Solar wing battery selecting and matching method and system based on three-dimensional layout

Abstract

The invention relates to the technical field of three-dimensional model processing, in particular to a solar wing battery selecting and matching method and system based on three-dimensional layout. The method comprises the following steps: acquiring three-dimensional data of a solar wing, and constructing a three-dimensional substrate model according to the three-dimensional data of the solar wing to obtain a three-dimensional model of the solar wing; performing structure-driven spreading strategy simulation on the three-dimensional model of the solar wing to obtain fixed simulation data, and performing fixed strategy simulation on the three-dimensional model of the solar wing to generate a disturbance type layout scheme to obtain random simulation data; performing regional structure domain analysis on the fixed simulation data and the random simulation data to obtain regional structure domain data, and performing optimal layout sorting on the regional structure domain data to obtain layout sorting data; and performing two-dimensional drawing export on the layout sorting data to obtain solar wing battery matching data. According to the method, the diversity of the scheme can be effectively expanded, the local adaptability is improved, and the high coverage rate and the optimal sorting output are realized under the condition of ensuring the structural constraint.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional model processing, and in particular to a solar wing battery selection method and system based on three-dimensional layout. Background Art

[0002] A solar wing structure refers to a deployable or fixed energy supply device installed on a spacecraft, satellite, or high-altitude aircraft. Its core function is to receive solar radiation through solar cells arranged on its surface and convert it into electrical energy, thereby providing a continuous power supply for the carrier platform. With the widespread application of flexible energy systems in scenarios such as aerospace, high-performance unmanned platforms, and satellite deployment, the complexity and diversity of solar wing structures continue to increase, placing higher demands on the layout efficiency, structural adaptability, and manufacturing coordination of cells. Traditional solar wing cell selection processes mostly use two-dimensional template mapping or manual rule-driven layout methods, relying on static arrangement strategies and empirical graphic design, which are difficult to accommodate the actual needs of special-shaped substrates, non-homogeneous structural areas, or dense layout of functional partitions. Summary of the Invention

[0003] In order to solve the above technical problems, the present invention proposes a solar wing battery selection method and system based on three-dimensional layout to solve at least one of the above technical problems.

[0004] This application provides a method for selecting solar wing batteries based on a three-dimensional layout, comprising the following steps:

[0005] Step S1: Acquire three-dimensional data of the solar wing, and construct a three-dimensional substrate model based on the three-dimensional data of the solar wing to obtain a three-dimensional model of the solar wing;

[0006] Step S2: performing a structure-driven deployment strategy simulation on the solar wing three-dimensional model to obtain fixed simulation data, and performing a perturbation layout scheme generation on the fixed strategy simulation on the solar wing three-dimensional model to obtain random simulation data;

[0007] Step S3: performing regional structure domain analysis on the fixed simulation data and the random simulation data to obtain regional structure domain data, and performing optimal layout sorting on the regional structure domain data to obtain layout sorting data;

[0008] Step S4: Export the layout sorting data into a two-dimensional drawing to obtain solar wing battery selection data.

[0009] In the present invention, step S1 achieves a high degree of restoration of the actual structural form by acquiring and constructing a three-dimensional model of the solar wing, providing an accurate geometric basis for subsequent layout simulation; step S2 introduces a perturbation layout generation mechanism based on fixed strategy simulation, effectively expanding the diversity and robustness of the layout scheme and avoiding falling into local optimality; step S3 jointly models the global and local characteristics of the layout scheme through regional structure domain analysis, improving the fine-grained discrimination ability of layout sorting; finally, step S4 automatically generates a two-dimensional drawing based on the three-dimensional simulation results, ensuring a seamless connection between three-dimensional optimization and production process. Compared with traditional two-dimensional manual layout or rule templates, this method can realize intelligent layout, data evaluation and output standardization under complex and irregular substrates, has higher layout adaptability, visual interactivity and manufacturing friendliness, and significantly improves the efficiency of solar wing battery selection and assembly quality.

[0010] Preferably, step S1 is specifically:

[0011] Obtain three-dimensional data of solar panels;

[0012] Perform heterogeneous 3D structure perception based on solar wing 3D data to obtain heterogeneous 3D structure data;

[0013] Perform layout adaptation domain perception on heterogeneous three-dimensional structure data to obtain layout adaptation domain data;

[0014] The base plate model is generated according to the layout adaptation domain data to obtain a three-dimensional model of the solar wing.

[0015] This step in the present invention not only obtains basic geometric data, but also deeply analyzes the heterogeneous features such as edge grooves, compression points, and hole distribution in the solar wing through heterogeneous three-dimensional structure perception technology, breaking through the limitation of traditional modeling that only constructs a unified grid model based on contour lines; through the layout adaptation domain perception algorithm, it can automatically identify and mark the spatial areas and prohibited areas suitable for battery cell layout, significantly improving the model's adaptability to actual assembly constraints; the generated three-dimensional substrate model has structural layering and regional adaptability, and can directly support high-precision layout simulation and strategy matching, avoiding manual division errors and improper rule configuration problems.

[0016] Preferably, the heterogeneous three-dimensional structure perception is specifically:

[0017] Perform heterogeneous data analysis based on the three-dimensional data of the solar wing to obtain heterogeneous data;

[0018] Perform structural semantic mapping on heterogeneous data to obtain structural semantic data;

[0019] A three-dimensional structure expression body is generated according to the structural semantic data to obtain heterogeneous three-dimensional structure data.

[0020] The present invention breaks the dependence on a single format or rule modeling data, is compatible with multi-source data input (such as CAD models, point clouds, scan layers, etc.), and is uniformly converted into structurally recognizable geometric elements through heterogeneous data analysis; based on structural semantic mapping technology, a mapping relationship is established between geometric elements and functional semantics (such as "compression points", "perforation areas", "surface corners", etc.), thereby giving the model a logical structure label; through the generation of three-dimensional structural expressions, the model not only has a geometric shape, but also carries high-order semantic information such as structural use and assembly restrictions. The present invention has extremely high scalability and interpretability, providing a solid foundation for layout strategy decision-making and regional routable analysis. Compared with traditional methods that only rely on geometric coordinate points or boundary lines for modeling, this technology has obvious differentiated advantages in structural understanding depth, semantic differentiation ability and automatic analysis accuracy, and is suitable for solar wing component design tasks with non-standard configurations or multi-functional integration.

[0021] Preferably, the deployment of the adaptation domain perception is specifically as follows:

[0022] Calculate the surface geometric paving ability based on heterogeneous 3D structural data to obtain preliminary geometric paving domain data;

[0023] Eliminate the structural functional area of the preliminary geometrically paved domain data to obtain functional mask data;

[0024] Perform semantic mask fusion based on functional mask data and heterogeneous 3D structure data to obtain semantically enhanced layout domain data;

[0025] The semantically enhanced layout domain data is screened for regional connectivity to obtain the layout adaptation domain data.

[0026] The present invention performs surface geometric paving feasibility calculation based on heterogeneous three-dimensional structural data, which can comprehensively consider geometric features such as local curvature, normal change, and unevenness, identify areas with actual paving feasibility, and avoid false paving planning caused by surface undulations in traditional methods; introduces a structural functional area elimination operation to effectively shield non-paveable areas with assembly / electrical functions such as compression points, bolt holes, and cable channels, thereby improving paving safety and functional compatibility; the semantic mask fusion step integrates functional semantics with geometric information to construct a paving layer with paving priority and risk level, thereby enhancing the system's perception of paving strategies for different structural areas; through regional connectivity screening, it ensures that the selected paving area is spatially coherent and has wiring accessibility, thereby avoiding isolated, broken, or unconnected paving configurations.

