain surfaces into micro-sloping surface unit meshes that highly match the terrain undulations, achieving precise quantitative expression of terrain features; based on the apparent annual solar motion trajectory, each micro-sloping surface unit is calibrated by radiation spatiotemporal integration and light-receiving efficiency classification, establishing a precise correspondence between terrain units and solar radiation receiving capacity, forming an objective and quantifiable basis for optimal layout; according to the priority of light-receiving efficiency from high to low, a backbone-based topology generation method is adopted. The sequential layout strategy of building and filling gaps prioritizes the construction of a main skeleton array that meets the spacing constraints in high-efficiency areas, and then uses variable efficiency units to fill the gaps in a compatible manner, maximizing land utilization while ensuring the core power generation revenue. Finally, based on the principle of geometric projection, a year-round shading simulation analysis is conducted to convert the degree of shading into a quantifiable unit-level shading coefficient. With shading threshold, slope limit, and site boundary as rigid constraints, the final topology layout meets all engineering requirements of anti-shading, slope limit, boundary protection, and spacing protection through iterative adjustment, translation and rotation, and boundary trimming.

[0068] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0069] Example:

[0070] This embodiment first provides a photovoltaic array topology generation method based on digital terrain analysis, see [link to relevant documentation]. Figure 1 It includes the following implementation steps:

[0071] S1. Acquire digital terrain elevation data, solar apparent position time series data, photovoltaic unit reference geometric data, micro-slope unit subdivision threshold parameters, and array layout constraint data;

[0072] In this step, the digital terrain elevation data is a discrete or continuous dataset describing the elevation distribution of the terrain surface; the solar apparent position time series data is a time series describing the changes in solar azimuth and elevation angles throughout the year; the photovoltaic unit reference geometric data is a set of geometric parameters describing the size and shape of standard photovoltaic modules; the micro-slope unit subdivision threshold parameter is a scalar value controlling the subdivision accuracy of the terrain surface; and the array layout constraint data is a set of rules describing the slope, spacing, and boundary constraints in the actual site.

[0073] In this embodiment, digital terrain elevation data is acquired through remote sensing or ground mapping techniques and is used to characterize the spatial distribution features of site topographic relief. This data stores elevation information in the form of point clouds or regular grids. The elevation information is used to define the three-dimensional spatial coordinates of the terrain surface and serves as input data for subsequent terrain surface curvature-guided gridding processing, enabling the calculation of local terrain geometric features based on the elevation distribution. The methods for acquiring and preprocessing the digital terrain elevation data are well-known to those skilled in the art and will not be elaborated upon here.

[0074] In this embodiment, the apparent solar position time-series data is generated based on astronomical calculation formulas and is used to reflect the apparent motion trajectory of the sun relative to the observation point. This data includes the solar azimuth and elevation angle values ​​over a continuous time series, which can be used to calculate the radiation receiving reference value for each micro-sloping surface unit; the relevant calculations require integrating the cosine of the angle between the solar ray direction and the unit normal over a continuous time domain. The method for generating the apparent solar position time-series data is a well-known prior art and will not be elaborated upon here.

[0075] In this embodiment, the photovoltaic unit reference geometric data includes geometric parameters such as the length, width, and outline shape of a standard photovoltaic module. These parameters are used to define the envelope rectangle corresponding to the photovoltaic module during the layout synthesis process. This envelope rectangle can be used for spacing constraint judgment and gap filling. Specifically, the size and shape parameters of the module directly determine its occupancy range on the micro-sloping unit mesh, thereby affecting the generation results of the backbone layout and the filling layout.

[0076] In this embodiment, the micro-sloping element subdivision threshold parameter is set based on the terrain curvature distribution field to guide the element size control during the meshing process. This parameter determines the density of the micro-sloping element mesh, and the mesh density directly affects the resolution of subsequent light-receiving performance calibration. Simultaneously, the micro-sloping element subdivision threshold parameter serves as the termination condition for mesh subdivision iteration, working in conjunction with the leading-edge propagation method to automatically generate triangular facet elements. Furthermore, the selection of this threshold parameter must balance computational accuracy and data processing efficiency, and can be flexibly adapted to the complexity of the site terrain and available hardware computing resources.

[0077] In this embodiment, the array layout constraint data is a set of rules characterizing the actual site slope, component spacing, and site boundary restrictions. Specifically, the slope constraint limits the allowable tilt angle range for photovoltaic modules; the spacing constraint sets the minimum spacing between modules, balancing shading prevention and future maintenance space requirements; and the boundary constraint delineates the legal deployment area for the photovoltaic array. The array layout constraint data guides the simulation and iterative adjustment of layout compliance throughout the process, ensuring that the final generated topology fully meets the requirements of actual engineering applications.

[0078] S2. Based on the micro-sloping unit subdivision threshold parameter, the digital terrain elevation data is subjected to curvature-guided meshing processing to generate a micro-sloping unit mesh;

[0079] In its specific implementation, this step includes the following sub-steps:

[0080] S21. Perform Gaussian curvature field calculation on the digital terrain elevation data to determine the principal curvature of each sampling point on the terrain surface, and obtain the Gaussian curvature value of each sampling point through differential geometric operations, thereby obtaining the terrain curvature distribution field;

[0081] Specifically, Gaussian curvature field calculations are performed on digital terrain elevation data to determine the principal curvature of each sampling point on the terrain surface. The Gaussian curvature value of each sampling point is then solved using differential geometric operations, comprising two steps: local surface fitting and differential invariant calculation. Local surface fitting uses elevation data from the sampling point and its neighborhood to construct a smooth surface patch to approximate the local geometry of the terrain surface. Differential invariant calculation, based on the fitted smooth surface patch, solves for the components of the curvature tensor, and then calculates the principal curvature and Gaussian curvature of the corresponding sampling point based on these curvature tensor components.

[0082] In this embodiment, digital terrain elevation data is provided in the form of a regular grid, with each grid point storing an elevation value. For any sampling point... Its elevation value is The coordinates are represented as The local surface fitting process uses a quadratic surface model, which defines an elevation function in the neighborhood of the sampling point. Specifically, the function Expressed as ,in , , , , , These are the fitting coefficients. The fitting process uses the least squares method, which calculates the coefficients based on the elevation data of all sampling points within the neighborhood. It should be noted that the neighborhood range is set based on the micro-sloping surface unit subdivision threshold parameter, and this setting affects the accuracy of the curvature calculation. Furthermore, local surface fitting processing is a method well-known to those skilled in the art, and will not be elaborated upon here.

[0083] In this embodiment, the differential invariant calculation is based on the fitted quadratic surface function. Calculate the partial derivatives. The partial derivatives include the first-order partial derivatives. and and second-order partial derivatives , and At the sampling point At this point, the partial derivative is obtained through analytical differentiation, for example, for... ,have , , , , Specifically, based on the partial derivatives, the components of the curvature tensor are calculated using the first and second fundamental forms. The coefficients of the first fundamental form are... , , The coefficient of the second basic form is , , Principal curvature and The maximum and minimum values ​​of the normal curvature are obtained by solving the equation. Gaussian curvature is obtained. It is the product of the principal curvatures, expressed as .

