Photovoltaic system multi-scene design method, system, equipment and medium

Abstract

The invention relates to the technical field of photovoltaic system design and intelligent planning, and provides a photovoltaic system multi-scene design method, system, device and medium, and the method comprises the steps: inputting a project address to automatically analyze longitude and latitude, and associating local or third-party irradiation and meteorological data; selecting a photovoltaic system application scene, and generating a unique project ID; drawing a region boundary or a facade contour on the base map through interactive operation; clicking on any edge of the drawn boundary and inputting the actually measured length, and triggering equal-proportion scaling calibration of the whole image to realize global scale unification; entering a facade mode to generate a three-dimensional model; obstacles are placed or imported in an interactive mode; generating a shadow analysis grid in real time according to solar path geometric deduction; calculating annual energy output, internal return rate, net present value and investment payback period based on the regional irradiation data, the component efficiency and the arrangement density; and performing model rotation, zooming and screenshot through the three-dimensional interaction interface to generate a standardized report.

Description

Technical Field

[0001] This application relates to the field of photovoltaic system design and intelligent planning technology, and in particular to a photovoltaic system multi-scenario design method, system, equipment and medium. Background Technology

[0002] In the field of photovoltaic system design, existing technologies generally suffer from multiple technical bottlenecks: 1. Insufficient data calibration and modeling accuracy: When drawing area boundaries based on satellite imagery or bitmaps, traditional methods lack a unified scale calibration mechanism and rely on manual measurement, resulting in significant geometric errors. In particular, they are weak in the ability to model elevation parameters for multi-facade scenes, making it difficult to meet the 0.01-meter accuracy requirement. 2. Poor adaptability to complex scenarios: Mainstream tools only support simple layouts on regular roofs. They lack the ability to automatically optimize component layouts for irregular shapes, curved surfaces, or multi-facade collaborative scenarios. They cannot adaptively adjust component size and spacing based on facade slope and vertex height difference. Furthermore, their avoidance strategies for obstacles such as parapet walls, skylights, and protruding structures are crude, which can easily leave behind the risk of obstruction. 3. Lack of full-process linkage: The results of power generation calculation, economic indicator evaluation and real-time scheduling are disconnected from each other, relying on offline estimation and unable to dynamically couple irradiation data, shading effects and cost parameters based on three-dimensional models for real-time iteration; 4. Low visualization and delivery efficiency: The lack of a 3D interactive verification environment means that communication of solutions relies on 2D drawings, and technical parameters, economic reports and visualization results need to be manually integrated, making it difficult to meet the requirements for standardized delivery.

[0003] Therefore, a method is urgently needed to solve at least one of the above problems. Summary of the Invention

[0004] This application provides a multi-scenario design method, system, equipment, and medium for photovoltaic systems, aiming to solve the problem that there is no existing fully automated method that can deeply integrate regional intelligent calibration, multi-facade parametric modeling, automatic obstacle avoidance, real-time revenue calculation, and 3D visualization verification.

[0005] Firstly, this application provides a multi-scenario design method for photovoltaic systems, including: The system automatically resolves latitude and longitude by inputting the project address, and associates local or third-party irradiance and meteorological data, including horizontal global irradiance, wind speed and temperature; selects the photovoltaic system application scenario, which includes rooftop scenario and facade scenario; enters the project name and basic constraints; and generates a unique project ID. Draw the area boundary or facade outline on the base map through interactive operations. The base map includes satellite imagery or bitmap. Click on any side of the drawn boundary and enter the measured length to trigger the full map to scale and calibrate proportionally, achieving global scale uniformity. Enter the facade mode, enter the elevation point by point for multiple facades, and generate a three-dimensional model. Obstacles can be placed or imported interactively, including parapet walls, fans, stairwells, ventilation shafts, windows, and protruding structures. A shadow analysis mesh is generated in real-time based on geometric deduction of the solar path. For rooftop scenarios, component specifications, row and column spacing, tilt angle, and azimuth angle are input to initialize the layout with a regular array, and optimization calculations are performed based on irradiance and shading cost functions. For facade scenarios, each facade is optimized individually or collaboratively, automatically adapting component sizes and array spacing based on the facade slope and the difference in vertex height to avoid obstacles and achieve component layout. Annual power generation is calculated based on regional irradiance data, component efficiency, and layout density. It is linked to local electricity prices, subsidies, initial investment costs, and operation and maintenance costs to calculate internal rate of return, net present value, and investment payback period, supporting comparison of multiple schemes. The model can be rotated, scaled, and screenshotted through a 3D interactive interface to generate a standardized report containing 3D renderings, layout diagrams, technical parameter tables, and revenue curves.

[0006] In some embodiments, the irradiance and shading cost function considers shading effects, row spacing, boundary constraints, and inverter proximity principles to avoid obstacles and achieve component arrangement. The optimization calculation based on the irradiance and shading cost function includes: after initializing the component arrangement with a regular array, determining the shading area according to the shadow analysis grid, and establishing a cost function model that includes irradiance loss weight, row spacing constraint weight, boundary distance weight, and inverter access distance weight; calculating the cost function value of each arrangement scheme by iteratively adjusting the row and column positions of the components, and selecting the arrangement scheme with the smallest cost function value that satisfies the obstacle avoidance condition.

[0007] In some embodiments, the step of rotating, scaling, and taking screenshots of the model through a 3D interactive interface to generate a standardized report containing a 3D rendering, layout diagram, technical parameter table, and revenue curve includes: rendering the 3D model from multiple angles using a 3D rendering engine, automatically extracting component layout coordinate data to generate a 2D layout diagram, extracting annual power generation, internal rate of return, net present value, and investment payback period from the 2D layout diagram to generate a technical parameter table and revenue curve, and integrating them according to a preset template to form a standardized report.

[0008] In some embodiments, the step of generating a shadow analysis grid in real time based on the geometric deduction of the solar path includes: calculating the solar altitude angle and azimuth angle for different seasons based on the input latitude, longitude and time parameters, and generating the solar path trajectory; constructing a geometric model of a shadow cone along the solar path with the obstacle vertex as a reference, dividing the target area into grid cells of preset precision, determining the occlusion state of each grid cell at different times through geometric intersection calculation, and generating a shadow analysis grid containing occlusion probability and occlusion duration.

[0009] In some embodiments, for a rooftop scenario, inputting component specifications, row and column spacing, tilt angle, and azimuth angle, and initializing the layout with a regular array includes: generating an initial component grid array within the rooftop area boundary based on the input component specifications, row and column spacing, tilt angle, and azimuth angle; performing boundary constraint verification on the initial array based on a preset boundary contraction distance and maintenance passage width, adjusting the positions of components that exceed the area boundary or do not meet the maintenance spacing, and forming an initial layout scheme that conforms to engineering constraints.

[0010] In some embodiments, the optimization of each facade scene individually or collaboratively, automatically adapting component size and array spacing based on the facade slope and the difference in vertex height to avoid obstacles and achieve component arrangement, includes: dividing the facade into segments according to slope changes, calculating the difference in vertex height for each segment, and dynamically adjusting the component arrangement direction and size based on the height difference and component specifications; at the junction of adjacent segments, calculating the horizontal and vertical spacing between components through an adaptive step distance algorithm to ensure that the component edges maintain a safe distance from obstacles, while simultaneously achieving collaborative arrangement optimization across facade segments.

[0011] In some embodiments, the calculation of annual power generation based on regional irradiance data, module efficiency, and layout density, in conjunction with local electricity prices, subsidies, initial investment costs, and operation and maintenance costs, to calculate the internal rate of return, net present value, and payback period includes: converting horizontal global irradiance data into irradiance received on the module mounting plane using an inclined plane irradiance conversion model, and calculating the actual power generation efficiency by combining the module temperature correction coefficient and system performance ratio; calculating the annual power generation based on the number of modules and the effective light-receiving area after layout; and retrieving local electricity prices, subsidy policies, initial investment costs, and operation and maintenance costs from a preset economic parameter database, calculating the internal rate of return, net present value, and payback period based on a cash flow model, and storing comparative data of economic parameters for multiple scenarios.