[0027] Preferably, the structure-driven spreading strategy simulation is specifically as follows:

[0028] Constructing a structural constraint diagram for the three-dimensional model of the solar wing to obtain structural constraint diagram data;

[0029] Performing paving behavior graph scheduling on the structural constraint graph data to obtain paving behavior graph data, wherein the paving behavior graph scheduling includes a horizontal filling strategy, a vertical filling strategy, a central symmetric filling strategy, and an S-shaped filling strategy;

[0030] The legitimacy of the laying behavior diagram data is verified to obtain fixed simulation data.

[0031] The present invention constructs a structural constraint graph, which can fully express multi-source restriction information such as geometric boundaries, local restrictions, functional zoning and prohibited areas in the three-dimensional structure of the solar wing, forming a unified structural constraint semantic map; through the laying behavior graph scheduling mechanism, multi-strategy graphic models such as horizontal filling, vertical filling, central symmetry and S-type are introduced to effectively take into account different structural forms and layout requirements, and realize highly adaptable graphic laying driven by strategy; the laying behavior graph, as a graph theory structure, has the advantages of high visualization, node coherence and traceability of behavior sequence, so that the strategy generation is not only regular, but also flexible in combination and optimization; through the legality verification operation, the layout generated by each strategy is subjected to collision detection, boundary verification and electrical accessibility check to ensure that each layout has physical feasibility and process feasibility.

[0032] Preferably, the perturbation layout solution is generated as follows:

[0033] Performing local graph partitioning on the structural constraint graph data to obtain local graph data;

[0034] Performing perturbation graph selection on the local graph data to obtain perturbation graph data;

[0035] Performing disturbance space selection on the paving behavior graph data according to the disturbance graph data to obtain disturbance space data;

[0036] Perform disturbance processing according to the disturbance space data to obtain disturbance result data;

[0037] The legitimacy of the disturbance result data is verified to obtain random simulation data.

[0038] The present invention performs local graph division based on the structural constraint graph. It can focus on key areas such as edge areas, corner areas or functional transition areas while maintaining the semantic consistency of the overall structure, perform targeted disturbance operations, and enhance the layout flexibility of the local structure. The disturbance graph selection mechanism enables the disturbance strategy to have graph-theoretic structural control capabilities, and can select the intervention range based on node importance, edge connectivity or layout density to avoid blind disturbance causing structural damage or local disorder. Through disturbance space selection, graphic disturbance operations such as spatial local replacement, rotation, offset, and dislocation are performed on the laying behavior graph to generate a controllable variant structural layout. After the disturbance processing, legality verification is performed to ensure that the random layout generated by the disturbance still meets the structural constraints, electrical accessibility and manufacturing rules, thereby ensuring practicality.

[0039] Preferably, the regional structure analysis is specifically as follows:

[0040] Constructing structure domain response tensors for fixed simulation data and random simulation data to obtain structure domain response tensor data;

[0041] Calculate the coverage rate of the routable area based on the structural domain response tensor data to obtain the coverage rate data of the routable area;

[0042] Calculating the local symmetry offset according to the structural domain response tensor data to obtain the local symmetry offset data;

[0043] Perform region label inference based on the deployable region coverage data and local symmetry offset data to obtain region label data;

[0044] According to the regional label data, the fixed simulation data and the random simulation data are sorted by domains to construct candidate maps, and the regional structure domain data is obtained.

[0045] In the present invention, by constructing a structural domain response tensor, it not only integrates the multi-dimensional spatial information of cloth distribution, structural constraints and paving adaptability in simulation data, but also provides a unified tensor expression framework for indicator calculation; by calculating the coverage rate of the paving area through the tensor, the actual utilization rate of the cloth in each domain can be accurately measured, reflecting the space filling effect; the local symmetry offset calculation can identify the degree of deviation of the cloth in the left-right, top-down or center-symmetrical structure, and effectively capture the local structural beauty and cloth balance; based on the above two types of key indicators, the system further performs regional label inference, and assigns semantic labels such as high coverage and high symmetry, high coverage and low symmetry, and low coverage and high symmetry to different layout domains to form structured regional portraits; these regional labels drive the construction and sorting of candidate maps of fixed and perturbed simulation data, and generate regional structural domain data that are both globally and locally optimal.

[0046] Preferably, the preferred layout order is specifically:

[0047] The regional weight tensor data is fused and constructed on the regional structure domain data to obtain the regional weight tensor data;

[0048] Perform nonlinear mapping and sorting on the regional weight tensor data to obtain preliminary layout sorting data;

[0049] The preliminary layout ranking data is enhanced with semantic annotation of candidate solutions to obtain layout ranking data.

[0050] In the present invention, the system constructs a regional weight tensor for the regional structure domain data, takes into account multiple spatial semantic indicators such as the layout rate, symmetry offset, connection accessibility and layout continuity of each region, realizes weighted aggregation of information in the tensor structure, and breaks the one-sidedness of the traditional single-indicator scoring model; by introducing a nonlinear mapping sorting algorithm, it adopts softmax normalization, Tanh compression or sorting perception network to dynamically transform the weights and sort the scores of each candidate solution, making the sorting process less sensitive to extreme values and the ranking results more stable and balanced; the system combines the regional structure semantic labels to enhance the semantic annotation of the candidate solutions in the sorting results, so that each layout result not only has a score, but also contains semantic descriptions such as "boundary balance priority", "center filling intensive", and "structural symmetry optimization", which is convenient for human understanding and interactive screening.

[0051] Preferably, step S4 is specifically:

[0052] Performing three-dimensional layout projection according to the layout sorting data to obtain three-dimensional layout projection data;

[0053] Perform surface expansion according to the three-dimensional layout projection data to obtain surface expansion data;

[0054] Generate drawing elements according to the surface expansion data to obtain drawing element data;

[0055] The drawing format is exported according to the drawing element data to obtain the solar wing battery selection data.

[0056] The system in the present invention performs three-dimensional layout projection operations based on layout sorting data, which can accurately preserve the spatial topological relationship of solar wing fabrics on curved surfaces or special-shaped structures, and project them onto the two-dimensional processing reference surface to ensure that the drawing expression is consistent with the actual assembly; through the patch unfolding operation, the fabric area with spatial curvature is geometrically unfolded, and the two-dimensional deformation errors caused by bending, corners or multiple curvature interference are automatically solved, thereby improving the consistency between the drawing and the physical substrate; the system generates drawing elements including numbers, boundary lines, docking ports, annotation symbols, etc. according to the unfolded patch data, automatically completes the embedding of standardized engineering graphics, and avoids errors and omissions in the manual drawing process; through the drawing format export module, the drawing element data can be quickly output to industrial common formats such as DWG and DXF, supporting direct calls by downstream CAD, electrical wiring and manufacturing systems.

[0057] Preferably, the present application further provides a solar wing battery selection system based on a three-dimensional layout, which is used to execute the solar wing battery selection method based on a three-dimensional layout as described above. The solar wing battery selection system based on a three-dimensional layout includes:

[0058] A three-dimensional substrate modeling module is used to obtain three-dimensional data of the solar wing and construct a three-dimensional substrate model based on the three-dimensional data of the solar wing to obtain a three-dimensional model of the solar wing;

[0059] A layout simulation generation module is used to simulate the structure-driven deployment strategy of the solar wing three-dimensional model to obtain fixed simulation data, and to generate a perturbation layout scheme by performing fixed strategy simulation on the solar wing three-dimensional model to obtain random simulation data;

[0060] The structural region analysis and optimal sorting module is used to perform regional structural domain analysis on fixed simulation data and random simulation data to obtain regional structural domain data, and to perform optimal layout sorting on the regional structural domain data to obtain layout sorting data;

[0061] The drawing generation and matching output module is used to export the layout sorting data into two-dimensional drawings and obtain the solar wing battery matching data.