[0084] Furthermore, Gaussian curvature The formula is given directly from the relationship between the second and first fundamental forms: Substituting into the partial derivative expression, Gaussian curvature... The calculation formula is simplified to This formula is used to calculate the Gaussian curvature value for each sampling point. It should be noted that the calculation involves standard operations in differential geometry, which are well-known to those skilled in the art and will not be elaborated upon here.

[0085] In this embodiment, the terrain curvature distribution field is represented by the spatial distribution of Gaussian curvature values, and this distribution field is a scalar field. Specifically, the terrain curvature distribution field is defined as a function... ,in These are topographical plane coordinates. It is the Gaussian curvature value at the corresponding point. The function... Stored in discrete form, each grid point corresponds to a curvature value, forming a curvature value matrix. ,in and It is a grid index. It is a point The Gaussian curvature at the location is used. The terrain curvature distribution field is used for the generation of anisotropic triangular meshes in subsequent steps, and the generation adjusts the mesh density according to the curvature value. Preferably, the Gaussian curvature field calculation also includes the calculation of principal curvature directions, which are used to guide the anisotropy of mesh generation. The principal curvature directions are obtained through the eigenvectors of the curvature tensor, where the eigenvectors represent the maximum and minimum curvature directions of the terrain surface. The calculation is well known to those skilled in the art and will not be described in detail here.

[0086] S22. Based on the terrain curvature distribution field and the micro-sloping element subdivision threshold parameter, perform anisotropic triangular mesh generation, wherein the terrain curvature distribution field is used to control the mesh density, the micro-sloping element subdivision threshold parameter is used as the stopping criterion for mesh subdivision, and a micro-sloping element mesh composed of triangular facets is generated by the leading edge advancing method.

[0087] In this embodiment, the anisotropic triangular mesh generation process uses the terrain curvature distribution field to control the mesh density, and the micro-sloping element subdivision threshold parameter as the stopping criterion for mesh subdivision. Generating a micro-sloping element mesh composed of triangular facets using the leading edge method includes mesh size field definition processing and leading edge method mesh generation processing. The mesh size field definition processing constructs a spatially varying size constraint function based on the terrain curvature distribution field and the micro-sloping element subdivision threshold parameter. This function specifies the maximum side length of the triangular elements allowed to be generated in a local region. The leading edge method mesh generation processing iteratively inserts new points and connects them into triangles under boundary constraints based on the size constraint function until all regions satisfy the size constraints. This process directly generates the micro-sloping element mesh.

[0088] In this embodiment, the goal of the mesh size field definition process is to generate a size field function. This function specifies a target grid edge length for each point on the terrain. Specifically, the size field... The absolute value of Gaussian curvature calculated in step S21 They exhibit a negative correlation. The relationship is defined as follows: ,in This is the micro-sloping element subdivision threshold parameter, a positive scalar value that directly controls the reference accuracy of the mesh. The smaller the value, the smaller the overall grid size. and These are the preset maximum and minimum allowable side lengths, used to prevent unreasonable dimensions from being generated in extremely flat or extremely curved areas. It is a very small positive number used to avoid division by zero errors and ensure numerical stability. The formula ensures that the value is within the Gaussian curvature. Large areas (such as ridges and valleys), target side length Small, the mesh is refined; in flat regions with low curvature, the target side length... Large, sparse mesh. The micro-sloping unit subdivision threshold parameter... As a global precision control variable, it is built into the size field calculation formula, thereby achieving the technical purpose of using it as a grid subdivision stopping criterion. Furthermore, the size field can be calculated at discrete grid points and then continuously defined through interpolation methods, the implementation of which is well known to those skilled in the art.

[0089] In this embodiment, the leading-edge propulsion method mesh generation process uses the aforementioned size field. To constrain the terrain, a triangular mesh is generated within the defined terrain boundary region. The process iterates from the initial set of boundary segments (front edge). Specifically, in each iteration, an active edge AB is selected from the current front edge. The target edge length of the midpoint M of this active edge in the size field is calculated. According to geometric rules, try to insert a new point at the ideal location. This ensures that the newly generated triangle is as close to equilateral as possible, and that its circumradius is equal to... Adaptation is key. It should be noted that the insertion of a new point must satisfy two core constraints: First, the distance between the new point and any existing grid point must not be less than a scaling factor (e.g., 0.8 × l) of the size field value *l* at that point, to avoid generating excessively small cells; second, the triangle formed by the new point and existing leading edges cannot intersect with the existing grid. If a feasible point satisfying the constraints is found, that point is inserted, the newly formed triangle is added to the grid, and the leading edge set is updated. If no feasible point is found, the currently active edge is marked as having met the requirements and no longer participates in subdivision. This iterative process continues until the leading edge set is empty, indicating that the grid throughout the region satisfies the density requirements defined by the size field. The specific algorithm implementation of the leading edge advancement method is well-known to those skilled in the art and will not be elaborated upon here.

[0090] In this embodiment, the micro-sloping element mesh generated by the above processing is represented as a two-dimensional manifold triangular mesh. , specifically It is a set of vertices, each vertex Includes its three-dimensional spatial coordinates ,in It is determined by interpolation from the original digital terrain elevation data. It is a set of edges that connect two vertices. It is a set of triangular facets, each of which is a triangular facet. Consists of three vertices Defined sequentially, and carrying a unit normal vector. The normal vector is obtained by cross-product operation and normalization of the coordinates of the three vertices, which is well known to those skilled in the art and will not be elaborated further. The triangular mesh M constitutes a piecewise planar approximation of the terrain surface, wherein each triangular facet... This is essentially a micro-sloping element. The density of this mesh is determined by the size field. Adaptive control ensures higher geometric resolution at terrain features with significant curvature variations, providing a discretized geometric basis for subsequent light-receiving performance calibration. Preferably, smoothing or optimization processing can be performed after mesh generation to improve cell quality. Please refer to the appendix. Figure 3 As shown, attached Figure 3 Using the spatial distance along the terrain profile line as the horizontal axis, two vertical axes represent the terrain Gaussian curvature value and the target mesh cell side length, respectively. The terrain Gaussian curvature curve shown in the figure exhibits a multi-peak, multi-valley undulating shape, with its peaks and troughs corresponding to high-curvature feature areas such as ridges and valleys on the terrain surface and low-curvature feature areas such as gentle slopes. In stark contrast, the target mesh cell side length curve shows a negative correlation, with its trend completely opposite to the Gaussian curvature curve: at the peak of the Gaussian curvature curve, the mesh size curve drops to the trough; at the trough of the Gaussian curvature curve, the mesh size curve rises to the peak. This mirror relationship intuitively verifies the design principle that the size and absolute value of curvature are negatively correlated in the mesh size field function. The figure further indicates typical high-curvature and low-curvature regions through background coloring. In high-curvature regions, the mesh size is significantly reduced, achieving mesh refinement; in low-curvature regions, the mesh size increases accordingly, achieving mesh sparseness. The baseline amplitude of the entire curve is globally controlled by the micro-slope cell subdivision threshold parameter. The mechanism of adaptively adjusting the mesh density based on the local geometric features (curvature) of the terrain is elucidated, which ensures that the generated micro-sloping unit mesh has higher geometric resolution in complex terrain features, while maintaining a reasonable computational scale in flat areas, thus providing an accurate and efficient discrete geometric basis for subsequent light-receiving performance calibration.