[0012] Secondly, this application provides a multi-scenario design system for photovoltaic systems, including: The scene selection unit is used to automatically parse latitude and longitude by inputting the project address, associate local or third-party irradiance and meteorological data, including horizontal global irradiance, wind speed and temperature; select the photovoltaic system application scene, which includes roof scene and facade scene; enter the project name and basic constraints; and generate a unique project ID. The outline drawing unit is used to draw the boundary or facade outline of a region on a base map through interactive operations. The base map includes satellite imagery or bitmap. Clicking on any side of the drawn boundary and inputting the measured length triggers the full map to scale and calibrate proportionally, achieving global scale uniformity. Entering the facade mode, the elevation of multiple facades is input point by point to generate a three-dimensional model. The optimization calculation unit is used to interactively place or import obstacles, including parapet walls, fans, stairwells, ventilation shafts, windows, and protruding structures. Based on the geometric deduction of the solar path, a shadow analysis mesh is generated in real time. For rooftop scenarios, component specifications, row and column spacing, tilt angle, and azimuth angle are input to initialize the arrangement with a regular array, and optimization calculations are performed based on irradiance and shading cost functions. For facade scenarios, each facade is optimized individually or collaboratively, automatically adapting component sizes and array spacing based on the facade slope and the difference in vertex height to avoid obstacles and achieve component arrangement. The report generation unit is used to calculate annual power generation based on regional irradiance data, component efficiency, and layout density, and to calculate internal rate of return, net present value, and investment payback period by linking local electricity prices, subsidies, initial investment costs, and operation and maintenance costs. It supports comparison of multiple scenarios and generates standardized reports containing 3D renderings, layout diagrams, technical parameter tables, and revenue curves through a 3D interactive interface for model rotation, scaling, and screenshotting.

[0013] Thirdly, this application also provides a computer device, comprising: Memory and processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the steps of the method described in the first aspect above.

[0014] Fourthly, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the steps of the method described in the first aspect above.

[0015] This application provides a multi-scenario design method, system, equipment, and medium for photovoltaic systems. The method uses a unique project ID to connect data input, modeling, layout, calculation, and report generation, achieving real-time data linkage at each stage and significantly shortening the design cycle. Based on proportional scaling calibration and 0.01-meter-level elevation parameterization, it ensures a high degree of matching between the 3D model and the actual scene, solving the problem of large errors in traditional manual calibration. It supports irregular roofs and multi-facade collaborative optimization, automatically and adaptively adjusting the component layout according to the geometric characteristics of the scene, and accurately avoiding obstacles by combining shadow analysis grids, thereby improving installed capacity and power generation efficiency. By deeply coupling irradiance data, layout density, and economic parameters, it outputs key indicators such as power generation and IRR in real time, supports dynamic comparison of multiple schemes, and provides quantitative basis for decision-making. The 3D interactive interface enables intuitive verification of the scheme and generates a standardized report containing technical, economic, and visualization results with one click, improving communication and approval efficiency.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart illustrating the steps of a multi-scenario design method for a photovoltaic system provided in one embodiment of this application; Figure 2 This is a schematic diagram of the structure of a photovoltaic system multi-scenario design system provided in one embodiment of this application; Figure 3 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0022] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0023] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0024] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0025] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0026] In the field of photovoltaic system design, existing technologies generally suffer from multiple technical bottlenecks: 1. Insufficient data calibration and modeling accuracy: When drawing area boundaries based on satellite imagery or bitmaps, traditional methods lack a unified scale calibration mechanism and rely on manual measurement, resulting in significant geometric errors. In particular, they are weak in the ability to model elevation parameters for multi-facade scenes, making it difficult to meet the 0.01-meter accuracy requirement. 2. Poor adaptability to complex scenarios: Mainstream tools only support simple layouts on regular roofs. They lack the ability to automatically optimize component layouts for irregular shapes, curved surfaces, or multi-facade collaborative scenarios. They cannot adaptively adjust component size and spacing based on facade slope and vertex height difference. Furthermore, their avoidance strategies for obstacles such as parapet walls, skylights, and protruding structures are crude, which can easily leave behind the risk of obstruction. 3. Lack of full-process linkage: The results of power generation calculation, economic indicator evaluation and real-time scheduling are disconnected from each other, relying on offline estimation and unable to dynamically couple irradiation data, shading effects and cost parameters based on three-dimensional models for real-time iteration; 4. Low visualization and delivery efficiency: The lack of a 3D interactive verification environment means that communication of solutions relies on 2D drawings, and technical parameters, economic reports and visualization results need to be manually integrated, making it difficult to meet the requirements for standardized delivery.

[0027] In the existing technology, there is no fully automated method that can deeply integrate regional intelligent calibration, multi-facade parametric modeling, automatic obstacle avoidance, real-time benefit calculation and 3D visualization verification.

[0028] To resolve the above issues, please refer to [link / reference]. Figure 1 This multi-scenario design method for photovoltaic systems can be implemented using computer equipment, which can be deployed on a single server or a server cluster. It can also be deployed on handheld terminals, laptops, wearable devices, or robots, etc.

[0029] Specifically, such as Figure 1 As shown, the provided photovoltaic system multi-scenario design method includes steps S101 to S104, which are detailed below: Step S101. Automatically parse latitude and longitude by inputting the project address, associate local or third-party irradiance and meteorological data, including horizontal global irradiance, wind speed and temperature; select the photovoltaic system application scenario, which includes rooftop scenario and facade scenario, enter the project name and basic constraints, and generate a unique project ID.

[0030] Specifically, by inputting the project address, the system automatically obtains the geographical coordinates, associates them with meteorological and irradiance data, defines the project scenario and basic information, and generates a unique identifier to connect the data throughout the entire process.

[0031] Address resolution and coordinate acquisition utilize geocoding technology to parse the user-input project address (e.g., "rooftop of a commercial complex") and automatically extract latitude and longitude coordinates (accurate to 6 decimal places, e.g., 22.3964°N, 113.5507°E). For each project, the corresponding regional surveying and mapping geographic information service interface is prioritized to ensure compliance.

[0032] Linking meteorological and irradiance data: The local database pre-stores typical meteorological year (TMY) data, including horizontal global irradiance (H, unit: kWh / m²). 2 It supports parameters such as monthly average wind speed (m / s) and temperature (°C); it also supports integration with third-party data platforms (such as Solargis and Meteonorm) to obtain real-time irradiance and meteorological data of the project location via API. The data association logic matches the nearest meteorological station based on latitude and longitude or generates customized meteorological datasets through spatial interpolation algorithms (such as Kriging interpolation).

[0033] Scene definition and project initialization provide an interactive interface for users to select the scene type (roof / facade), enter the project name, owner information, building type (commercial / industrial / residential), and roof load limit (e.g., 2.0 kN / m²). 2 Basic constraints include fire lane width (e.g., 1.2m). A unique project ID (e.g., "PV20251104001") is automatically generated and used as a data index throughout the entire process of modeling, layout, and calculation, ensuring that data from each module is linked through a unified ID.

[0034] Step S102. Draw the area boundary or facade outline on the base map through interactive operation. The base map includes satellite imagery or bitmap. Click on any side of the drawn boundary and input the measured length to trigger the full map to scale and calibrate it proportionally, so as to achieve global scale uniformity. Enter the facade mode, input the elevation point by point for multiple facades, and generate a three-dimensional model.

[0035] Specifically, the regional boundaries are drawn based on the base map, and the global scale is calibrated by actual length measurement to generate a high-precision multi-faceted 3D model.

[0036] Boundary drawing and interactive calibration include: Base map loading: Supports importing satellite imagery (such as Google Earth tiles, UAV orthophotos) or bitmaps (CAD drawings, JPG / BMP formats), providing zoom and pan interactive tools. Users can drag and drop to draw polygonal area boundaries (roof outlines or facade outlines). Scale calibration: Users double-click on any boundary line segment, input the measured length (e.g., "actual length of a side is 20.5m"), and the system calculates the pixel length of the line segment in the base map, generating a global scaling factor (Scale Factor = actual length / pixel length), and scaling the entire map coordinates proportionally to ensure geometric dimension accuracy error ≤0.1%.