[0062] The present invention offers the following advantages: During the data modeling phase, heterogeneous parsing and semantic recognition of 3D solar wing data are performed to construct a 3D representation model that can express local functional constraints, such as compression points, hole areas, and boundary interferences. This addresses the difficulty of traditional methods in accurately perceiving complex boundaries and functional areas. Furthermore, through layout adaptation domain awareness, the system automatically identifies paving areas and generates semantically enhanced layout masks, providing high-precision support for subsequent simulation optimization. During the simulation phase, a structural constraint graph is constructed and multiple layout behavior graphs are scheduled to generate fixed simulation samples that conform to physical constraints. A local variation mechanism is introduced through the perturbation strategy generation module, effectively expanding the layout solution space. During the optimization phase, a structural domain response tensor is constructed, integrating coverage and symmetry features. Combined with regional label inference and weight tensor sorting, this improves optimal interpretation and multi-objective adaptability. The system automatically performs 3D projection, mesh expansion, and drawing generation, creating standard drawing elements and supporting export to industrial formats. This enables a closed-loop selection process from modeling to manufacturing, offering the technical advantages of strong structural adaptability and intelligent evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0064] Figure 1 A flowchart showing the steps of a solar wing battery selection method based on a three-dimensional layout according to an embodiment is shown;

[0065] Figure 2 A flowchart showing the steps of a three-dimensional substrate modeling method according to an embodiment is shown;

[0066] Figure 3A flowchart showing the steps of a structure-driven spreading strategy simulation method according to an embodiment is shown;

[0067] Figure 4 A flowchart showing the steps of a method for generating a perturbation-type layout solution according to an embodiment is shown;

[0068] Figure 5 A flowchart showing the steps of a drawing generation and matching output method according to an embodiment is shown. DETAILED DESCRIPTION

[0069] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative work are within the scope of protection of the present invention.

[0070] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0071] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0072] See also Figures 1 to 5 , this application provides a solar wing battery selection method based on three-dimensional layout, comprising the following steps:

[0073] Step S1: Acquire three-dimensional data of the solar wing, and construct a three-dimensional substrate model based on the three-dimensional data of the solar wing to obtain a three-dimensional model of the solar wing;

[0074] In one embodiment, the three-dimensional structural data of the solar wing is obtained. The three-dimensional data can be obtained in two ways: first, a structured light scanner or a laser radar (LiDAR) system is used to perform high-precision point cloud scanning on the physical solar wing to obtain its spatial geometric shape; second, three-dimensional model data is extracted from an existing computer-aided design (CAD) model. The model format may include standard .stp (STEP), .igs (IGES) and other commonly used industrial file formats. A three-dimensional substrate model of the solar wing is constructed based on the point cloud data. In this embodiment, it is preferred to use a Poisson surface reconstruction algorithm or an α-Shape method to generate a continuous surface mesh, wherein the former is suitable for smooth surface modeling of dense point cloud data, and the latter is suitable for structural recovery scenarios with boundary preservation requirements. After the construction is completed, the main load-bearing structural surface of the solar wing is further extracted, and a unified reference coordinate system is established based on its geometric center and laying direction, thereby generating a normalized, structurally complete three-dimensional geometric substrate model.

[0075] Step S2: performing a structure-driven deployment strategy simulation on the solar wing three-dimensional model to obtain fixed simulation data, and performing a perturbation layout scheme generation on the fixed strategy simulation on the solar wing three-dimensional model to obtain random simulation data;

[0076] In one embodiment, the generated three-dimensional substrate model is input into a finite element simulation platform, such as ANSYS, ABAQUS, or other commercial simulation tools that support multibody dynamics and nonlinear material properties. During the simulation setup, the solar wing's deployment starting point and deployment direction are set, gravity load parameters are appropriately applied, and the stiffness coefficients of each structural unit are set. Based on the actual structural characteristics of the solar wing's span, key hinge points in the mechanism's motion are defined, and the connection and activation sequences between nodes are set to reflect the dynamic behavior of the solar wing's gradual deployment. During the simulation, the system solves the structural evolution in a time-stepping manner, outputting data such as configuration changes, node stress distributions, displacement vector fields, and deployment progress status at each time step. These simulation results, collectively referred to as fixed simulation data, can be used to describe standard deployment behavior under ideal structural control conditions. A perturbation-based layout solution generation process is then executed. This process, based on the fixed strategy simulation, considers non-ideal perturbations during the deployment process and conducts an extended analysis of deployment stability and layout robustness. Specifically, micro-perturbation parameters are introduced at the fixed simulation boundary or the initial state of the node, including the displacement perturbation δx with a value range of ±2 mm and the angle perturbation δθ with a value range of ±3 degrees. A multi-round random simulation method (such as the Monte Carlo simulation strategy) is used to construct multiple perturbation input samples, and the same dynamic spreading process as the fixed strategy is performed. In each round of perturbation simulation, the system outputs the configuration evolution path after the perturbation and calculates whether it meets the structural stability constraints (such as stress limit, spreading integrity, node non-overlap, etc.). Only the perturbation samples that meet the stability requirements are retained as random simulation data, and key evaluation indicators such as the actual layout position, energy dissipation value, and response peak of each sample are recorded.

[0077] Step S3: performing regional structure domain analysis on the fixed simulation data and the random simulation data to obtain regional structure domain data, and performing optimal layout sorting on the regional structure domain data to obtain layout sorting data;

[0078] In one embodiment, a spatial partitioning method is used to divide the surface of a three-dimensional model into several regional units. The partitioning method can be an octree partitioning method or an equal voxel grid partitioning strategy. For each divided area, the system calculates the following structural evaluation indicators, such as the load uniformity coefficient, which is defined as the sum of the stress values of all nodes in the area divided by the area volume, that is, where μ i is the load uniformity coefficient of the i-th regional unit, j is the node order term, the upper limit is the node number data, σ j is the equivalent stress value at the jth node in the region, V i is the volume or area of the regional unit to which the jth node belongs. The threshold value for calculating structural deformation is defined as the maximum difference between the displacement offset value of all nodes in the region and the reference displacement in the initial state, i.e. Δi =max(|d ij ―d i0 |), Δ i is the structural deformation offset threshold of the i-th regional unit, max is the maximum value function, d ij is the displacement value of the jth node in the region in the current spreading state, d i0 The reference displacement value in the initial state. According to the above two indicators, the regional screening rules are set. If the load uniformity coefficient μ of a certain area is i Higher than the set stress mean threshold, and its structural deformation Δ i If the deformation limit is less than the preset maximum allowable deformation limit, the area is determined to be a suitable area. The ranking index of the suitable area is calculated, S i =w1·μ i +w2·(1 / Δ i )+w3·A i , where S i is the ranking index of the suitable distribution area, w1 is the weight data of the uniformity coefficient data, which is 0.4, μ i is the load uniformity coefficient data, w2 is the weight data of the load uniformity coefficient data, and its value is 0.3, Δ i is the structural deformation data, w3 is the weight data of the effective paving area data of the area, and its value is 0.3, A i is the effective paving area data of the region, and the weight data is obtained by fitting the empirical data. i Sort in descending order to form a layout optimization sequence.

[0079] Step S4: Export the layout sorting data into a two-dimensional drawing to obtain solar wing battery selection data.

[0080] In one embodiment, each preferred area in the three-dimensional layout sorting data is mapped to a unified drawing reference plane, and projection modeling is performed to form a series of structured drawing views, including but not limited to the following: a grid wiring diagram showing the grid boundaries, connection relationships, and spatial alignment relationships of the battery cell layout units in each area; a wiring diagram indicating the electrical connection method between battery modules, such as series or parallel logic and wire direction; and an installation point coordinate diagram marking the specific assembly coordinate position of each battery cell module or component to facilitate processing positioning and on-site installation. The above drawing files can be exported to standard engineering file formats, including .dwg, .dxf, .svg, etc., supporting seamless integration with process planning systems and production manufacturing systems. Generate solar wing battery selection data. For each preferred layout area, the system automatically matches the most suitable PV module type based on its structural parameters, deformation tolerance, and area size. It then generates a set of structured selection parameters, including the area number, which uniquely identifies the layout area; the area's centroid coordinates, which use three-dimensional coordinates [x, y, z] to represent the area's spatial location within the substrate model; the effective layout area, which represents the actual surface area available for cell layout (in square meters); the deformation tolerance, which represents the maximum structural displacement of the area when laid (in millimeters); the matching cell type, such as the "PERC-158×158-72" high-efficiency crystalline silicon module; and the expected power value, estimated in watts based on the module's power density and the layout area. After summarizing this data, the system outputs a recommended cell selection table containing all the selected areas and simultaneously generates a corresponding configuration atlas to assist the project implementation unit with subsequent process review, electrical layout, and component procurement.