[0091] S3. Based on the micro-sloping unit grid and the solar apparent position time series data, the light-receiving efficiency of each micro-sloping unit is calibrated to obtain the light-receiving efficiency calibration results of each micro-sloping unit;

[0092] In its specific implementation, this step includes the following sub-steps:

[0093] S31. Based on each cell in the micro-sloping cell grid and the solar apparent position time series data, perform spatiotemporal integration processing to obtain the normal direct radiation component of the cell, and the normal direct radiation component is used as the radiation receiving reference value; wherein, the spatiotemporal integration processing is to integrate the product of the cosine value of the angle between the normal direction of the cell surface and the solar ray direction and the theoretical value of the normal solar irradiance outside the atmosphere with time over the continuous time domain of the solar apparent trajectory.

[0094] In this embodiment, the spatiotemporal integration processing involves integrating the product of the cosine of the angle between the surface normal direction and the solar ray direction of the unit cell and the theoretical value of the external atmospheric normal solar irradiance over a continuous time domain along the apparent solar trajectory. This integration includes solar ray direction vector construction and radiant flux time integration. The solar ray direction vector construction is based on the solar altitude and azimuth sequence in the apparent solar position time series data, calculating the unit direction vector of the solar rays in three-dimensional space at each moment. The radiant flux time integration is performed by taking the dot product of the surface unit normal vector and the solar ray direction vector at each moment for each cell in the micro-sloping cell grid to obtain a cosine value. This cosine value is then multiplied by the theoretical value of the external atmospheric normal solar irradiance at the corresponding moment, and numerically integrated over a set annual time domain. The integration result is the normal direct radiation component of that cell.

[0095] In this embodiment, the solar ray direction vector construction process transforms the solar apparent position time series data into a vector form that can be used for geometric calculations. Specifically, the solar apparent position time series data provides a series of discrete time points. (For example, the solar altitude angle corresponding to the hour on the hour every day of the year) and solar azimuth The solar azimuth angle North is defined as 0 degrees, and clockwise is positive. For any given time... The unit vector of the direction of solar rays The Cartesian components in the local horizontal coordinate system (east-north-zenith direction) are calculated as follows: It should be noted that the solar altitude angle mentioned... and azimuth The calculation is based on a standard astronomical formula, the input of which includes geographical latitude, date and time. The calculation method is well known to those skilled in the art and will not be elaborated on here.

[0096] In this embodiment, the radiation flux time integration processing is applied to the micro-sloping element mesh. Each triangular facet unit in Execution. The unit. It has a unit normal vector The normal vector is determined after the mesh is generated in step S22, and its direction is usually conventionally upward, away from the terrain surface. For each time point... ( Solar apparent position time series data provides information in the evaluation time domain. Calculate the solar ray direction vector (a series of discrete moments within the time frame). With element normal vector The dot product of these two values ​​yields the cosine of the angle of incidence. .when At this time, it indicates that the sun is below the horizon of the unit or its rays enter from the back of the unit, and the direct radiation contribution at this moment is zero. The theoretical value of the external atmospheric normal solar irradiance... As the Earth-Sun distance changes, according to the formula Given, among which, The solar constant, for The date ordinal number within the year in which the moment occurs. The unit. normal direct radiation component The discrete integral formula is obtained by summing and integrating the contributions at all valid time points: ,in For The time interval centered on the summation covers a preset annual evaluation period. It should be noted that this summation is a continuous integral. The numerical approximation is given by the time step. When sufficiently small, the discrete time point converges to a continuous integral value. Generation and time step The determination is based on the output of the solar apparent position time series data and the selected numerical integration scheme, which is a method of implementation well known to those skilled in the art.

[0097] Furthermore, the normal direct radiation component This is represented as a scalar field, defined over all triangular facet elements of the micro-sloping element mesh. Specifically, for each element in the mesh... There exists a unique scalar value Correspondingly, this value quantifies the total direct solar radiation energy received by the unit throughout the year under ideal conditions, ignoring atmospheric attenuation and surrounding obstruction. The set of scalar values... This forms the basis for the radiation receiving reference value data used in the subsequent step S32 for performance calibration. Preferably, the time integration can employ a more refined numerical integration method (such as the adaptive Simpson's rule) to improve accuracy. The selection and implementation of the numerical integration method are well-known to those skilled in the art. It should be noted that the calculation process assumes that the micro-sloping surface cell mesh has been aligned with the spatiotemporal reference frame where the apparent solar position data is located through necessary coordinate transformations. This alignment process is well-known to those skilled in the art.

[0098] S32. Based on the normal direct radiation component, a performance calibration process is performed using a preset energy flux density threshold to generate the unit's light-receiving performance calibration; wherein, the performance calibration process includes comparing the normal direct radiation component of each unit with a preset threshold range, classifying the units into high-performance units, variable-performance units, or low-performance units according to the comparison results, and assigning a unique calibration identifier to each category.

[0099] In this embodiment, the performance calibration process includes comparing the normal direct radiation component of each unit with a preset threshold range. Based on the comparison results, the units are classified into high-performance units, variable-performance units, or low-performance units, and a unique calibration identifier is assigned to each category. This includes threshold interval division processing and unit classification and identifier assignment processing. The threshold interval division processing is based on a preset energy flux density threshold, defining two ordered threshold values ​​on the real number axis to divide the radiation value range into three continuous, non-overlapping intervals. The unit classification and identifier assignment processing involves determining the corresponding performance category of each unit based on the specific threshold interval into which its normal direct radiation component, calculated in step S301, falls, and assigning the unit a unique calibration identifier corresponding to the category.

[0100] In this embodiment, the threshold interval division process is based on photovoltaic engineering experience and energy return targets, setting two ordered threshold parameters: a high threshold and a low threshold. and low threshold And satisfy Specifically, these two threshold parameters will affect the possible normal direct radiation component values. (The symbol definition is consistent with step S31) The value space is divided into three intervals: high-efficiency interval Variable performance range and low-efficiency range The threshold parameter and The specific values ​​are determined based on historical radiation data of the climate zone where the site is located, photovoltaic module performance, and economic benefit analysis models. This determination method is a routine design step that is well known to those skilled in the art based on project requirements, and will not be elaborated on further here.

[0101] Furthermore, when determining the threshold parameter, the statistical distribution characteristics of the normal direct radiation component can be used as an auxiliary decision-making factor. Specifically, the cumulative distribution function of the normal direct radiation component of all units can be calculated. This function describes the proportion of units with radiation values ​​less than or equal to any given radiation value in the total unit set. The cumulative distribution function curve monotonically increases from 0% to 100%, and its shape reflects the concentration and dispersion of the unit radiation potential distribution. Based on the cumulative distribution function curve, the corresponding threshold value can be derived by setting a desired unit category ratio (e.g., determining that high-efficiency units should account for the top 20% of the total number of units), thereby achieving quantitative setting of the threshold parameter based on statistical objectives. The calculation and analysis of the cumulative distribution function provides a data-based decision-making basis for the threshold interval division process. The threshold interval division process not only relies on a preset fixed threshold parameter but can also be dynamically adapted based on the statistical distribution characteristics of the normal direct radiation component. Specifically, the statistical distribution characteristics are described by two complementary representations: first, the probability density distribution of the normal direct radiation component, which visually presents the clustering of unit numbers within different radiation value ranges in the form of a histogram, demonstrating the spatial heterogeneity of radiation potential; second, the cumulative distribution function obtained based on the integral of the probability density distribution, which is strictly monotonically increasing, and its function value represents the proportion of units with radiation values ​​not exceeding a certain given value among all units. In practical engineering decisions, the expected proportion of each type of unit can be set according to the design objectives (e.g., high-efficiency units account for the top 25% of the total). By finding the radiation value mapped to the corresponding proportion on the cumulative distribution function curve, that radiation value can be determined as the high threshold. Similarly, the low threshold is determined based on the expected proportion. Thus, the quantitative conversion from the proportion target to the threshold value gives the threshold division the dual attributes of engineering experience guidance and data statistical support.