[0037] Multi-facade parametric modeling includes: Facade mode activation: For the facade scene, the user switches to the "Facade Editing" interface and marks the elevation (Z-axis coordinate, accuracy 0.01m) point by point on the two-dimensional outline. For example, the elevations of the four vertices of the south facade of the office building are "100.00m, 100.00m, 105.50m, 105.50m". The system generates a three-dimensional facade mesh model through the triangulation algorithm.

[0038] Model generation technology: Using a 3D modeling engine (such as Three.js, Unity) or CAD kernel (such as OpenCASCADE), the spatial relationship between each facade is automatically associated to generate a parametric 3D model that includes roof slope and facade inclination angle, and supports exporting OBJ / STL format for subsequent analysis.

[0039] Step S103. Place or import obstacles interactively. The obstacles include parapet walls, fans, stairwells, ventilation shafts, windows, and protruding structures. Generate a shadow analysis mesh in real time based on the geometric deduction of the solar path. For the roof scene, input the component specifications, row and column spacing, tilt angle, and azimuth angle, initialize the arrangement with a regular array, and perform optimization calculations based on the irradiance and shading cost function. For the facade scene, optimize each facade individually or collaboratively, automatically adapt the component size and array spacing according to the facade slope and the difference in vertex height, and avoid obstacles to achieve component arrangement.

[0040] Specifically, obstacles are labeled and a shadow analysis mesh is generated. An intelligent layout algorithm is executed based on the scene type (roof / facade) to avoid occlusion risks.

[0041] Obstacle labeling and shadow analysis include: Obstacle input: Supports two methods: Interactive placement: Drag and drop preset obstacles (parapet walls, fans, skylights, etc.) from the component library, setting height, size, and position parameters; Batch import: Parses CAD drawings (DXF / DWG format) or BIM models (IFC format), automatically identifies obstacle geometric features, and maps them to the 3D model. Shadow mesh generation: Calculates the annual solar altitude angle and azimuth angle of the project location based on solar path algorithms (such as Solar Position Algorithm, SPA), generates hourly shadow cones, and constructs an occlusion cost matrix through mesh division (such as 0.5m × 0.5m resolution), marking the average annual occlusion duration of each mesh cell.

[0042] The intelligent rooftop layout includes: Initial layout: A regular array is generated based on the module specifications (e.g., 182mm×182mm monocrystalline modules), preset row and column spacing (e.g., 1.2m horizontally, 2.0m vertically), and tilt angle (default 180° facing south). The boundary automatically shrinks to a safe distance (e.g., 0.5m from the parapet wall). Optimization algorithm: A constrained non-dominated sorting genetic algorithm (NSGA-II) is used, with the objective functions of "maximizing annual irradiance reception" and "minimizing shading cost". The module row and column positions are iteratively adjusted to avoid the shadow areas of obstacles, while also meeting constraints such as proximity of inverter wiring and maintenance channel width (≥1m).

[0043] Intelligent facade layout includes: Facade-specific optimization: Each facade is individually gridded, and component sizes are automatically adapted based on facade slope (e.g., 30° tilt angle) and vertex height difference (e.g., 1.5m difference between adjacent vertices) (supporting component cutting or selection of different specifications). A dynamic programming algorithm is used to determine the optimal arrangement direction (vertical / horizontal). Cross-facade collaboration: For scenarios with multiple connected facades (e.g., L-shaped buildings), shared boundary constraints ensure that component layout on adjacent facades does not obstruct each other, optimizing component density at cross-facade joints. Obstacle avoidance: Buffer zones are generated around obstacles (e.g., a 0.3m safety distance is reserved at window edges), collision detection algorithms exclude unsuitable installation areas, and component density is prioritized on high-irradiance facades.

[0044] Step S104. Calculate the annual power generation based on regional irradiance data, component efficiency, and layout density. In conjunction with local electricity prices, subsidies, initial investment costs, and operation and maintenance costs, calculate the internal rate of return, net present value, and investment payback period, supporting comparison of multiple schemes. Through a 3D interactive interface, rotate, scale, and screenshot the model to generate a standardized report containing 3D renderings, layout diagrams, technical parameter tables, and revenue curves.

[0045] Specifically, it links irradiance, arrangement, and economic parameters to calculate benefit indicators, and provides 3D interactive visualization and standardized report generation.

[0046] The power generation and economic calculations include: The irradiance conversion model calculates the plane received irradiance (POA): Based on the component tilt angle, azimuth angle and shadow grid data, the horizontal plane irradiance is converted into the inclined plane irradiance. The formula is: POA=H×(cosθ+ρ / 2(1-cosθ)); where θ is the solar altitude angle and ρ is the ground reflectivity (default 0.2).

[0047] Considering temperature correction: Module efficiency decreases as temperature increases, and the correction formula is: η=ηSTC×[1-αT(Tcell-25℃)]; where αT is the temperature coefficient (-0.34% / ℃), and Tcell is the module operating temperature (calculated from meteorological data).

[0048] Economic indicator calculation: Annual power generation E = ∑(POA × η × A × PR), where A is the total area of ​​the modules and PR is the performance ratio (default 0.85); Linked input parameters: electricity price (e.g., 0.8 yuan / kWh), subsidy (e.g., 0.3 yuan / kWh), initial investment (CAPEX, including equipment / installation costs), operation and maintenance cost (OPEX, 1% of CAPEX annually), calculated using the net present value (NPV) formula: ; Where Ct is the net income in year t, r is the discount rate (default 8%), and the internal rate of return (IRR) is obtained by solving the discount rate when NPV=0.

[0049] Visualization and report generation include: 3D interaction: A 3D view based on WebGL technology, supporting mouse drag-and-drop rotation, scroll wheel zoom, and component selection to view detailed parameters (such as occlusion rate and power generation contribution), providing multi-view screenshot functionality (front view, bird's-eye view, elevation unfolded view). Standardized reports: A built-in template engine (such as Puppeteer to generate PDF) automatically integrates the following: Visualization results: 3D renderings, shadow analysis heatmaps, component layout vector maps; Technical parameters: Installed capacity, number of components, occlusion rate, average annual power generation, PR value; Economic calculations: IRR, NPV, payback period, and return curves (annual returns for the first 20 years); Compliance documents: Boundary coordinate tables, load verification tables, and fire safety spacing specifications. Multi-scheme comparison: Supports saving ≥2 schemes, generating comparison tables highlighting key indicator differences (such as a 5% increase in power generation, a 1.2% increase in IRR), assisting in decision-making.

[0050] In some embodiments, the irradiance and shading cost function considers shading effects, row spacing, boundary constraints, and inverter proximity principles to avoid obstacles and achieve component arrangement. The optimization calculation based on the irradiance and shading cost function includes: after initializing the component arrangement with a regular array, determining the shading area according to the shadow analysis grid, and establishing a cost function model that includes irradiance loss weight, row spacing constraint weight, boundary distance weight, and inverter access distance weight; calculating the cost function value of each arrangement scheme by iteratively adjusting the row and column positions of the components, and selecting the arrangement scheme with the smallest cost function value that satisfies the obstacle avoidance condition.

[0051] In the arrangement of components in the rooftop scene, by constructing a cost function model with multiple constraint weights and combining obstacle avoidance conditions, the row and column positions of components are iteratively optimized to achieve intelligent arrangement that takes into account both irradiance benefits and engineering constraints.

[0052] The cost function model construction includes: Irradiance loss weight (Wirr): Based on the average annual shading duration at each component location in the shading analysis grid, the irradiance loss ratio is calculated (the weight increases as the shading rate > 10%), using the following formula: (Gpoa,i is the actual received irradiance, and Gpoa,max is the unobstructed irradiance).