[0081] Preferably, step S1 is specifically:

[0082] Step S11: Acquire three-dimensional data of the solar wing;

[0083] In one embodiment, the three-dimensional model data can be directly extracted from the existing engineering design drawings, and the drawing formats include but are not limited to common CAD file formats such as IGES, STEP and STL. Alternatively, a three-dimensional modeling operation is performed on the physical solar wing sample, preferably using a laser scanner (such as FARO Focus) or a structured light scanner (such as Artec Leo) for spatial data acquisition to obtain high-precision point cloud data, covering the surface contours and key structural details of the solar wing. If an image-based three-dimensional reconstruction solution is adopted, image sequences of the solar wing at different angles can be collected, and the structural self-motion algorithm can be executed in sequence to estimate the camera pose and sparse three-dimensional structure; a multi-view stereo reconstruction algorithm is used to obtain dense point cloud data to achieve complete three-dimensional reconstruction.

[0084] Step S12: performing heterogeneous three-dimensional structure perception based on the solar wing three-dimensional data to obtain heterogeneous three-dimensional structure data;

[0085] In one embodiment, the heterogeneous three-dimensional structure refers to a composite structural unit composed of multiple different materials within the overall solar wing structure. These typically include heterogeneous components such as a carbon fiber reinforced plastic (CFRP) substrate, flexible photovoltaic cells, and an aluminum alloy support frame. These components exhibit significant differences in geometric characteristics, reflective properties, and spectral response. The system integrates multi-source sensors to collect material features from each point cloud segment within the solar wing's three-dimensional model. These features primarily include RGB color reflectance, used to identify differences in surface color or coating; laser reflection intensity, reflecting the laser reflectivity of different materials; and infrared response characteristics, used to identify areas with varying thermal radiation, such as flexible photovoltaic cells. Based on these collected multimodal point cloud attributes, feature vectors are extracted, including color distribution, texture characteristics, and surface normal direction. Subsequently, a clustering algorithm (such as K-Means or the density-based DBSCAN algorithm) is used to classify the point cloud data into material categories, identifying heterogeneous material regions and completing coarse-grained material partitioning. For heterogeneous regions identified in the material classification results, their geometric features are identified. For example, the flexible battery cells in solar wings are often rectangular or curved sheets, while the connecting structures are elliptical rods or conical shafts. To achieve this recognition process, the system uses edge contour-based geometric modeling algorithms, such as RANSAC plane fitting or cylindrical surface fitting methods, to perform high-confidence modeling and segmentation of typical geometric fragments in point cloud data. A structural annotation set is established for each identified structural unit. The annotation content includes but is not limited to the following elements: component type (such as "main wing skeleton", "connecting pin"); material (such as "carbon fiber reinforced composite material", "aluminum alloy"); geometric contour type (such as three-dimensional cuboid, three-dimensional cylinder); spatial coordinate range, that is, the three-dimensional bounding box range occupied by the component in a unified coordinate system.

[0086] Step S13: performing layout adaptation domain perception on the heterogeneous three-dimensional structure data to obtain layout adaptation domain data;

[0087] In one embodiment, the layout adaptation domain refers to the effective area on the surface of the solar wing that meets the requirements for laying out the solar cells. This area needs to meet multiple layout conditions such as geometric flatness, light adaptability and laying feasibility judgment, so as to ensure the high reliability deployment of photovoltaic cells in terms of spatial structure and energy efficiency. The geometric flatness analysis is specifically that the system divides the surface of the solar wing into multiple local small units (triangular grids or equidistant sampling surface elements), and calculates the angle θ between the normal vector of each small unit and the reference normal vector of the overall solar wing span. If the normal deviation angle θ of a small unit is less than a preset threshold (for example, 10°), the unit is judged to be a flat area and meets the basic layout stability requirements. The laying feasibility judgment is to further perform local meshing processing (such as Delaunay triangulation) on the above-mentioned flat area, and calculate the surface average curvature κ on each grid segment to evaluate whether the area has excessive undulation or deformation. When the average curvature κ of a certain area is less than or equal to the preset threshold (for example, 0.05m ―1 ), it can be determined that the area has sufficient geometric flatness and structural ductility, and can be included in the layout adaptation domain as a feasible layout area. Under the set light source direction (such as the typical solar incidence angle), the system calculates the angle α between this direction and the surface normal vector of each unit to evaluate the efficiency of the area receiving light. If the incident angle α of a unit is less than the preset light adaptation threshold (for example, 45°), the area is identified as a high-efficiency lighting area, which is suitable for the priority layout of photovoltaic units to improve energy utilization. Based on the joint judgment of the above three dimensions, the system marks the set of layout adaptation domains that meet the layout flatness, deformation tolerance and lighting efficiency.

[0088] Step S14: Generate a base plate model according to the layout adaptation domain data to obtain a three-dimensional model of the solar wing.

[0089] In one embodiment, all local geometric fragments determined as routable areas in the layout adaptation domain are fused and spliced to reconstruct an integrated substrate structure. This aggregation process enables the spatial position relationship and edge matching relationship of each region in a unified reference coordinate system to ensure the geometric consistency of the generated model in the macro structure. For boundary discontinuities or gaps that exist in the region splicing process, a repair strategy based on boundary curve interpolation is adopted, which specifically includes a geometric stretching smoothing method to connect regions with similar curvatures; Bezier surface interpolation technology is used to insert continuous surface fragments at boundary transitions to improve the surface smoothness and structural integrity of the substrate. Based on the generated continuous substrate geometry, a topological structure expression is constructed, for example, a "node-edge-face" ternary topological network is constructed using a Half-Edge data structure. This topological structure facilitates efficient patch operations, local editing, and attribute binding. For each patch area in the topological mesh, its key attribute information is annotated, including the material type (such as flexible battery layer, support base, etc.); layout level (for example, divided into high, medium and low levels according to lighting adaptability or laying priority); local deformation constraints (such as thermal expansion compensation area, edge tensioning area, etc.). According to the spatial distribution density of the layout adaptation domain, the system performs mesh refinement on the three-dimensional model. Specifically, in areas with high layout density and complex structural features, the mesh resolution is improved to enhance the model's expressiveness; in areas with flat structures, the mesh density can be appropriately reduced to optimize the model volume and computational efficiency. Output standard three-dimensional files (such as .obj, .stl) or program-parseable formats (such as .vtk, .glTF).

[0090] Preferably, the heterogeneous three-dimensional structure perception is specifically:

[0091] Perform heterogeneous data analysis based on the three-dimensional data of the solar wing to obtain heterogeneous data;

[0092] In one embodiment, the system performs heterogeneous attribute analysis on three-dimensional solar wing data to extract multi-source structural information. The 3D data can be in industry-standard formats such as .STL, IGES, and STEP, or in the form of a custom mesh model containing metadata such as material identification, assembly information, and naming conventions. Its structure can take the form of an attribute tree in a CAD system. For geometric structure analysis, the system extracts the model's mesh patches or voxel elements (such as B-rep) and constructs a spatial index structure based on their spatial coordinates to support accurate identification of local features and neighbor relationships. For material attribute extraction, if the model contains color coding or explicit material labels, the corresponding material type (e.g., aluminum alloy, carbon fiber reinforced composite material, etc.) is directly recorded. If label information is lacking, a preliminary material type determination is made using density parameters, reflectivity characteristics, and thickness data (based on a knowledge engine built from historical empirical data). A machine learning classification model (deep learning training based on historical data and corresponding labels) can also be introduced to enhance material recognition accuracy. For assembly hierarchy extraction, the system reads the assembly structure description information in the model and constructs a component hierarchy, including the main wing assembly, secondary components, and connecting brackets. This structure is expressed in a tree or graph format, supporting structural function dependency analysis and layout logic modeling.