[0102] In this embodiment, the cell classification and identifier assignment process traverses all triangular facet cells in the micro-sloping surface cell mesh M. For each unit Obtain the normal direct radiation component calculated in step S31. The classification rule is implemented through a segmented discriminant function. Then the unit Classified as a high-efficiency unit; if Then the unit Classified as a variable performance unit; if Then the unit The units are categorized into low-efficiency unit classes. After classification, each category is assigned a unique, discrete identifier. Preferably, the identifier uses integer encoding; for example, the identifier for high-efficiency unit classes is defined as 2, the identifier for variable-efficiency unit classes is defined as 1, and the identifier for low-efficiency unit classes is defined as 0. Furthermore, the identifier is used as the basis for logical judgment in subsequent steps S41 and S42 to filter high-efficiency unit classes for constructing the main skeleton unit set and to filter variable-efficiency unit classes for gap filling.

[0103] In this embodiment, the light-receiving performance calibration of the unit is represented as a discrete scalar field function defined on all units of the micro-sloping unit mesh. Specifically, suppose there exists a mapping This mapping will transform each triangular facet unit. Mapped to an integer calibrator ,Right now According to the above classification rules, The set of values ​​is The mapping The expression for can be fully defined by the aforementioned piecewise discrimination rule. The discrete scalar field This is a complete representation of the unit's light-receiving performance, encapsulating the classification information of each unit based on its radiative receiving potential, providing a data structure foundation for the conversion from continuous radiative energy values ​​to discrete layout decisions. It should be noted that the classification and label allocation process is deterministic and non-iterative; its output directly drives the serialized layout synthesis logic in subsequent step S4. Please refer to the appendix. Figure 4 As shown, attached Figure 4The figure displays the function curves of the light-receiving efficiency distribution of micro-sloping units versus threshold calibration. The horizontal axis represents the numerical range of the normal direct radiation component, derived from the radiation receiving reference value obtained by performing spatiotemporal integration on each micro-sloping unit in step S31. The left vertical axis represents the number of units, and the right vertical axis represents the cumulative distribution percentage. The blue curve is a histogram of the normal direct radiation component distribution of micro-sloping units, showing the distribution of the number of micro-sloping units within different radiation value ranges. The shape of the blue curve reflects the statistical distribution characteristics of the unit radiation receiving potential across the entire terrain. The purple curve is the cumulative distribution function curve, representing the proportion of micro-sloping units with radiation values ​​less than or equal to any normal direct radiation component value on the horizontal axis in the total unit set. This curve monotonically increases from 0% to 100%. The two red vertical dashed lines in the figure indicate the preset low and high thresholds, respectively. The two threshold parameters are set based on photovoltaic engineering experience and energy return targets, dividing the space of the normal direct radiation component on the horizontal axis into three continuous and non-overlapping intervals. Specifically, the area to the left of the low threshold is the low-efficiency interval, the area between the low and high thresholds is the variable-efficiency interval, and the area to the right of the high threshold is the high-efficiency interval. These three intervals are illustrated in the figure with red, orange, and green background areas, respectively. The threshold interval division directly corresponds to the classification rules in the unit classification and label allocation process. For any micro-sloping unit, its normal direct radiation component value is compared with the threshold interval: if its value is in the low-efficiency interval, the unit is classified as a low-efficiency unit; if its value is in the variable-efficiency interval, the unit is classified as a variable-efficiency unit; if its value is in the high-efficiency interval, the unit is classified as a high-efficiency unit. The area ratio of the blue curve under different threshold intervals in the figure intuitively reflects the proportion of units in each category, while the vertical axis value of the purple curve at the threshold point gives the cumulative percentage of units in the corresponding category.

[0104] S4. Based on the light-receiving efficiency calibration results and the photovoltaic unit reference geometric data, sequential layout synthesis is performed according to efficiency priority to generate a synthesized topology layout;

[0105] In its specific implementation, this step includes the following sub-steps:

[0106] S41. Construct a set of main skeleton units based on the high-efficiency unit class in the unit light-receiving efficiency calibration and the photovoltaic unit reference geometric data;

[0107] The process of constructing the main skeleton unit set involves selecting a unit from the high-efficiency unit class as the starting unit, and iteratively selecting and connecting neighboring high-efficiency units that meet the repulsion relationship based on the spatial repulsion relationship between the component envelope shapes defined by the photovoltaic unit reference geometric data, until no unit that meets the condition can be connected.

[0108] In this embodiment, constructing the main skeleton unit set involves selecting a unit from the high-efficiency unit class as the starting unit. Based on the spatial repulsion relationship between the component envelope shapes defined by the photovoltaic unit reference geometry data, neighboring high-efficiency units that meet the repulsion relationship are iteratively selected and connected until no unit that meets the condition can be connected. This includes unit envelope space definition processing and unit iterative selection processing based on spatial repulsion. The unit envelope space definition processing calculates a two-dimensional envelope region associated with the plane of each micro-sloping unit based on the component size and shape defined in the photovoltaic unit reference geometry data. The envelope region represents the projected space range occupied by the unit if it is selected as the installation location of the photovoltaic module. The unit iterative selection processing based on spatial repulsion uses the spatial repulsion relationship as a constraint condition. From the high-efficiency units that meet the constraint, units are iteratively selected and added to an ordered set, which constitutes the main skeleton unit set.

[0109] In this embodiment, the element envelope space definition processing is applied to any triangular patch element in the micro-sloping element mesh. Specifically, let the reference geometry of the photovoltaic unit define the length L and width W of a standard photovoltaic module. For a given unit... The local plane in which the element resides is determined. The element envelope space is defined as a rectangle on this local plane, the center of which is perpendicular to the element. centroid The projections on this plane coincide. The long side of the rectangle is aligned with a pre-designed reference installation direction within the unit plane, which typically points due south or is determined based on the overall site orientation. This rectangular area is represented as a set of points on the plane. ,in It is a unit The plane in which it is located This indicates that the vector is projected onto a local plane coordinate system spanned by the reference mounting direction and its perpendicular direction. and Let represent the absolute distances along the long and short sides in this local coordinate system, respectively. Representing a plane Any point on the bounding rectangle. The calculation is based on the centroid of the unit. A local rectangular coordinate system is established within the unit plane, with the positive X-axis pointing towards the preset installation direction. Based on the component length L and width W, the coordinates of the four vertices of the envelope rectangle in this local coordinate system are calculated. Through coordinate transformation, the local coordinates of these four vertices are converted back to the global three-dimensional coordinate system, thereby obtaining the precise position and orientation of the envelope rectangle in three-dimensional space. The calculation and determination of the envelope rectangle are methods well-known to those skilled in the art.