[0053] Row spacing constraint weight (Wrow): Ensures that the row and column spacing is greater than or equal to the design value (e.g., 1.2m horizontally and 2.0m vertically). If the spacing is insufficient, the cost is increased proportionally, as shown in the formula: (dmin is the minimum allowable spacing, dactual is the actual spacing).

[0054] The boundary distance weight (Wboundary) is the minimum safe distance (e.g., 0.5m) that must be maintained between the component edge and the roof boundary. Exceeding this range increases the cost linearly, as shown in the formula: (bmin is the minimum boundary distance, bactual is the actual distance).

[0055] Inverter connection distance weighting (Winverter): Prioritizes bringing components closer to the inverter location (preset coordinates), with higher costs for greater distances. The formula is: (xi,yi are component coordinates, xinv,yinv are inverter coordinates).

[0056] Total cost function: Cost = Wirr * Lirr + Wrow * Lrow + Wboundary * Lboundary + W inverter * Linverter; weighting coefficients can be user-defined (default Wirr = 0.5, Wrow = 0.2, Wboundary = 0.2, Winverter = 0.1).

[0057] The iterative optimization process includes: Initialization: Generate an initial layout according to a rule array (e.g., a 5×10 component matrix) and mark obstacle positions (e.g., air conditioner unit locations). Neighborhood search: In each iteration, randomly adjust the positions of 1-2 rows / columns of components (offset ≤ 5% of row / column spacing) to generate a new solution. Feasibility verification: The new solution must meet the obstacle buffer distance requirement (e.g., ≥ 0.3m from the obstacle edge); otherwise, skip it. Optimal selection: Retain the top 20% of solutions with the lowest Cost values ​​and repeat the iteration until the Cost value converges (rate of change < 1%).

[0058] In some embodiments, the step of rotating, scaling, and taking screenshots of the model through a 3D interactive interface to generate a standardized report containing a 3D rendering, layout diagram, technical parameter table, and revenue curve includes: rendering the 3D model from multiple angles using a 3D rendering engine, automatically extracting component layout coordinate data to generate a 2D layout diagram, extracting annual power generation, internal rate of return, net present value, and investment payback period from the 2D layout diagram to generate a technical parameter table and revenue curve, and integrating them according to a preset template to form a standardized report.

[0059] By using 3D rendering and data extraction technologies, 3D models, layout data, and revenue indicators are integrated into a standardized report, supporting multi-view visualization and parameter export.

[0060] 3D rendering and interaction include: Rendering engine: Using the Three.js or Unity engine, 3D models are rendered with lighting and shadows (supporting PBR materials, such as component glass reflection effects). Users can rotate the model with the right mouse button, zoom with the scroll wheel, and pan with the left mouse button. Viewpoint management: Preset commonly used viewpoints (top view, 45° side view, elevation view), which can be automatically switched by clicking a button; supports selecting components to display real-time parameters (such as "Component ID: C001, Occlusion rate: 5%, Annual power generation: 120kWh").

[0061] Data extraction and report generation include: 2D layout diagram: Extracting component coordinates (X,Y plane projection) from the 3D model, automatically generating a DWG format vector map, and labeling component numbers, obstacle locations, and maintenance access routes (width ≥ 1m). Technical parameter table: Extracting core indicators from the 2D layout diagram, including: installed capacity (kWp), number of components, and effective light-receiving area (m²). 2 Annual power generation (kWh), PR value (performance ratio), first-year degradation rate (%); shading rate (percentage of shaded modules), boundary compliance rate (percentage of modules meeting safety distance requirements). Yield curve: Based on 20-year lifecycle cash flow data, a line chart is generated (horizontal axis: year, vertical axis: NPV / IRR / payback period), supporting export to PNG / SVG format. Template integration: Using LaTeX or HTML templates, content is arranged by chapter: cover (project name / ID), table of contents, 3D renderings (3-5 typical views), layout diagram (roof / facade layered display), technical parameter table (with units), yield curve comparison, appendix (coordinate data / load calculation sheet), finally generating a PDF report (supporting A3 / A4 format).

[0062] In some embodiments, the step of generating a shadow analysis grid in real time based on the geometric deduction of the solar path includes: calculating the solar altitude angle and azimuth angle for different seasons based on the input latitude, longitude and time parameters, and generating the solar path trajectory; constructing a geometric model of a shadow cone along the solar path with the obstacle vertex as a reference, dividing the target area into grid cells of preset precision, determining the occlusion state of each grid cell at different times through geometric intersection calculation, and generating a shadow analysis grid containing occlusion probability and occlusion duration.

[0063] Based on the solar path algorithm and geometric modeling, a high-precision shadow analysis grid is generated, quantifying the occlusion probability and duration of each grid cell, providing a data foundation for intelligent layout.

[0064] The solar path calculation includes: Input parameters: using latitude and longitude, project year (e.g., 2025), and time precision (hourly / 15-minute intervals), calling SPA (Solar Position Algorithm) to calculate the solar altitude angle (α) and azimuth angle (γ), with the formula: sinα=sinδsin +cosδcos cosω (δ is the solar declination, (where ω is latitude and ω is hour angle).

[0065] The shadow cone construction and mesh generation process includes: Obstacle modeling: Abstracting obstacles into cubes / cylinders, extracting vertex coordinates (such as the parapet wall vertex coordinate sequence), and constructing a shadow cone along the sun's path (from sunrise to sunset) (the base is the obstacle outline, and the height varies with the sun's altitude angle). Mesh generation: Dividing the target area into 0.5m × 0.5m grids (with customizable precision), with the center point of each grid serving as a detection point. Occlusion calculation: For each grid point, determining hourly whether it is located within the obstacle's shadow cone (using a ray intersection algorithm, such as the Möller-Trumbore algorithm, to detect the intersection of rays with the obstacle surface), recording the occlusion duration (unit: hours / year) and occlusion probability (occlusion duration / 8760 hours).

[0066] The grid data output generates a two-dimensional matrix file (CSV format), with each row corresponding to a grid cell, containing coordinates (X,Y), average annual occlusion duration (h), and occlusion rate (%), which is used by the intelligent layout engine (e.g., grids with an occlusion rate >20% are marked as uninstallable areas).

[0067] In some embodiments, for a rooftop scenario, inputting component specifications, row and column spacing, tilt angle, and azimuth angle, and initializing the layout with a regular array includes: generating an initial component grid array within the rooftop area boundary based on the input component specifications, row and column spacing, tilt angle, and azimuth angle; performing boundary constraint verification on the initial array based on a preset boundary contraction distance and maintenance passage width, adjusting the positions of components that exceed the area boundary or do not meet the maintenance spacing, and forming an initial layout scheme that conforms to engineering constraints.

[0068] In the rooftop scenario, an initial layout plan is generated based on component specifications and engineering constraints. Boundary contraction and inspection channel verification are then used to ensure that the layout complies with installation specifications.

[0069] The regular array generation includes: parameter input via user input of component dimensions (e.g., 1.82m × 0.91m), row and column spacing (horizontal spacing 1.2m, vertical spacing 2.0m), tilt angle (30°), and azimuth angle (180° south). Mesh generation uses the component's geometric center as the reference point, generating a mesh within the roof projection area according to the row and column spacing (rows along the azimuth direction, columns along the tilt angle perpendicular direction), and calculating the maximum number of rows that can be arranged (Nrow = |(Ww) / drow| + 1, where W is the area width, w is the component width, and drow is the horizontal spacing).

[0070] Boundary constraint verification and adjustment include: Boundary contraction: The edge of the component must maintain a safe distance (e.g., 0.5m) from the roof boundary. The outermost component exceeding the boundary will automatically move inward (the amount of movement = the distance exceeded) until all components are within the boundary. Maintenance access verification: A blank area ≥1.2m wide is reserved along the main roof access direction (e.g., fire access). If the initial array occupies the access, the components in that area are deleted, and the spacing between adjacent rows and columns is recalculated (a ±10% fluctuation in spacing is allowed to ensure compliance with access width regulations).

[0071] The initial scheme output generates an initial layout list containing component coordinates, tilt angles, and azimuth angles, and marks the boundary compliance status (compliant / needs adjustment), which serves as the input basis for subsequent optimization algorithms.