[0093] Perform structural semantic mapping on heterogeneous data to obtain structural semantic data;

[0094] In one embodiment, the system constructs a structural semantic mapping model based on heterogeneous structural data to generate structural semantic data. This process aims to assign semantic labels to each three-dimensional structural component, thereby realizing the knowledge-based expression of the structural unit at the functional and material levels. The semantic label system includes the following three dimensions: (1) structural role labels, which are used to identify the functional positioning of the component in the overall structure, such as the main load-bearing component, support frame, hinged end, battery embedding area, etc.; (2) functional area feature labels, which are used to describe the functional attributes of the structural segment, such as the expansion area, fixed area and adjustable area; (3) material semantic labels, which are used to reflect the physical properties of the structural unit, such as lightweight and high-rigidity area, flexible area and shielding area, etc. In the label assignment process, the system can adopt two semantic mapping methods. The first is a method based on expert knowledge rules, which maps component attributes to semantic labels according to predefined conditional rules. For example, if a component is made of carbon fiber reinforced composite material and its main side length exceeds 500 mm, the component can be labeled as the main load-bearing component. The second type is a classification method based on machine learning. By training historically labeled data, a structural component recognition model (such as a model based on a random forest, graph convolutional network, or support vector machine) is constructed. The model input includes the component's geometric features (such as aspect ratio, thickness), material type, and its spatial position relationship. The output is a probability distribution of each semantic label. The system selects the label with the highest confidence as the semantic classification result for the component.

[0095] A three-dimensional structure expression body is generated according to the structural semantic data to obtain heterogeneous three-dimensional structure data.

[0096] In one embodiment, the system constructs a structural representation based on structural semantic data and a three-dimensional mesh model, thereby generating a three-dimensional structural representation model with heterogeneous semantics, spatial boundaries, and physical properties, referred to as a structural representation. A structural representation refers to a geometric unit with clear spatial boundaries, structural semantic labels, and behavioral attributes, which can serve as an input unit for tasks such as layout calculation, collision detection, and path simulation. During the construction process, the system establishes a spatial representation object for each structural component. Specifically, it includes a local coordinate system to describe the relative spatial position and posture of the component within the overall system; a bounding volume structure, such as a minimum circumscribed rectangular box, convex hull, or composite, to describe its physical boundaries; and contact surface information, which records the boundary surfaces that contact or interfere with other structural units to support collision detection and connection modeling. The system binds the geometric information of each component to its corresponding semantic label and physical characteristic parameters to form a complete structural representation unit. This unit can be abstracted as a structural triple, represented as [geometric entity, semantic label, physical attribute]. Semantic labels include structural role, functional area type, and material semantics, while physical attributes include density, stiffness coefficient, thermal expansion characteristics, etc. During the relationship modeling phase, the system determines structural connectivity based on spatial proximity. If the length of the shared edge region between two entities exceeds a preset ratio (e.g., 5%) of their boundary length, a connecting edge is added to the structural connectivity graph. This structural connectivity graph represents the spatial coupling between the entities.

[0097] Preferably, the deployment of the adaptation domain perception is specifically as follows:

[0098] Calculate the surface geometric paving ability based on heterogeneous 3D structural data to obtain preliminary geometric paving domain data;

[0099] In one embodiment, a geometric paving feasibility analysis of the solar wing surface is performed based on heterogeneous 3D structural data to identify areas where structural paving is feasible and generate preliminary geometric paving domain data. This process includes the following specific steps: the 3D surface model of the solar wing is subjected to facet processing. Preferably, a triangular meshing algorithm is used to divide the entire surface into facets, forming a facet set T = t1, t2, ..., t n , where each triangle t i Considered as the basic paving analysis unit. For each patch t i Extract geometric paving index, including the following three key parameters, such as calculating the normal deviation angle θ i , which is used to measure the angle between the normal direction of the patch and the main deployment direction of the entire solar wing. It is calculated as follows: is the normal direction of the film, is the overall main direction. Calculate the local curvature κ i, used to evaluate the degree of geometric curvature of a patch within its neighborhood. It is preferred to use the rate of change of the angle between the normal vectors of adjacent patches to approximate the Gaussian curvature or mean curvature value. A smaller curvature value indicates that the surface of the area is relatively flat and is conducive to paving. Calculate the standard deviation of the patch area σ a , used to evaluate the uniformity of local mesh division. By calculating the standard deviation of the area of all facets within a certain neighborhood around each facet, if the standard deviation is small, it means that the mesh density in the area is uniform, which is beneficial for layout control and manufacturing accuracy. If: -θ i <10° (small directional deviation), ―κ i <0.05 (locally relatively flat), ―σ a <20 (uniform grid), the triangle is marked as preliminarily paving-ready.

[0100] Eliminate the structural functional area of the preliminary geometrically paved domain data to obtain functional mask data;

[0101] In one embodiment, the goal of this process is to identify and eliminate structural and functional areas on the solar wing surface that are unsuitable for photovoltaic cell placement, such as connector mounting areas, hinge areas, locking mechanism areas, and access holes. This identification method is based on the structural annotation information extracted from the heterogeneous 3D structural data in the previous step. This information includes the structural type label, geometric outline description, and spatial boundaries of each component in the 3D model. In practice, the system first extracts labeled areas of structural components with functional attributes, including but not limited to "connection areas," "hinge areas," "locking mechanism areas," and "access holes." Each type of functional area corresponds to a structural unit with a clear physical meaning and layout constraints in 3D space. Based on the geometric envelope information of these functional components, the system constructs a projection of the corresponding functional area in the model surface coordinate system. Preferably, the 3D boundaries are mapped to a 2D surface mesh using axis-aligned bounding boxes or polygonal projection outlines, resulting in precise coverage of the original layout area. Based on this, the system constructs a binary functional mask matrix corresponding to the solar wing surface. In this mask matrix, each surface unit location (such as a triangle or grid node) is assigned a binary status flag based on whether it is located in a functional area: if the location belongs to a non-paveable functional area, the corresponding mask value is set to 0, indicating that battery deployment is not allowed at this location; if the location is not covered by the functional area, the corresponding mask value is set to 1, indicating that the location can continue to remain in the paveable domain. The output functional mask data is used to screen the preliminary geometric paveable domain point by point, thereby retaining only the high-availability areas that are structurally deployable.

[0102] Perform semantic mask fusion based on functional mask data and heterogeneous 3D structure data to obtain semantically enhanced layout domain data;

[0103] In one embodiment, multiple semantic enhancement variables are assigned to each surface triangular facet unit in the three-dimensional model, including but not limited to the following types, such as material type, such as fiberglass, flexible copper base, etc.; thermal deformation level, which is divided into three levels of thermal stability: high, medium, and low according to the thermal stress response results of the region in the spreading simulation; electromagnetic shielding area mark, which is used to identify whether it is a high electromagnetic interference area and it is not recommended to place sensitive electronic devices such as photovoltaic cell modules. Based on the introduction of the above semantic variables, the system constructs a fusion function for each facet to fuse the geometric adaptability evaluation results with the semantic credibility results, such as S i =w g GS i +w s SS i , where S i is the fusion index data, w g The weight for the geometric score is 0.7, GS i is the geometric paving score (normalized by the scores of the first two steps), w s The semantic score weight is 0.3, SS i The system then assigns semantic relevance scores (e.g., avoiding high-heat areas and electromagnetic interference zones). After the fusion index is calculated, a unified threshold is set (e.g., a score greater than 0.6). Patches with fusion indexes above this threshold are considered semantically enhanced patches. The system then aggregates all patches that meet these criteria to generate semantically enhanced patch domain data.