[0110] In this embodiment, the spatial repulsion-based iterative selection process is based on the cell light-receiving performance calibration generated in step S32. This calibration will assign each cell... Mapped to a calibration identifier The process starts from all those that satisfy... From the units of the high-efficiency unit class, select one unit as the starting unit. And add it to an initially empty set of main skeleton units. The selection rules for the starting units aim to ensure that the main frame layout has a good foundation for expansion. Preferred rules include, but are not limited to: selecting the unit closest to the geometric center of the high-efficiency unit class; or selecting the unit with the most neighboring high-efficiency units in the high-efficiency unit class; or selecting the unit with the highest score after weighting the normal direct radiation component value and the positional centrality index. The spatial repulsion relationship is defined by the spacing rules in the array layout constraint data, specifically: for any two different units... and If their corresponding envelope rectangles and The minimum Euclidean distance between projections on the horizontal plane Less than the preset minimum spacing value If these two units violate the spatial exclusion relationship, they cannot coexist in the main skeleton unit set. The distance... The calculation is a routine operation in computational geometry.

[0111] Furthermore, the iterative selection process defines a set of frontier units. Initially, it only contained In each iteration, from the frontier set... Take out a unit Search for all neighboring high-performance cells (i.e., calibration markers) in the micro-sloping cell mesh. (The units). For each candidate unit found. Examine its envelope rectangle With the main skeleton unit set Each existing unit Envelope rectangle The minimum distance between them. If for all All satisfied Then Add to main skeleton unit set And simultaneously add the frontier set This allows for further expansion based on it in subsequent iterations. When the frontier set... When the value is empty, the iteration terminates, indicating that under the constraint of satisfying the spatial repulsion relation, it is no longer possible to find new connectable high-efficiency units in the neighborhood of the current set.

[0112] In this embodiment, the set of main skeleton units is represented as an ordered set. This set is a subset of the set of micro-sloping units F, which satisfies the following two core conditions: First, for any Its calibration marks meet Second, for any two distinct units in the set... , The corresponding envelope rectangle satisfies the spatial exclusion constraint. The set The generation process ensures that the main body of the photovoltaic array is preferentially arranged in the area with the highest radiation receiving potential, while strictly adhering to the minimum spacing requirements between components, defining a structural boundary framework for subsequent gap filling processing. It should be noted that the iterative search for neighboring units is achieved by querying the topological connectivity of the micro-sloping unit mesh, based on defining spatially neighboring units as units sharing edges or vertices in the triangular mesh; the calculation of the envelope rectangle distance projects the three-dimensional spatial rectangle envelope corresponding to each unit onto the horizontal plane, characterizing the spatial repulsion relationship by calculating the minimum Euclidean distance between two two-dimensional projected polygons, where the minimum distance is the lower bound of the distance between any pair of points on the two polygons; the implementation logic of the specific starting unit selection rule is to construct an objective function, which combines the normal direct radiation component with the spatial centrality index of the unit in the high-efficiency unit class (e.g., the reciprocal of the distance from the unit to the geometric center of the same type of unit, or the number of neighboring high-efficiency units of the unit) through linear or nonlinear weighted summation, selecting the unit that optimizes the objective function value as the starting unit. This process provides an initial position that balances radiation performance and expansion potential for layout expansion. The processing method and principle described herein are conventional computational geometry, spatial analysis, and decision-making techniques well-known to those skilled in the art, and will not be elaborated upon further here.

[0113] S42. Based on the main skeleton unit set, the variable efficiency unit class in the unit light-receiving efficiency calibration, and the photovoltaic unit reference geometric data, a synthetic topology layout is generated by filling the gap regions between the main skeleton unit sets;

[0114] The process of generating the synthetic topology layout involves identifying a continuous unoccupied area defined by adjacent unit pairs in the main skeleton unit set, selecting units from the variable performance unit class whose component envelope shape is located within this area and does not overlap with the component envelope shape of any main skeleton unit, and adding the selected units to the fill unit set in descending order of light reception performance.

[0115] The synthetic topology layout is formed by merging the main skeleton unit set and the filling unit set.

[0116] In this embodiment, generating the synthetic topology layout involves identifying continuous unoccupied regions defined by adjacent units in the main skeleton unit set, selecting units from the variable performance unit class whose component envelope shapes lie within these regions and do not overlap with the component envelope shapes of any main skeleton unit, and adding these selected units to the filling unit set in descending order of light reception efficiency. This includes gap region identification processing and variable unit selection and filling processing. The gap region identification processing is based on the spatial proximity relationship between units in the main skeleton unit set, calculating the gaps between envelope rectangles and merging them to form continuous unoccupied regions. The variable unit selection and filling processing is based on the geometric boundaries of these continuous unoccupied regions, selecting units from the variable performance unit class whose envelope rectangles completely lie within these regions and do not overlap with the envelope rectangles of any main skeleton unit, and then sorting and iteratively adding them in descending order based on their normal direct radiation component values ​​to construct the filling unit set.

[0117] In this embodiment, the gap region identification process uses the set of main skeleton units and their corresponding set of envelope rectangles. For input, specifically, the process first determines the adjacency relationship between the main skeleton units. The adjacency determination is based on the minimum Euclidean distance between the projections of the envelope rectangles onto the horizontal plane; a neighborhood determination threshold is set. If two units The envelope rectangle satisfies If the threshold value is found to be adjacent, then the two units are determined to be adjacent. The value is related to the module size and minimum spacing in the photovoltaic unit reference geometry data. Related, the preferred value is Where k is a constant greater than 1. For each pair of adjacent units, the gap region defined by their envelope rectangles is calculated. The calculation logic for the gap region is as follows: an outer buffer zone is constructed for each envelope rectangle, the distance of which is the minimum spacing value. Then, the intersection of the two buffer zones is taken, and the portion of this intersection region after removing the two original envelope rectangles is the gap region. The continuous unoccupied regions are formed by merging the corresponding gap regions of all adjacent units through a geometric union operation, and then subtracting the union portion of the envelope rectangles of all main skeleton units, forming a set of unconnected two-dimensional polygonal regions. The area Represented as a set of points on a plane, satisfying It is a non-empty set. It should be noted that the geometric algorithms for determining adjacency, calculating gap regions, and merging regions are based on spatial indexing, geometric buffer construction, and polygon Boolean operations, and these implementation methods are well known to those skilled in the art.

[0118] In this embodiment, the variable cell filtering and filling process traverses all calibration identifiers in the cell light-receiving performance calibration defined in step S32. Variable performance unit For each such cell, obtain its envelope rectangle. and the normal direct radiation component The process performs two geometric condition checks: first, a position containment check to verify the envelope rectangle. Is it affected by a certain gap area? Complete inclusion, that is Second, non-overlapping checks are performed to verify the envelope rectangle. Envelope rectangle of all main skeleton units (in There is no overlap between them, that is The non-empty set is for all Established. Variable performance units that simultaneously satisfy both of these conditions are included in the candidate unit set. Furthermore, regarding the candidate unit set... The element in the middle is radiated by the normal direct radiation component. Sort in descending order to get an ordered list. ,in Initialize the fill cell set. Non-empty set. Each candidate unit is processed iteratively according to the order of the ordered list. In each iteration, check the current cell. Envelope rectangle Is it still completely contained within a region in the currently updated gap region G and not joined with any region? The envelope rectangles of any unit overlap. If the condition is satisfied, then... Add to fill cell set And subtract from the gap region G The area occupied, i.e., the update For the corresponding ( A continuous planar region not occupied by any main skeleton unit enclosing rectangle. The iteration continues until the ordered list has been traversed.