[0072] In some embodiments, the optimization of each facade scene individually or collaboratively, automatically adapting component size and array spacing based on the facade slope and the difference in vertex height to avoid obstacles and achieve component arrangement, includes: dividing the facade into segments according to slope changes, calculating the difference in vertex height for each segment, and dynamically adjusting the component arrangement direction and size based on the height difference and component specifications; at the junction of adjacent segments, calculating the horizontal and vertical spacing between components through an adaptive step distance algorithm to ensure that the component edges maintain a safe distance from obstacles, while simultaneously achieving collaborative arrangement optimization across facade segments.

[0073] To address the complex slopes and height differences of the facade, the size and spacing of components are dynamically adjusted through segmented optimization and adaptive step distance algorithms, enabling cross-segment collaborative layout and obstacle avoidance.

[0074] Facade segmentation and parameter calculation include: Slope segmentation: Detecting slope changes along the facade height (automatic segmentation when slope difference > 15°), generating independent planar equations for each segment (e.g., segment 1 slope 30°, segment 2 slope 45°). Height difference handling: Calculating the height difference between the top and bottom vertices of each facade segment (e.g., a height difference of 5m between the top and bottom vertices of segment 1), and dynamically selecting the component arrangement direction based on component specifications (e.g., standard component height 1m) (vertical arrangement is preferred when height difference > 2m, otherwise horizontal arrangement).

[0075] Adaptive step distance and obstacle avoidance include: Spacing calculation: Horizontal step distance between adjacent components sh=max(0.3m, hdiff / tanα+0.1m) (hdiff is the height difference between the upper and lower components, α is the minimum solar altitude angle), vertical step distance sv=component height+0.2m (maintenance spacing).

[0076] Obstacle buffering generates a circular buffer zone with a radius of 0.3m around obstacles such as windows and protruding structures. The edges of components must not enter this zone, and the feasibility of the arrangement is verified by a collision detection algorithm (such as AABB bounding box).

[0077] Cross-segment collaborative optimization forces the components of adjacent segments to align at the junctions (error ≤ 0.05m), and selects the optimal joint component size through a dynamic programming algorithm (e.g., 1.6m component for segment 1, 1.2m component for segment 2, and gap at the joint ≤ 0.1m) to ensure that there are no visual breaks in the cross-segment layout and that no additional occlusion is added.

[0078] In some embodiments, the calculation of annual power generation based on regional irradiance data, module efficiency, and layout density, in conjunction with local electricity prices, subsidies, initial investment costs, and operation and maintenance costs, to calculate the internal rate of return, net present value, and payback period includes: converting horizontal global irradiance data into irradiance received on the module mounting plane using an inclined plane irradiance conversion model, and calculating the actual power generation efficiency by combining the module temperature correction coefficient and system performance ratio; calculating the annual power generation based on the number of modules and the effective light-receiving area after layout; and retrieving local electricity prices, subsidy policies, initial investment costs, and operation and maintenance costs from a preset economic parameter database, calculating the internal rate of return, net present value, and payback period based on a cash flow model, and storing comparative data of economic parameters for multiple scenarios.

[0079] By coupling the irradiation conversion model with economic parameters, real-time linkage between power generation and revenue indicators is achieved, supporting the comparison and storage of economic parameters from multiple scenarios.

[0080] The power generation calculation model includes: POA conversion: converting horizontal surface irradiance H into module tilt surface irradiance Gpoa, the formula is: Gpoa=Hb×Rb +Hd×(1+cosβ) / 2+H×ρ×(1-cosβ) / 2 (Hb is direct irradiance, Hd is diffuse irradiance, β is module tilt angle, ρ is ground reflectivity).

[0081] Efficiency Correction: Considering the effect of component temperature, the actual efficiency η = ηSTC × [1 - αT (Ta + ΔT) - 25℃], where ΔT = (Gpoa × SNOCT) / (800W / m 2(SNOCT is the irradiation at the nominal operating temperature). Annual power generation E = ∑(Gpoa,i×ηi×Ai×PR)×365, where Ai is the area of ​​a single component and PR is the system performance ratio (default 0.85).

[0082] Economic indicator calculations include: Cash flow model: Net income in year t Ct = Et × (electricity price + subsidy) - OPEXt, where OPEXt = CAPEX × 1% (average annual operation and maintenance cost), and initial investment CAPEX = number of components × unit price + installation cost. IRR and NPV: The discount rate when NPV = 0 is obtained by solving the Newton-Raphson iteration method, which is the IRR. The formula is: The payback period is the year in which the cumulative net present value first reaches or exceeds 0.

[0083] The multi-scheme comparison storage is achieved by establishing an SQLite database table with fields including: scheme ID, project ID, power generation (kWh), IRR (%), NPV (yuan), payback period (years), and creation time. It supports querying historical schemes by project ID and generating comparison tables (e.g., scheme A has a 1.5% higher IRR than scheme B, but its CAPEX is 8% higher), to assist users in making decisions.

[0084] In some embodiments, a GAN model with engineering constraint awareness is constructed for complex roof / facade scenarios. An irregular array layout scheme is automatically generated by a generator, and the feasibility of the scheme (occlusion, spacing, boundary constraints) is verified by a discriminator, thereby achieving a balance between "creative layout" and "engineering compliance".

[0085] Data preparation and model architecture include: Training data: Collecting historical compliant layout schemes (100,000+ cases), extracting component coordinates, obstacle positions, and constraint parameters (such as row and column spacing, boundary distance) as input, labeled as "compliant / non-compliant". Generator G: Employing a Convolutional Neural Network (CNN) + Transformer structure, the input is a noise vector (100-dimensional) and scene parameters (roof outline, obstacle coordinates), outputting a component coordinate matrix (X, Y, angle). Discriminator D: A dual-channel CNN that simultaneously judges the irradiance rationality of the scheme (based on shadow grid prediction of power generation) and engineering compliance (boundary distance, row and column spacing verification), outputting a 0-1 compliance probability.

[0086] Adversarial training and constrained embedding: Loss function: LG=-Ex~G[logD(x)]+λ*L constraint; where Lconstraint is the penalty term for violations of boundary distance and row and column distance (e.g., 0.1 loss is added for each violation).

[0087] Engineering constraint mandatory verification: After the scheme is generated, the components are verified by geometric algorithm to see if they exceed the boundary (error ≤ 0.05m) and whether the row and column spacing is ≥ the minimum threshold (allowing ±5% fluctuation). Schemes that violate the rules are directly filtered out.

[0088] Iterative optimization and solution output generate 100 candidate solutions each time. The top 20 solutions with a compliance probability > 0.8 are selected by D-straining. The solutions are then input into the cost function calculator (same as in Example 1) to calculate the comprehensive score. The three solutions with the highest scores are selected for user decision-making. Irregular arrays (such as L-shaped or circular arrangements) are supported to avoid complex obstacles.

[0089] In some embodiments, by designing an intelligent mesh accuracy adjuster, the mesh density of shadow analysis is dynamically adjusted based on reinforcement learning (RL), automatically balancing computational efficiency and accuracy, thus solving the problem of "computational redundancy in simple areas and insufficient accuracy in complex areas" in traditional fixed meshes.

[0090] The state-action space definition includes: State S: The geometric complexity of the current region (number of obstacles, edge curvature), computation time, and preset precision threshold (e.g., ≤10ms / mesh). Action A: Adjust mesh precision (coarse 1m → fine 0.25m, a total of 5 levels). Reward R: +10 for achieving the precision target (e.g., fine granularity in complex regions satisfies occlusion calculation error <5%), -5 for computation timeout penalty, and +5 for saving computational power in simple regions using coarse granularity.

[0091] The RL model and training include: Algorithm: The PPO (Proximal Policy Optimization) algorithm is used. The neural network input is the scene point cloud features (obstacle distribution density extracted through PointNet), and the output is the selection probability for each precision level. Training data: 1000 scenes (simple flat roof / complex multi-obstacle facade) are manually labeled, and the computation time and occlusion error at different precision levels are recorded to construct a mapping relationship between "scene complexity - optimal precision".