[0104] The semantically enhanced layout domain data is screened for regional connectivity to obtain the layout adaptation domain data.

[0105] In one embodiment, a connected domain extraction operation is performed. The triangular facets of the layout surface are used as graph nodes. If two facets have a common edge relationship, that is, they share a boundary edge, a connecting edge is established between the nodes. The connection relationship graph thus constructed reflects the topological adjacency of the paving area in the geometric space. After the graph structure is constructed, a standard graph traversal algorithm (such as breadth-first traversal BFS or depth-first traversal DFS) is used to identify the connectivity of the above connection graph, and all structurally connected sub-areas are extracted, which are recorded as a connected sub-graph set C = c1, c2, ..., c m , where each subgraph c j Represents a local continuous layout area. For each connected subgraph c j Calculate the following three key geometric indicators, including the effective area A j , which represents the total area of the connected area that can be used for layout, in square meters; the maximum connected diameter D j , represents the maximum Euclidean distance between any two patches in the connected area, used to evaluate the farthest boundary length of the paved path, in millimeters; boundary complexity Bj , defined as the ratio of the total length of the outer contour line of the area to the area of the region, reflects whether the shape of the region edge is too complex, without units. The conditions for retention are A j > Amin (the minimum layable area threshold, such as 0.05 m²); D j > Dmin (the minimum connected distance threshold, such as 150 mm); B j < Bmax (the maximum allowable complexity threshold, such as 1.8).

[0106] Preferably, the structure-driven spreading strategy simulation is specifically as follows:

[0107] Step S21: Construct a structure constraint diagram for the three-dimensional model of the solar wing to obtain structure constraint diagram data;

[0108] In one embodiment, the key structure points in the three-dimensional substrate model are defined as the nodes of the graph, mainly including grid points (i.e., the vertices of the surface triangular grid or polygon grid); regional seam points (the connection edge points of different structural materials or modules); the end points of the rotating shaft or the moving joint points. Each node v i contains coordinates and structural attributes (such as "fixed point", "edge point", "rotatable point"). The construction edges between adjacent structural patches establish graph edges e ij , and each edge represents the structural constraint path between nodes. The types of structural edges are divided into rigid edges (non-spreadable areas); hinged edges (with angular constraints); transitional edges (allowing laying changes); and the associated parameters for each edge are the maximum spreading angle θ max , the allowable deformation amount δ max , and the connection type.

[0109] Step S22: Schedule the laying behavior graph for the structure constraint graph data to obtain laying behavior graph data, where the laying behavior graph scheduling includes a horizontal filling strategy, a vertical filling strategy, a central symmetry filling strategy, and an S-shaped filling strategy;

[0110] In one embodiment, each node represents a "laying unit"; the node contains attributes, such as the current state (not laid / laying in progress / completed), activation conditions (such as "the number of adjacent paved areas reaches 3 sides"); the horizontal filling strategy (progress in the X-axis direction) is that the laying operation is carried out in sequence along the long side direction of the solar wing (X-axis), starting from one wing end and advancing to the other end; the activation condition is that the unit is activated after the adjacent unit on the left of each unit is completed. The longitudinal filling strategy (progress in the Y-axis direction) is that the laying operation is carried out along the short side direction of the solar wing (Y-axis), advancing from the wing root (close to the body) to the wing tip (outer edge); the activation condition is that the adjacent unit on the inside (close to the wing root) has been completed. The central symmetric filling strategy is to use the geometric center of the solar wing as the starting point, and simultaneously expand the laying path symmetrically along the main axis on both sides to ensure that the laying load is symmetrically distributed; the required conditions are that the target node is structurally symmetrical and its symmetric nodes simultaneously meet the laying activation conditions. The S-shaped filling strategy (snake path) is used to lay the round-trip path of the simulated print head, and the laying order alternates between "left→right" and "right→left" directions to lay the next row; it is suitable for laying large strip areas and can effectively reduce the reversal frequency and execution delay of the motion system. The dynamic execution process of the laying behavior graph is represented by a finite state machine or a directed acyclic graph model. The state transition of each node is controlled by the activation condition; state transfer events include operations such as "laying activation", "state transition" and "path completion". In each round of simulation, the system traverses the graph structure according to the current state of the laying graph, identifies the current set of activatable nodes, updates its state and records the laying behavior path, providing input basis for control execution.

[0111] Step S23: Verify the legitimacy of the laying behavior diagram data to obtain fixed simulation data.

[0112] In one embodiment, after each spreading step, the splicing error Δx between the paved surfaces is checked to be less than ε (e.g., 0.5 mm); iterative Boolean operations are used to verify whether there are any intersecting, overhanging, or broken areas between the paved surfaces. Structural simulation is performed on the current spreading state, such as setting node boundary conditions (fixed / free / elastic); applying loads (gravity, solar pressure); and calculating the maximum stress σ. max and allowable stress σ allow , requiring σ max <0.9·σ allow If not, roll back the strategy or reset the paving order. Determine whether the paving path traverses all target areas; check whether the strategy dependency chain has no closed loops (to avoid logical deadlocks); if there are path islands or non-convergence behavior, mark it as "non-convergence simulation."

[0113] Preferably, the perturbation layout solution is generated as follows:

[0114] Step S24: performing local graph division on the structural constraint graph data to obtain local graph data;

[0115] In one embodiment, the overall structure diagram is divided according to spatial proximity or functional partitioning attributes, specifically including the following two typical strategies: (1) With the key node in the structure diagram as the center, the set of k-order adjacent nodes is expanded outward to form a local subgraph. k is an integer, representing the maximum connection depth between nodes in the diagram (for example, k = 2 means starting from the core node, retaining adjacent nodes connected by at most two layers). A subgraph area with local compactness and strong connectivity in space is constructed for local perturbation testing. (2) Based on the functional area division of the solar wing structure, the nodes in the diagram are partitioned and mapped according to their structural functions, such as the main wing section, auxiliary deployment mechanism, terminal structure area, etc. Each functional segment is mapped to an independent local graph. Strategy optimization and simulation deduction are carried out based on functional consistency to improve the local accuracy of structural function simulation. Regardless of the partitioning strategy adopted, the system ensures that the generated local graph maintains connectivity in the structural topology and has the characteristics of being able to be independently perturbed and analyzed logically, which facilitates the diversity generation and local stability screening in the subsequent layout scheme.

[0116] Step S25: performing disturbance map selection on the local map data to obtain disturbance map data;

[0117] In one embodiment, the system quantitatively scores each local graph based on the following three perturbation sensitivity indicators: structural degrees of freedom (DOF), which is the total number of unconstrained nodes in the local graph, reflecting the potential deformation capacity of the region. The higher the DOF, the more likely the local region is to produce a significant response to external forces or spreading perturbations; node distribution density, which is the ratio of the number of nodes to their area in the main projection direction. This density indicator reflects the complexity and tightness of the local structure; and structural force concentration, which is obtained based on historical simulation data or structural static analysis results and reflects the maximum stress distribution trend of the local region under load, is an important basis for judging the sensitivity of the structure's response. Based on these three indicators, the system constructs a weighted scoring function, converting the sensitivity of each local graph into a set of perturbation scores. By setting the weighting coefficients for each indicator (for example, according to the actual simulation strategy to configure the emphasis on nonlinear deformation or structural strength response, such as increasing randomness with weights of 0.5, 0.25, and 0.25), the system scores and ranks all local graphs. The system can adopt one of the following strategies to select perturbation map data: (1) select the top-k local maps with the highest scores as candidate perturbation areas; (2) randomly select several local maps whose scores exceed the set threshold; (3) execute the full coverage strategy to perform perturbation simulation on all local maps.