[0119] In this embodiment, the synthetic topology layout is represented as a discrete set L, which is the union of the main skeleton unit set and the filling unit set. This set satisfies the following geometric constraints: for any two distinct units... The projection of its bounding rectangle onto the horizontal plane satisfies the minimum spacing constraint. Furthermore, all filling units The calibration mark is And there exists an initial gap region. Make Established. The synthesized topology layout L constitutes a complete topological distribution of the photovoltaic array on the micro-sloping unit grid. This distribution ensures the preferential utilization of high-efficiency regions and the complementary filling of gap spaces, providing an initial configuration for the layout compliance simulation and adjustment in step S5. It should be noted that the geometry includes checks, rectangle intersection tests, and region update operations, which are conventional computational geometry techniques well known to those skilled in the art.

[0120] S5. Based on the principle of geometric projection, perform occlusion analysis on the synthesized topology layout, and combine slope constraints and boundary constraints for iterative adjustment to obtain the corrected synthesized topology layout;

[0121] In its specific implementation, this step includes the following sub-steps:

[0122] S51. Based on the synthetic topology layout and the spacing rules defined in the array layout constraint data, a geometric projection occlusion analysis is performed. The process calculates and superimposes the projection areas of each unit in the typical solar altitude angle domain to determine the occlusion time ratio of each component unit and generate a set of unit-level occlusion coefficients.

[0123] In this embodiment, the processing calculates and superimposes the projection areas of each unit within a typical solar altitude angle domain to determine the shading time proportion of each component unit. This includes time-series projection area calculation processing and time-integral shading determination processing. The time-series projection area calculation processing calculates the projection area of ​​the envelope rectangle on the ground plane for each photovoltaic module unit in the composite topology layout, based on its position and normal on the micro-sloping surface grid, under a selected typical solar altitude angle sequence. The time-integral shading determination processing, based on the spatial superposition relationship of the projection areas, calculates the proportion of time during which its projection area is covered by the projections of other units in the solar runtime sequence for each unit; this proportion is the shading coefficient of that unit.

[0124] In this embodiment, the temporal projection region calculation process uses the synthesized topology layout L generated in step S42 as input. Each cell in the layout With a known envelope rectangle (Defined in step S41) and unit normal vector (Defined in step S22). Let the selected typical solar altitude angle sequence for analysis be... This sequence covers the effective solar altitude angle range from sunrise to sunset, with each altitude angle... Corresponding to one or more solar azimuth angles Together, these constitute a time series of the apparent position of the sun. For a given unit... and the position of the sun at a certain moment Solar ray direction vector The calculation follows the definition in step S31. The element envelope rectangle. Projection area on the ground plane (i.e., horizontal plane) It is obtained through parallel projection calculation. Specifically, it involves calculating the four corner points of the enclosing rectangle. In the opposite direction of the solar rays Projected onto the horizontal plane The coordinates of the projection point are calculated as follows: ,in Corner point elevation, The direction vector of solar rays The zenith component (Z component). Consists of four projection corner points. The enclosed quadrilateral (which may be a convex quadrilateral or a degenerate form) is the projection region at that moment. The calculation of the projection area falls under the category of perspective geometry and coordinate transformation applications well known to those skilled in the art.

[0125] In this embodiment, the time-integrated occlusion determination process is based on the projection area. Perform the calculation. For any two elements in the composite topology layout. and ( ), in the position of the sun Next, the judgment unit projection area Is it for the unit? projection area This causes occlusion. The geometric condition for determining occlusion is: element Any point within the projection area, its relationship with the element Does a unit exist along the spatial line connecting itself (i.e., in the opposite direction of the solar ray direction)? The solid part. In the parallel projection model, this determination simplifies to checking the projection area. and Does there exist an intersection, and are the units... Relative to unit Closer in the direction of the solar rays. Distance comparison is performed by comparing the coordinates of the unit centroid in the direction of the solar rays. A binary indicator function is defined. :

[0126] ;in and Units and The centroid coordinates. When "Time" indicates the moment. unit Unit Occlusion. Furthermore, the unit... At any moment The overall occlusion state is determined by the function express: The function takes a value of 1 to indicate that At this moment, it is obscured by at least one other unit; a value of 0 indicates that it is not obscured. The typical solar altitude angle sequence... Together with the corresponding azimuth angle changes, they discretize to represent the apparent solar motion trajectory throughout the year. (Unit) The proportion of the annual cumulative time that is obscured, i.e., its occlusion coefficient. Through the The weighted integral over the time domain yields: ,in The weighting factor is used to reflect the difference in the length of the time period corresponding to different solar positions and the contribution of solar irradiance to the total annual radiation. The determination of the weighting factor is a routine design carried out by those skilled in the art based on meteorological and solar energy engineering knowledge.

[0127] In this embodiment, the set of cell-level occlusion coefficients is represented as a scalar value mapping defined on the synthetic topology layout L. Specifically, for each unit There exists a unique real number. ,and The definition of this mapping Z is derived from the above. The integral formula completely describes the set. This is a set of unit-level shading coefficients, which quantifies the risk of effective sunshine time loss for each component unit in the initial layout due to shading by the array in front, providing a clear, data-based adjustment target for constraint satisfaction adjustment in step S52. It should be noted that the above projection calculation, polygon intersection judgment, and weighted integral operation are all conventional techniques in computational geometry and numerical analysis.

[0128] S52. Based on the set of unit-level occlusion coefficients and the slope and boundary constraints in the array layout constraint data, constraint satisfaction adjustment processing is performed. The processing identifies units whose occlusion coefficients exceed a preset threshold as units to be adjusted, performs position translation or axial rotation on the units to be adjusted based on slope constraints, and trims layout parts that exceed the permissible range based on boundary constraints, thereby generating a corrected topology layout that satisfies all given constraints.

[0129] In this embodiment, the process identifies units with occlusion coefficients exceeding a preset threshold as units to be adjusted. Based on slope constraints, these units are translated or rotated axially. Furthermore, layout portions exceeding permissible limits are trimmed based on boundary constraints, thereby generating a corrected topology layout that satisfies all given constraints. This process includes unit identification and initial correction, and boundary constraint trimming and final compliance verification. The unit identification and initial correction process involves filtering severely occluded units based on the set of unit-level occlusion coefficients and a preset occlusion coefficient threshold. Under the constraints of slope and spacing, the position or orientation of these units is locally adjusted to reduce the occlusion coefficient. The boundary constraint trimming and final compliance verification process, after completing the local adjustments, checks whether the envelope rectangle of all units in the entire layout exceeds the permissible site boundary. Any portions exceeding the boundary are trimmed or removed, and the adjusted layout is re-verified to ensure it meets all given constraints, thus outputting the final corrected topology layout.