[0092] The dynamic mesh generation process takes a 3D scene model as input and first predicts the optimal mesh precision using an RL model (e.g., 1m mesh for simple areas, automatically switching to 0.25m around obstacles). For obstacle-sensitive shadow areas (e.g., a 5m radius around occlusions with a height > 2m), adaptive refinement is initiated: point-by-point verification is performed using a ray tracing algorithm; if the occlusion change rate > 15%, the mesh is further refined to 0.1m. Finally, a variable-precision shadow mesh (stored as a sparse matrix) is generated, improving computational efficiency by 30%-50% and reducing occlusion prediction errors in complex areas to within 3%.

[0093] In some embodiments, by constructing a domain-specific NLP model, unstructured user input requirements (such as "generate an aesthetically pleasing report suitable for industrial plants") are automatically parsed, and personalized reports are dynamically generated in combination with photovoltaic design data, supporting multi-language and multi-format output.

[0094] Demand parsing and intent recognition include: Model: Fine-tuning of the GPT-4 photovoltaic-specific version (training data includes 100,000+ project reports and industry standards), inputting user commands (such as "emphasize comparative analysis of operation and maintenance costs"), and outputting structured demand tags (such as {report type: comparative, key module: operation and maintenance cost, format: PPT}). Entity extraction: Extracting key parameters, such as "internal rate of return > 15%" and "must include 3D walkthrough animation," through Named Entity Recognition (NER), and converting them into constraints for report generation.

[0095] Dynamic template generation and content population: Template library: 10+ predefined report templates (feasibility study version, construction version, presentation version), each template contains configurable modules (such as the number of 3D views, the granularity of economic indicators).

[0096] Content generation includes: Text description: Automatically generates technical specifications based on the layout plan (e.g., "This plan has been optimized through 23 iterations, avoiding 3 ventilation ducts, and achieving a row and column spacing compliance rate of 98%"). Chart customization: Dynamically adjusts curve dimensions according to user needs (e.g., simultaneously displaying IRR comparisons under different electricity price scenarios), and supports adding custom annotations (e.g., "The spacing here is adjusted due to fire lane requirements").

[0097] Multimodal output and interaction include: generating 3D walkthrough videos with voice narration (automatically adapting to 16:9 / 4:3 aspect ratios), and highlighting interactive hotspots in key areas (click to view component parameters). It also supports natural language querying of report content: users can input "view the payback period for Option 2," and the system will directly locate and highlight the relevant section in the report, improving readability for non-technical personnel.

[0098] In some embodiments, by constructing a photovoltaic design knowledge graph, integrating multi-source data such as industry standards, equipment parameters, and case experience, and using graph neural networks (GNNs) to achieve intelligent reasoning of "design-constraint-benefit", potential problems can be automatically identified (such as "the distance between a certain type of inverter and the module is too far, resulting in excessive cable costs").

[0099] The knowledge graph construction includes: entity types: modules (power / size), inverters (capacity / access distance limitations), obstacles (type / safe distance), and standards (GB 50797-2012 Photovoltaic Design Code). Relationship modeling includes: "Module-Adapter-Inverter" (e.g., if a module's maximum series connection count is ≤15, corresponding to the inverter's MPPT channel count), and "Obstacle-Forced Constraint-Layout Spacing" (e.g., fire lanes require a 1.5m clear width).

[0100] Intelligent reasoning and conflict detection include: Rule Engine: After the layout plan is generated, SPARQL queries are used to verify: "Are there any components with inverter access distance > 50m?" (triggering cable cost warning) "Are fall arrestors used in areas with tilt angle > 45°?" (automatically associating with regulatory clauses). Case Reasoning: Input scene features (such as "sloping roof + multiple chimneys"), retrieve similar historical cases (similarity > 80%), and recommend optimal row and column spacing parameters (such as the average row spacing in historical cases + 10% to avoid chimney shadows).

[0101] Decision support outputs include generating a design compliance report: listing the compliance status of each specification, with links to the original clauses for violations (e.g., "Insufficient boundary distance, violating GB 50797 Clause 6.2.3"). It also provides a recommended equipment list: based on the number of components and series-parallel connections in the layout scheme, it predicts the optimal inverter model using a GNN (accuracy > 90%), along with cost comparisons (e.g., the recommended model saves 15% on cable costs compared to the user's initial selection).

[0102] This application provides a multi-scenario design method, system, equipment, and medium for photovoltaic systems. The method uses a unique project ID to connect data input, modeling, layout, calculation, and report generation, achieving real-time data linkage at each stage and significantly shortening the design cycle. Based on proportional scaling calibration and 0.01-meter-level elevation parameterization, it ensures a high degree of matching between the 3D model and the actual scene, solving the problem of large errors in traditional manual calibration. It supports irregular roofs and multi-facade collaborative optimization, automatically and adaptively adjusting the component layout according to the geometric characteristics of the scene, and accurately avoiding obstacles by combining shadow analysis grids, thereby improving installed capacity and power generation efficiency. By deeply coupling irradiance data, layout density, and economic parameters, it outputs key indicators such as power generation and IRR in real time, supports dynamic comparison of multiple schemes, and provides quantitative basis for decision-making. The 3D interactive interface enables intuitive verification of the scheme and generates a standardized report containing technical, economic, and visualization results with one click, improving communication and approval efficiency.

[0103] Please see Figure 2 As shown, Figure 2This is a schematic diagram of the structure of the photovoltaic system multi-scenario design system 200 provided in the embodiments of this application. The photovoltaic system multi-scenario design system 200 is used to execute the steps of the photovoltaic system multi-scenario design method shown in the above embodiments. The photovoltaic system multi-scenario design system 200 can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0104] like Figure 2 As shown, the photovoltaic system multi-scenario design system 200 includes: The scene selection unit 201 is used to automatically parse latitude and longitude by inputting the project address, associate local or third-party irradiance and meteorological data, including horizontal global irradiance, wind speed and temperature; select the photovoltaic system application scene, which includes roof scene and facade scene; enter the project name and basic constraints; and generate a unique project ID. The outline drawing unit 202 is used to draw the boundary or facade outline of a region on a base map through interactive operations. The base map includes satellite imagery or bitmap. Clicking on any side of the drawn boundary and inputting the measured length triggers the full map to scale and calibrate proportionally, achieving global scale uniformity. Entering the facade mode, the elevation of multiple facades is input point by point to generate a three-dimensional model. The optimization calculation unit 203 is used to place or import obstacles interactively. These obstacles include parapet walls, fans, stairwells, ventilation shafts, windows, and protruding structures. Based on the geometric deduction of the solar path, a shadow analysis mesh is generated in real time. For the roof scene, the component specifications, row and column spacing, tilt angle, and azimuth angle are input to initialize the arrangement with a regular array, and optimization calculations are performed based on the irradiance and shading cost function. For the facade scene, each facade is optimized individually or collaboratively, and the component size and array spacing are automatically adapted according to the facade slope and the difference between the top and bottom heights to avoid obstacles and achieve component arrangement. The report generation unit 204 is used to calculate annual power generation based on regional irradiance data, component efficiency, and layout density, and to calculate internal rate of return, net present value, and investment payback period by linking local electricity prices, subsidies, initial investment costs, and operation and maintenance costs. It supports comparison of multiple schemes and generates a standardized report containing 3D renderings, layout diagrams, technical parameter tables, and revenue curves through a 3D interactive interface for model rotation, scaling, and screenshotting.

[0105] In some embodiments, the irradiance and shading cost function considers shading effects, row spacing, boundary constraints, and inverter proximity principles to avoid obstacles and achieve component arrangement. The optimization calculation based on the irradiance and shading cost function includes: after initializing the component arrangement with a regular array, determining the shading area according to the shadow analysis grid, and establishing a cost function model that includes irradiance loss weight, row spacing constraint weight, boundary distance weight, and inverter access distance weight; calculating the cost function value of each arrangement scheme by iteratively adjusting the row and column positions of the components, and selecting the arrangement scheme with the smallest cost function value that satisfies the obstacle avoidance condition.