[0118] Step S26: performing disturbance space selection on the paving behavior graph data according to the disturbance graph data to obtain disturbance space data;

[0119] In one embodiment, the system maps the set of structural nodes contained in the perturbation graph to the corresponding set of laying units in the laying behavior graph. If the activation condition of a laying unit depends on the local structural response path, that is, there is a behavioral chain dependency structure, then the unit is determined to be a target area with perturbation potential and marked as a perturbable space. The perturbation space can be divided into the following types according to the laying behavior characteristics and perturbation mode it assumes in the behavior graph, such as the path perturbation type, which realizes the perturbation rearrangement of the chain behavior path by adjusting the execution order of the local laying units in the behavior graph; the delayed perturbation type, which introduces a controlled delay in the activation process of the local behavior unit to simulate the intervention of the response lag or the structural feedback mechanism on the laying process; the strategy replacement type, which partially replaces the original set laying method, for example, replacing the sequential or S-shaped laying method with a symmetrical or alternating layout strategy, thereby constructing a multi-strategy perturbation space.

[0120] Step S27: performing disturbance processing according to the disturbed spatial data to obtain disturbance result data;

[0121] In one embodiment, the system applies various types of disturbances to the corresponding units of the laying behavior graph, and then constructs a variety of layout candidate paths and control schemes after disturbance, and evaluates their stability and robustness during the structural response process. Specific disturbance types and operations include but are not limited to the following categories, including sequential disturbances, which introduce changes in the local path sequence by exchanging the execution order of adjacent laying units to explore the impact of the laying logic sequence on the overall response of the structure; angle disturbances, which introduce directional offsets within a certain range at the structural node level to simulate the angle error or structural installation deviation when the node is laid; position disturbances, which apply tiny spatial displacement disturbances to local nodes in the laying behavior graph to simulate the geometric position changes caused by physical deviations during the laying process; strategy disturbances, which replace the original laying strategy path, for example, replacing the "S-type laying method" with a "bidirectional symmetric strategy" to evaluate the structural energy efficiency, laying path length and response robustness under different strategy forms. During the disturbance processing process, the system gradually advances the simulation process based on the behavior graph after disturbance. The simulation module will update the structural model status in real time, dynamically record the structural displacement field distribution, the evolution process of the laying state diagram and the change trend of the energy function under each disturbance step, and generate complete structural evolution trajectory data.

[0122] Step S28: Verify the legitimacy of the disturbance result data to obtain random simulation data.

[0123] In one embodiment, the system makes a legality judgment from two dimensions: geometric integrity and mechanical stability. The geometric legality verification is specifically a spatial topological consistency check of the disturbed structural model by the system, which mainly includes face-to-face overlap detection, whether the layout units are suspended or disconnected, and whether the structural surfaces are damaged or broken. Specifically, the system calculates the minimum distance between all adjacent structural surfaces. If the minimum face spacing is greater than 0.5 mm, the structure is considered to be geometrically legal, otherwise the sample will be eliminated. The mechanical legality verification is specifically a physical response analysis of the disturbed structure using finite element simulation tools, focusing on two parameters: (1) the maximum stress value of the structure under loading conditions; (2) the maximum response displacement of the structure. The system evaluates the difference between the maximum stress value in the disturbed sample and the stress reference value in the undisturbed state. If the absolute value of the stress change is less than the set threshold (for example, 15 MPa) and the maximum displacement does not exceed the upper limit allowed by the system (for example, 5 mm), the mechanical response is judged to be reasonable. The perturbation results that meet the above geometric and mechanical conditions will be confirmed as valid perturbation samples, constituting the random simulation data set used by the system for strategy optimization, diverse layout generation and robustness evaluation.

[0124] Preferably, the regional structure analysis is specifically as follows:

[0125] Step S31: constructing a structure domain response tensor for the fixed simulation data and the random simulation data to obtain structure domain response tensor data;

[0126] In one embodiment, the system extracts key structural response parameters of each laying unit in the fixed simulation data and the random simulation data. Each laying unit can be a triangular facet, a rectangular tile unit or other meshable surface unit. For each unit, the system records the following response fields, stress value σ, displacement δ, laying state mark s (laid / unlaid), laying angle θ, and local energy dissipation index E. The system divides the entire simulation area into a two-dimensional space grid. For each grid unit, the system integrates its structural response fields in a fixed order to form a structural response vector. If the laying behavior is carried out in a multi-step simulation process, and there is an evolutionary trend in the response of each time step, the system can superimpose the response tensor at each time step into a third-order tensor structure. The dimension of this tensor can be expressed as m is the spatial dimension, n is the temporal dimension, and d is the number of feature dimensions (such as 5 or 6).

[0127] Step S32: Calculating the coverage of the routeable area based on the structural domain response tensor data to obtain coverage data of the routeable area;

[0128] In one embodiment, in the structural domain response tensor, a binary field representing the unit laying state is extracted and recorded as s i,j, where i and j represent the position index in the two-dimensional space grid, s i,j The value of 1 indicates that the unit has been laid out, and the value of 0 indicates that the unit has not been laid out. The system divides the entire laying area into multiple sub-areas of fixed size, denoted as R k , each sub-region contains several adjacent grid cells. Preferably, each sub-region can be set as a grid block with five rows and five columns (i.e., 5×5) for local area statistics. In each sub-region R k In the system, the number of deployed units is counted and the ratio is calculated with the total number of units in the area to obtain the coverage rate of the area that can be deployed, which is recorded as $\rho_k$. The calculation method is: k The sum of the number of cells with a laying state of 1 in the area is divided by the total number of cells in the area, that is, the coverage rate ρ k is equal to the ratio of the number of laid out units to the total number of units in the area. k ≥0.75, marked as "high coverage area"; if 0.4≤ρ k <0.75, marked as “medium coverage area”; otherwise, marked as “low coverage area”.

[0129] Step S33: performing local symmetry offset calculation based on the structural domain response tensor data to obtain local symmetry offset data;

[0130] In one embodiment, a reference symmetry axis is set for the entire paving surface or solar wing structure. Preferably, the geometric center line or the X axis of the structural coordinate system can be selected as the symmetry reference axis, which is denoted as the symmetry axis x = x0. This axis corresponds to the solar wing deployment direction or the main symmetry line in the design drawing. Then, for each local layout area R k , the system calculates the coordinates of the two-dimensional geometric center of mass of the area, which is (x c ,y c ), where x c Indicates the center of mass position of the area in the X direction, y c Indicates the center of mass position in the Y direction. The center of mass is obtained by weighted average calculation of the position coordinates of all laid units in the area. After obtaining the center of mass position, the system uses x=x0 as the symmetry reference axis and defines the first-order symmetry offset of the area as the absolute difference Δ between the center of mass of the area and the symmetry axis in the X direction. sym (R k )=|x c―x0|. The closer this value is to zero, the closer the layout of the area is to symmetry, and theoretically the force is more balanced, which is suitable for the layout of battery modules with high requirements for structural symmetry. In this embodiment, a second-order symmetry offset metric is further defined to evaluate the overall degree of difference between the shape of the area and its mirror image about the axis of symmetry. The regional outline or layout shape is mirrored with the axis of symmetry as the reference, and the difference in spatial distribution between the original area and the mirror area is calculated in the form of mean square error as a mirror offset metric. The second-order symmetry offset can be defined as the mean square error value between the original area and its mirror area, which reflects the degree of consistency of the overall shape symmetry. The smaller the value, the more symmetrical the regional morphology. Through the joint calculation of the above-mentioned first-order and second-order symmetry offsets, the system can quantify the symmetry performance of each local area in multiple dimensions.