[0130] In this embodiment, the identification and initial correction of the unit to be adjusted uses the set of unit-level occlusion coefficients generated in step S51 as input. An occlusion coefficient threshold is set. This threshold is determined based on photovoltaic system performance tolerance and engineering experience. For each unit in the synthetic topology layout L... If its shading coefficient Then mark the unit as the unit to be adjusted, forming a set of units to be adjusted. For each unit to be adjusted The process attempts to improve the occlusion situation through position translation or axial rotation. Position translation refers to moving the center of the envelope rectangle of the element along a specific direction by a finite distance within the local plane of the micro-sloping element. Axial rotation refers to changing the direction of the long side of the element's envelope rectangle, i.e., adjusting the element's installation azimuth angle so that the element's normal vector rotates around the zenith axis by a certain angle. The specific adjustment strategy is based on gradient descent or heuristic search methods: calculating the sensitivity of the cell occlusion coefficient to its position or orientation, and making small adjustments along the direction that reduces the occlusion coefficient. After each adjustment, it is necessary to immediately verify whether the adjusted cell still satisfies the slope and spacing constraints. The slope constraint requires that the angle between the normal vector of the cell's plane and the normal vector of the horizontal plane (i.e., the slope) does not exceed the preset maximum slope angle. The spacing constraint requires that the minimum distance between the envelope rectangle of the adjusted element and the envelope rectangles of all other elements in layout L still satisfy the condition. If any constraint is violated after adjustment, the adjustment is revoked and other directions or adjustment methods are tried. The adjustment process is repeated iteratively until a unit is reached. Occlusion coefficient Drop to threshold The maximum number of iterations may be reached. It should be noted that the specific implementation of the gradient descent or heuristic search is an optimization technique well-known to those skilled in the art.

[0131] In some implementations, boundary constraint pruning and final compliance verification are based on boundary constraints defined in the array layout constraint data. These boundary constraints are defined as two-dimensional polygonal regions. This indicates that the area represents the permissible projection of the site onto the horizontal plane. This applies to the layout after initial modifications. (i.e., the adjusted layout), iterate through each unit within it. Examine its envelope rectangle Does the projection on the horizontal plane lie entirely within the boundary polygon? Internally. Specifically, using a polygonal inclusion test algorithm, if the enclosing rectangle... Any corner point located at Besides, or and If a cell intersects with the boundary, it is determined that the cell has exceeded the boundary. There are two approaches to handling out-of-bounds cells: First, if the out-of-bounds portion is small, attempt a small translation of the cell to bring it completely within the boundary; second, if translation cannot satisfy the boundary constraints, remove the cell from the layout. Remove units from the layout. Removing units may create new isolated units or violate spacing rules, therefore the spacing of the remaining units in the layout needs to be recalculated, and it needs to be checked whether the minimum spacing is still satisfied. Constraints. If not satisfied, the affected cells are further repositioned or removed. This process is iterated until all remaining cells in the layout satisfy both boundary and spacing constraints. Finally, the final layout is determined. Repeat the geometric projection occlusion analysis in step S51, calculate the occlusion coefficient for each element, and ensure that the occlusion coefficients of all elements are equal. If any unit still does not meet the requirements, repeat the above adjustment and pruning process until convergence is achieved or the preset maximum number of iterations is reached.

[0132] In this embodiment, the modified topology layout is represented as a discrete set. This set is a subset or modified set obtained after the original synthetic topology layout L has undergone the above series of constraint adjustment processes, i.e. (This may involve element replacement, i.e., changes in cell position or orientation). This set satisfies the following four engineering constraints: First, for any cell... The slope of the micro-sloping unit it is located in does not exceed Second, for any two distinct units The projection of its bounding rectangle onto the horizontal plane satisfies the minimum spacing constraint. (The minimum Euclidean distance between these two two-dimensional projection areas must be greater than or equal to the preset minimum spacing value.) and Each refers to a unit With unit The envelope rectangle defined in step S41, specifically, for the synthesized topology layout Any two distinct units in and , Representation unit The two-dimensional geometric region formed by the projection of the corresponding envelope rectangle onto the horizontal plane. Representation unit The corresponding two-dimensional geometric region formed by the projection of the envelope rectangle onto the horizontal plane. Third, for any element... The projection of its envelope rectangle onto the horizontal plane lies entirely within the boundary polygon of the site. Internal; Fourth, for any unit Its shading coefficient The modified topology layout This is the final photovoltaic array topology generated by this method, which achieves comprehensive optimization of lighting efficiency, engineering safety, and site constraints on complex terrain. It should be noted that the specific implementation of the polygon inclusion test, iterative adjustment, and verification process falls under computer-aided design and optimization techniques well-known to those skilled in the art.

[0133] Based on the description of the above embodiments of the photovoltaic array topology generation method based on digital terrain analysis, this embodiment also discloses a photovoltaic array topology generation system based on digital terrain analysis. The photovoltaic array topology generation system based on digital terrain analysis can be a computer program that runs the aforementioned photovoltaic array topology generation method based on digital terrain analysis. Please see the appendix. Figure 2As shown, the photovoltaic array topology generation system based on digital terrain analysis provided in this embodiment includes:

[0134] The acquisition unit is used to acquire digital terrain elevation data, solar apparent position time series data, photovoltaic unit reference geometric data, micro-slope unit subdivision threshold parameters, and array layout constraint data;

[0135] A meshing processing unit is used to perform curvature-guided meshing processing on digital terrain elevation data based on the micro-slope unit subdivision threshold parameters to generate a micro-slope unit mesh.

[0136] The efficiency calibration unit is used to calibrate the light-receiving efficiency of each micro-sloping unit based on the micro-sloping unit grid and the solar apparent position time series data, and obtain the light-receiving efficiency calibration results of each micro-sloping unit.

[0137] Topology synthesis unit is used to generate a synthesized topology layout by sequentially arranging and synthesizing the photovoltaic unit based on the light-receiving efficiency calibration results and the photovoltaic unit reference geometric data according to the efficiency priority.

[0138] The compliance adjustment unit is used to perform occlusion analysis on the synthetic topology layout based on the principle of geometric projection, and to perform iterative adjustment in combination with slope constraints and boundary constraints to obtain the corrected synthetic topology layout.

[0139] Since the functions of each functional unit in this system correspond to the steps in the photovoltaic array topology generation method based on digital terrain analysis, the specific implementation methods of each step in the method are also applicable to each functional unit in this system. Therefore, the specific implementation methods of each functional unit in this system will not be elaborated here.

[0140] Although embodiments of the present invention have been described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present invention, and all such changes and alterations shall not depart from the protection scope of the present invention.