[0106] In some embodiments, the step of rotating, scaling, and taking screenshots of the model through a 3D interactive interface to generate a standardized report containing a 3D rendering, layout diagram, technical parameter table, and revenue curve includes: rendering the 3D model from multiple angles using a 3D rendering engine, automatically extracting component layout coordinate data to generate a 2D layout diagram, extracting annual power generation, internal rate of return, net present value, and investment payback period from the 2D layout diagram to generate a technical parameter table and revenue curve, and integrating them according to a preset template to form a standardized report.

[0107] In some embodiments, the step of generating a shadow analysis grid in real time based on the geometric deduction of the solar path includes: calculating the solar altitude angle and azimuth angle for different seasons based on the input latitude, longitude and time parameters, and generating the solar path trajectory; constructing a geometric model of a shadow cone along the solar path with the obstacle vertex as a reference, dividing the target area into grid cells of preset precision, determining the occlusion state of each grid cell at different times through geometric intersection calculation, and generating a shadow analysis grid containing occlusion probability and occlusion duration.

[0108] In some embodiments, for a rooftop scenario, inputting component specifications, row and column spacing, tilt angle, and azimuth angle, and initializing the layout with a regular array includes: generating an initial component grid array within the rooftop area boundary based on the input component specifications, row and column spacing, tilt angle, and azimuth angle; performing boundary constraint verification on the initial array based on a preset boundary contraction distance and maintenance passage width, adjusting the positions of components that exceed the area boundary or do not meet the maintenance spacing, and forming an initial layout scheme that conforms to engineering constraints.

[0109] In some embodiments, the optimization of each facade scene individually or collaboratively, automatically adapting component size and array spacing based on the facade slope and the difference in vertex height to avoid obstacles and achieve component arrangement, includes: dividing the facade into segments according to slope changes, calculating the difference in vertex height for each segment, and dynamically adjusting the component arrangement direction and size based on the height difference and component specifications; at the junction of adjacent segments, calculating the horizontal and vertical spacing between components through an adaptive step distance algorithm to ensure that the component edges maintain a safe distance from obstacles, while simultaneously achieving collaborative arrangement optimization across facade segments.

[0110] In some embodiments, the calculation of annual power generation based on regional irradiance data, module efficiency, and layout density, in conjunction with local electricity prices, subsidies, initial investment costs, and operation and maintenance costs, to calculate the internal rate of return, net present value, and payback period includes: converting horizontal global irradiance data into irradiance received on the module mounting plane using an inclined plane irradiance conversion model, and calculating the actual power generation efficiency by combining the module temperature correction coefficient and system performance ratio; calculating the annual power generation based on the number of modules and the effective light-receiving area after layout; and retrieving local electricity prices, subsidy policies, initial investment costs, and operation and maintenance costs from a preset economic parameter database, calculating the internal rate of return, net present value, and payback period based on a cash flow model, and storing comparative data of economic parameters for multiple scenarios.

[0111] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the photovoltaic system multi-scenario design system and its modules described above can be referred to the corresponding processes in the embodiments of the photovoltaic system multi-scenario design method described above, and will not be repeated here.

[0112] The aforementioned multi-scenario design method for photovoltaic systems can be implemented as a computer program, which can be used in scenarios such as... Figure 2 It runs on the system shown.

[0113] Please see Figure 3 , Figure 3 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. The computer device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0114] The storage medium can store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform any multi-scenario design method for a photovoltaic system.

[0115] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0116] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to execute any multi-scenario design method for photovoltaic systems.

[0117] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 3The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the terminal to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0118] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0119] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: The system automatically resolves latitude and longitude by inputting the project address, and associates local or third-party irradiance and meteorological data, including horizontal global irradiance, wind speed and temperature; selects the photovoltaic system application scenario, which includes rooftop scenario and facade scenario; enters the project name and basic constraints; and generates a unique project ID. Draw the area boundary or facade outline on the base map through interactive operations. The base map includes satellite imagery or bitmap. Click on any side of the drawn boundary and enter the measured length to trigger the full map to scale and calibrate proportionally, achieving global scale uniformity. Enter the facade mode, enter the elevation point by point for multiple facades, and generate a three-dimensional model. Obstacles can be placed or imported interactively, including parapet walls, fans, stairwells, ventilation shafts, windows, and protruding structures. A shadow analysis mesh is generated in real-time based on geometric deduction of the solar path. For rooftop scenarios, component specifications, row and column spacing, tilt angle, and azimuth angle are input to initialize the layout with a regular array, and optimization calculations are performed based on irradiance and shading cost functions. For facade scenarios, each facade is optimized individually or collaboratively, automatically adapting component sizes and array spacing based on the facade slope and the difference in vertex height to avoid obstacles and achieve component layout. Annual power generation is calculated based on regional irradiance data, component efficiency, and layout density. It is linked to local electricity prices, subsidies, initial investment costs, and operation and maintenance costs to calculate internal rate of return, net present value, and investment payback period, supporting comparison of multiple schemes. The model can be rotated, scaled, and screenshotted through a 3D interactive interface to generate a standardized report containing 3D renderings, layout diagrams, technical parameter tables, and revenue curves.

[0120] In some embodiments, the irradiance and shading cost function considers shading effects, row spacing, boundary constraints, and inverter proximity principles to avoid obstacles and achieve component arrangement. The optimization calculation based on the irradiance and shading cost function includes: after initializing the component arrangement with a regular array, determining the shading area according to the shadow analysis grid, and establishing a cost function model that includes irradiance loss weight, row spacing constraint weight, boundary distance weight, and inverter access distance weight; calculating the cost function value of each arrangement scheme by iteratively adjusting the row and column positions of the components, and selecting the arrangement scheme with the smallest cost function value that satisfies the obstacle avoidance condition.

[0121] In some embodiments, the step of rotating, scaling, and taking screenshots of the model through a 3D interactive interface to generate a standardized report containing a 3D rendering, layout diagram, technical parameter table, and revenue curve includes: rendering the 3D model from multiple angles using a 3D rendering engine, automatically extracting component layout coordinate data to generate a 2D layout diagram, extracting annual power generation, internal rate of return, net present value, and investment payback period from the 2D layout diagram to generate a technical parameter table and revenue curve, and integrating them according to a preset template to form a standardized report.

[0122] In some embodiments, the step of generating a shadow analysis grid in real time based on the geometric deduction of the solar path includes: calculating the solar altitude angle and azimuth angle for different seasons based on the input latitude, longitude and time parameters, and generating the solar path trajectory; constructing a geometric model of a shadow cone along the solar path with the obstacle vertex as a reference, dividing the target area into grid cells of preset precision, determining the occlusion state of each grid cell at different times through geometric intersection calculation, and generating a shadow analysis grid containing occlusion probability and occlusion duration.

[0123] In some embodiments, for a rooftop scenario, inputting component specifications, row and column spacing, tilt angle, and azimuth angle, and initializing the layout with a regular array includes: generating an initial component grid array within the rooftop area boundary based on the input component specifications, row and column spacing, tilt angle, and azimuth angle; performing boundary constraint verification on the initial array based on a preset boundary contraction distance and maintenance passage width, adjusting the positions of components that exceed the area boundary or do not meet the maintenance spacing, and forming an initial layout scheme that conforms to engineering constraints.

[0124] In some embodiments, the optimization of each facade scene individually or collaboratively, automatically adapting component size and array spacing based on the facade slope and the difference in vertex height to avoid obstacles and achieve component arrangement, includes: dividing the facade into segments according to slope changes, calculating the difference in vertex height for each segment, and dynamically adjusting the component arrangement direction and size based on the height difference and component specifications; at the junction of adjacent segments, calculating the horizontal and vertical spacing between components through an adaptive step distance algorithm to ensure that the component edges maintain a safe distance from obstacles, while simultaneously achieving collaborative arrangement optimization across facade segments.

[0125] In some embodiments, the calculation of annual power generation based on regional irradiance data, module efficiency, and layout density, in conjunction with local electricity prices, subsidies, initial investment costs, and operation and maintenance costs, to calculate the internal rate of return, net present value, and payback period includes: converting horizontal global irradiance data into irradiance received on the module mounting plane using an inclined plane irradiance conversion model, and calculating the actual power generation efficiency by combining the module temperature correction coefficient and system performance ratio; calculating the annual power generation based on the number of modules and the effective light-receiving area after layout; and retrieving local electricity prices, subsidy policies, initial investment costs, and operation and maintenance costs from a preset economic parameter database, calculating the internal rate of return, net present value, and payback period based on a cash flow model, and storing comparative data of economic parameters for multiple scenarios.

[0126] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement the steps of the photovoltaic system multi-scenario design method provided in the above embodiments of this application.

[0127] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0128] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A multi-scenario design method for photovoltaic systems, characterized in that, include: The system automatically resolves latitude and longitude by inputting the project address, and associates local or third-party irradiance and meteorological data, including horizontal global irradiance, wind speed, and temperature. Select the photovoltaic system application scenario, which includes rooftop and facade scenarios; enter the project name and basic constraints; and generate a unique project ID. Draw the regional boundary or elevation outline on the base map through interactive operations. The base map includes satellite imagery or bitmap. Click on any side of the drawn boundary and enter the measured length to trigger the full map to scale and calibrate proportionally, achieving global scale uniformity. Enter the facade mode, input the elevation point by point for multiple facades, and generate a 3D model; Obstacles can be placed or imported interactively, including parapet walls, fans, stairwells, ventilation shafts, windows, and protruding structures; a shadow analysis mesh is generated in real time based on the geometric deduction of the solar path; for roof scenes, component specifications, row and column spacing, tilt angle, and azimuth angle are input, the arrangement is initialized with a regular array, and optimization calculation is performed based on the irradiance and shading cost function; For facade scenarios, each facade is optimized individually or collaboratively, and the component size and array spacing are automatically adapted according to the facade slope and the height difference of the apex, avoiding obstacles to achieve component arrangement; Annual power generation is calculated based on regional irradiance data, component efficiency, and layout density. It is linked to local electricity prices, subsidies, initial investment costs, and operation and maintenance costs to calculate internal rate of return, net present value, and investment payback period, and supports comparison of multiple schemes. The model can be rotated, scaled, and screenshotted through a 3D interactive interface, generating a standardized report that includes 3D renderings, layout diagrams, technical parameter tables, and profit curves.

2. The method according to claim 1, characterized in that, The irradiance and shading cost function considers shading effects, row spacing, boundary constraints, and inverter proximity principles to avoid obstacles and achieve component arrangement; the optimization calculation based on the irradiance and shading cost function includes: After initializing the component layout with a regular array, the shading area is determined according to the shadow analysis grid, and a cost function model is established that includes irradiance loss weight, row spacing constraint weight, boundary distance weight, and inverter access distance weight. By iteratively adjusting the row and column positions of components, the cost function value of each layout scheme is calculated, and the layout scheme with the smallest cost function value that meets the obstacle avoidance condition is selected.

3. The method according to claim 1, characterized in that, The process of rotating, scaling, and screenshotting the model through a 3D interactive interface to generate a standardized report containing 3D renderings, layout diagrams, technical parameter tables, and profit curves includes: The 3D model is rendered from multiple angles using a 3D rendering engine. The component layout coordinate data is automatically extracted to generate a 2D layout diagram. The annual power generation, internal rate of return, net present value, and investment payback period are extracted from the 2D layout diagram to generate a technical parameter table and a revenue curve. The data are then integrated into a standardized report according to a preset template.

4. The method according to claim 1, characterized in that, The process of generating a shadow analysis mesh in real time based on the geometric deduction of the solar path includes: Based on the input latitude, longitude and time parameters, the solar altitude angle and azimuth angle for different seasons are calculated to generate the solar path trajectory. Using the vertices of obstacles as a reference, a geometric model of a shadow cone along the sun's path is constructed. The target area is divided into grid cells with a preset precision. The occlusion state of each grid cell at different times is determined through geometric intersection calculations, generating a shadow analysis grid that includes occlusion probability and occlusion duration.

5. The method according to claim 1, characterized in that, For the rooftop scenario, the input component specifications, row and column spacing, tilt angle, and azimuth angle are used to initialize the arrangement as a regular array, including: Based on the input component specifications, row and column spacing, tilt angle, and azimuth angle, an initial component mesh array is generated within the roof area boundary; The initial array is checked for boundary constraints based on the preset boundary contraction distance and maintenance channel width. The positions of components that exceed the area boundary or do not meet the maintenance spacing are adjusted to form an initial layout scheme that meets the engineering constraints.

6. The method according to claim 1, characterized in that, For facade scenarios, the optimization is performed individually or collaboratively on each facade, automatically adapting component size and array spacing based on the facade slope and the height difference at the apex, and avoiding obstacles to achieve component arrangement, including: The facade is divided into segments according to the slope variation. The height difference at the top of each segment is calculated. The arrangement direction and size of the components are dynamically adjusted according to the height difference and component specifications. At the junction of adjacent segments, the horizontal and vertical spacing between components is calculated through an adaptive step distance algorithm to ensure that the edges of the components maintain a safe distance from obstacles, while achieving collaborative layout optimization across segment facades.

7. The method according to claim 1, characterized in that, The calculation of annual power generation based on regional irradiance data, component efficiency, and layout density, in conjunction with local electricity prices, subsidies, initial investment costs, and operation and maintenance costs, includes the calculation of internal rate of return, net present value, and investment payback period, including: The horizontal global irradiance data is converted into the irradiance received by the module mounting plane through the inclined plane irradiance conversion model, and the actual power generation efficiency is calculated by combining the module temperature correction coefficient and the system performance ratio. Based on the number of components and the effective solar-receiving area after the arrangement, the annual power generation is calculated; by calling local electricity prices, subsidy policies, initial investment costs and operation and maintenance costs from the preset economic parameter database, the internal rate of return, net present value and investment payback period are calculated based on the cash flow model, and the economic parameter comparison data of multiple schemes are stored.

8. A multi-scenario design system for photovoltaic systems, characterized in that, include: The scene selection unit is used to automatically parse latitude and longitude by inputting the project address, associate local or third-party irradiance and meteorological data, including horizontal global irradiance, wind speed and temperature; select the photovoltaic system application scene, which includes roof scene and facade scene; enter the project name and basic constraints; and generate a unique project ID. The outline drawing unit is used to draw the boundary or elevation outline of a region on a base map through interactive operations. The base map includes satellite imagery or bitmap. Clicking on any side of the drawn boundary and entering the measured length triggers the full map to be scaled proportionally, thereby achieving global scale uniformity. Enter the facade mode, input the elevation point by point for multiple facades, and generate a 3D model; An optimization calculation unit is used to interactively place or import obstacles, including parapet walls, fans, stairwells, ventilation shafts, windows, and protruding structures; a shadow analysis mesh is generated in real time based on the geometric deduction of the solar path; for the roof scene, the component specifications, row and column spacing, tilt angle, and azimuth angle are input, the arrangement is initialized with a regular array, and optimization calculation is performed based on the irradiance and shading cost function; For facade scenarios, each facade is optimized individually or collaboratively, and the component size and array spacing are automatically adapted according to the facade slope and the height difference of the apex, avoiding obstacles to achieve component arrangement; The report generation unit is used to calculate annual power generation based on regional irradiance data, component efficiency, and layout density, and to calculate internal rate of return, net present value, and investment payback period by linking local electricity prices, subsidies, initial investment costs, and operation and maintenance costs, and to support comparison of multiple scenarios. The model can be rotated, scaled, and screenshotted through a 3D interactive interface, generating a standardized report that includes 3D renderings, layout diagrams, technical parameter tables, and profit curves.

9. A computer device, characterized in that, The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to implement the method as described in any one of claims 1 to 7.

Images