Claims

1. A photovoltaic array topology generation method based on digital terrain analysis, characterized in that, Includes the following steps: S1. Acquire digital terrain elevation data, solar apparent position time series data, photovoltaic unit reference geometric data, micro-slope unit subdivision threshold parameters, and array layout constraint data; S2. Based on the micro-sloping element subdivision threshold parameter, the digital terrain elevation data is subjected to curvature-guided meshing to generate a micro-sloping element mesh: S21. Perform Gaussian curvature field calculation on the digital terrain elevation data to determine the principal curvature of each sampling point on the terrain surface, and obtain the Gaussian curvature value of each sampling point through differential geometric operations, thereby obtaining the terrain curvature distribution field; S22. Based on the terrain curvature distribution field, control the grid density and use the micro-sloping element subdivision threshold parameter as the grid subdivision stopping criterion. Use the leading edge propagation method to generate anisotropic triangular meshes to obtain micro-sloping element meshes composed of triangular facets. S3. Based on the micro-sloping surface unit grid and the solar apparent position time series data, the light-receiving efficiency of each micro-sloping surface unit is calibrated to obtain the light-receiving efficiency calibration results for each micro-sloping surface unit: S31. For each cell in the micro-sloping cell grid, perform spatiotemporal integration processing in conjunction with the solar apparent position time series data to obtain the normal direct radiation component of the cell, and use it as the radiation receiving reference value. S32. Based on the normal direct radiation component and the preset energy flux density threshold, the performance is calibrated to obtain the light-receiving performance calibration results for each micro-sloping unit; S4. Based on the light-receiving efficiency calibration results and the photovoltaic unit reference geometric data, sequential layout synthesis is performed according to efficiency priority to generate a synthesized topology layout: S41. Construct a set of main skeleton units based on the high-efficiency unit class in the unit light-receiving efficiency calibration and the photovoltaic unit reference geometric data; S42. Based on the main skeleton unit set, the variable efficiency unit class in the unit light-receiving efficiency calibration, and the photovoltaic unit reference geometric data, a synthetic topology layout is generated by filling the gap regions between the main skeleton unit sets; S5. Based on the principle of geometric projection, perform occlusion analysis on the synthesized topology layout, and combine slope constraints and boundary constraints for iterative adjustment to obtain the corrected synthesized topology layout.

2. The photovoltaic array topology generation method based on digital terrain analysis as described in claim 1, characterized in that, In step S1, the digital terrain elevation data is a discrete or continuous dataset describing the elevation distribution of the terrain surface; the solar apparent position time series data is a time series describing the changes in solar azimuth and elevation angles throughout the year; the photovoltaic unit reference geometric data is a set of geometric parameters describing the size and shape of standard photovoltaic modules; the micro-slope unit subdivision threshold parameter is a scalar value controlling the subdivision accuracy of the terrain surface; and the array layout constraint data is a set of rules describing the slope, spacing, and boundary constraints in the actual site.

3. The photovoltaic array topology generation method based on digital terrain analysis as described in claim 1, characterized in that, In step S31, the spatiotemporal integration process is to integrate the product of the cosine of the angle between the unit surface normal and the solar rays and the theoretical value of the normal solar irradiance outside the atmosphere over the continuous time domain of the solar apparent trajectory. In step S32, the performance calibration includes: comparing the normal direct radiation component of each unit with a preset threshold range, classifying the units into high-performance unit type, variable-performance unit type or low-performance unit type according to the comparison results, and assigning a unique calibration identifier to each category.

4. The photovoltaic array topology generation method based on digital terrain analysis as described in claim 1, characterized in that, In step S41, the construction of the main skeleton unit set involves selecting a unit from the high-efficiency unit class as the starting unit, and iteratively selecting and connecting neighboring high-efficiency units that meet the repulsion relationship based on the spatial repulsion relationship between the component envelope shapes defined by the photovoltaic unit reference geometric data, until no unit that meets the condition can be connected. In step S42, the generation of the synthetic topology layout is achieved by identifying the continuous unoccupied area defined by adjacent unit pairs in the main skeleton unit set, selecting units from the variable performance unit class whose component envelope shape is located in the area and does not overlap with the component envelope shape of any main skeleton unit, and adding the selected units to the filling unit set in descending order of light reception performance. The synthetic topology layout is formed by merging the main skeleton unit set and the filling unit set.

5. The photovoltaic array topology generation method based on digital terrain analysis as described in claim 1, characterized in that, In step S5, occlusion analysis is performed on the synthesized topology layout based on the principle of geometric projection, and iterative adjustments are made in conjunction with slope constraints and boundary constraints to obtain the corrected synthesized topology layout, including: S51. Based on the synthetic topology layout and the spacing rules defined in the array layout constraint data, perform geometric projection occlusion analysis to generate a set of cell-level occlusion coefficients. S52. Based on the set of unit-level occlusion coefficients and the slope and boundary constraints in the array layout constraint data, perform constraint satisfaction adjustment processing to obtain the corrected synthetic topology layout.

6. The photovoltaic array topology generation method based on digital terrain analysis as described in claim 5, characterized in that, In step S51, the geometric projection occlusion analysis process calculates and superimposes the projection areas of each unit in the typical solar altitude angle domain to determine the occlusion time ratio of each component unit and generate a set of unit-level occlusion coefficients. In step S52, the constraint satisfaction adjustment process involves identifying units whose occlusion coefficient exceeds a preset threshold as units to be adjusted, performing position translation or axial rotation on the units to be adjusted based on slope constraints, and trimming layout portions that exceed the permissible range based on boundary constraints, thereby generating a corrected topology layout that satisfies all given constraints.

7. A photovoltaic array topology generation system based on digital terrain analysis, used to implement the photovoltaic array topology generation method based on digital terrain analysis as described in any one of claims 1 to 6, characterized in that, The system includes: The acquisition unit is used to acquire digital terrain elevation data, solar apparent position time series data, photovoltaic unit reference geometric data, micro-slope unit subdivision threshold parameters, and array layout constraint data; The meshing unit is used to perform curvature-guided meshing on digital terrain elevation data based on the micro-slope element subdivision threshold parameter to generate a micro-slope element mesh: Gaussian curvature field calculation is performed on the digital terrain elevation data to determine the principal curvature of each sampling point on the terrain surface, and the Gaussian curvature value of each sampling point is obtained through differential geometric operations, thereby acquiring the terrain curvature distribution field; the mesh density is controlled based on the terrain curvature distribution field, and the micro-slope element subdivision threshold parameter is used as the mesh subdivision stopping criterion. Anisotropic triangular mesh generation is performed using the leading-edge propagation method to obtain a micro-slope element mesh composed of triangular facets; The efficiency calibration unit is used to calibrate the light-receiving efficiency of each micro-sloping unit based on the micro-sloping unit grid and the solar apparent position time series data, and obtain the light-receiving efficiency calibration results for each micro-sloping unit: for each unit in the micro-sloping unit grid, spatiotemporal integration processing is performed in combination with the solar apparent position time series data to obtain the normal direct radiation component of the unit, and this is used as the radiation receiving reference value; efficiency calibration is performed based on the normal direct radiation component and a preset energy flux density threshold to obtain the light-receiving efficiency calibration results for each micro-sloping unit; A topology synthesis unit is used to perform sequential layout synthesis based on the light-receiving efficiency calibration results and the photovoltaic unit's reference geometric data, according to efficiency priority, to generate a synthesized topology layout: A main skeleton unit set is constructed based on the high-efficiency unit class in the unit's light-receiving efficiency calibration and the photovoltaic unit's reference geometric data; a synthesized topology layout is generated by filling the gaps between the main skeleton unit set based on the main skeleton unit set, the variable-efficiency unit class in the unit's light-receiving efficiency calibration, and the photovoltaic unit's reference geometric data. The compliance adjustment unit is used to perform occlusion analysis on the synthetic topology layout based on the principle of geometric projection, and to perform iterative adjustment in combination with slope constraints and boundary constraints to obtain the corrected synthetic topology layout.

Images