[0131] Step S34: performing region label inference based on the deployable region coverage data and the local symmetry offset data to obtain region label data;

[0132] In one embodiment, the system pre-sets a set of regional label sets to characterize the structural adaptability level of the region. The label set includes the following three categories, such as the preferred area, which indicates that the area has both high layout density and good structural symmetry, and is suitable for priority battery layout; the available area: indicates that the area is at a medium level in terms of layout density and symmetry, and can be selected when resources are limited or the layout requirements are more complex; the risk area indicates that the area has significant defects, such as low coverage or poor symmetry, and is not recommended as the preferred layout object. The reasoning process of the above labels is based on two key indicators: coverage, which indicates the ratio of laid units to layable units in the area, reflecting the layout density; symmetry offset, which indicates the degree of displacement of the center of mass of the area relative to the symmetry axis, reflecting the structural stability. The system uses the following classification rules for label inference: if the coverage of an area is not less than 75% and the symmetry offset does not exceed 5 mm, the area is marked as a preferred area, indicating that it excels in both laying efficiency and structural symmetry; if the coverage of an area is between 40% and 75% and the symmetry offset does not exceed 15 mm, the area is marked as a usable area, indicating that it has basic deployment value; if the area does not meet any of the above conditions, that is, the coverage is less than 40% or the symmetry offset exceeds 15 mm, the area is marked as a risk area and is not recommended for priority use.

[0133] Step S35: construct candidate maps for domain sorting of the fixed simulation data and the random simulation data according to the region label data to obtain region structure domain data.

[0134] In one embodiment, the system uses each structural area as a graph node, and archives the multi-dimensional structural attributes and response characteristics corresponding to the area as node attributes to form the basic unit of the regional candidate map. Each regional node contains the following feature fields, such as the coverage value of the routable area, which reflects the layout density of the area; the symmetry offset, which indicates the degree of deviation of the regional layout relative to the symmetry axis; the regional average stress value, which indicates the structural load response intensity of the area during the simulation process (which can be calculated based on historical experience or based on finite element calculation of simulation data obtained based on preset parameters and strategies); regional label type, such as "preferred area", "available area" or "risk area", representing the structural adaptability level of the area; recommended paving direction, such as along the X direction, Y direction or diagonal direction, reflecting the appropriate paving path planning strategy. While constructing the nodes, the system further establishes the connection relationship between the regional nodes in the map (i.e., the edge structure) to express the spatial adjacency or physical response similarity between regions. The system determines whether two regions are connected based on any of the following conditions: spatial connectivity: if the Euclidean distance between the geometric centers of two regions is less than a set spatial threshold, they are considered spatially adjacent and a connecting edge is established; response gradient continuity: if the main indicators of the simulation response (such as stress or displacement values) of the two regions have strong gradient continuity, they are considered to have a response coupling relationship and a connecting edge can be established. The system organizes all candidate regions, connection relationships, and label information into a complete regional structure domain atlas data.

[0135] Preferably, the preferred layout order is specifically:

[0136] Step S36: constructing regional weight tensor fusion on the regional structure domain data to obtain regional weight tensor data;

[0137] In one embodiment, the multi-dimensional attributes of the region (such as coverage, stress, symmetry, etc.) are integrated to form a weighted region tensor as the basis for layout sorting. k Extract the following characteristic indicators, such as the coverage rate ρ k , symmetry offset Δ sym,k , mean stress Number of units N k , the laying direction consistency score is C k (Extract path vector; for each pair of adjacent path vectors in the path vector, calculate the angle cosine of the unit vector; according to the angle cosine cosθ of the unit vector i Calculate the average directional consistency as If all directions are consistent, C k →1 represents strong consistency; if the direction changes drastically, C k →0 represents poor coherence), label confidence P kAll indicators are normalized by min-max, kept in the range of [0,1], and constructed in tensor form such as each region corresponds to a multidimensional weight vector All regions form a three-dimensional tensor structure Where n is the number of regions and d is the feature dimension.

[0138] Step S37: performing nonlinear mapping sorting on the regional weight tensor data to obtain preliminary layout sorting data;

[0139] In one embodiment, a nonlinear scoring mapping strategy is used to map the weight vectors of each region. (Refer to the previous step for parameter names and meanings) to sort and form a preliminary priority result. Use weighted nonlinear mapping functions such as Sigmoid, ReLU or exponential functions to enhance response sensitivity. Sorting score function where x i is the regional weight tensor data, f(x i ) is a nonlinear transformation (e.g. f(x) = log(1 + αx), α can be set to 5); weight w i It can be set according to user preferences or system goals, such as coverage priority → w1 = 0.4, symmetry priority → w2 = 0.2, the smaller the stress, the better → w3 = 0.2, and the remaining balance items → w4 + w5 + w6 = 0.2. For all R k According to S k Sort from largest to smallest; retain the index, region ID, and sort score.

[0140] Step S38: Perform semantic annotation enhancement of candidate solutions on the preliminary layout ranking data to obtain layout ranking data.

[0141] In one embodiment, based on the sorting, a semantic description is added to each candidate area to assist manual review or system explainability. The system reversely infers the structural stability level of the area during the actual layout process based on the area's routable coverage and average stress index. If a region has the attribute combination of "high coverage + low average stress" (judged based on the threshold set by the experience value), it is marked as a "stable area"; if the region shows "low coverage + high stress" (judged based on the threshold set by the experience value), it is marked as a "high-risk area"; the intermediate situation is marked as "acceptable risk". The geometric shape of the boundary shape of each region is identified based on the convex hull analysis of the region contour, aspect ratio calculation and boundary connectivity identification. For example, the convex area has basically no depressions on the boundary and is suitable for symmetrical paving; the strip area has a long side much larger than the short side, which is suitable for S-shaped paving strategy; the concentrated block area has a relatively compact boundary and is suitable for expansion layout starting from the center. The system combines the structural characteristics of the region in the sorting tensor to recommend the type of paving path that is most suitable for the region. If an area exhibits "ribbon-like and high continuity," an "adaptive S-shaped paving path" is recommended; if the area has a centrally symmetrical shape and a stable structure, "adaptive centrally symmetrical paving" is recommended; if there are no distinct features or fragmented boundaries, "locally adaptive paving" may be recommended. The system extracts the key factors that most influence the score from the ranking scoring function to generate readable explanations for the recommendations. If a region's score is dominated by high coverage and low symmetry offsets, the statement "Due to the high coverage and good symmetry of this region, paving is recommended as a priority" can be automatically generated; if the primary factor is stress, the statement "This region has a low stress level and good structural bearing capacity" can be generated.

[0142] Preferably, step S4 is specifically:

[0143] Step S41: performing three-dimensional layout projection according to the layout sorting data to obtain three-dimensional layout projection data;

[0144] In one embodiment, if the region information in the sorted data directly corresponds to a predefined local grid segment of the solar wing surface, its grid index and node information can be directly read without coordinate transformation.

[0145] In one embodiment, if the layout area is given by a two-dimensional plane analysis or an external coordinate system result, it needs to be mapped to the three-dimensional model surface through the following steps: using the nearest point projection method, the center of mass coordinates and boundary point set of the layout area are projected onto the three-dimensional model surface grid to determine its mapping target facet set. During the projection process, ensure that the normal direction of the layout area (denoted as ) and the normal direction of the solar wing surface (denoted as ) are in the same direction and meet the following directional consistency conditions If this condition is not met, the regional posture is flipped or reflected to ensure surface conformity. The laying plane in the external reference system is mapped to the model local coordinate system.

[0146] Step S42: performing surface expansion according to the three-dimensional layout projection data to obtain surface expansion data;

[0147] In one embodiment, the layout area patches in the three-dimensional space are geometrically unfolded to convert them into a two-dimensional plane structure suitable for drawing expression. A geometric unfolding algorithm (such as LSCM) is used to maintain angle distortion; for complex surface patches, partition unfolding or physical simulation unfolding (spring model) can be used; after unfolding, each patch F i Contains 2D vertex coordinates, edge lengths, and original 3D reference information. Unfolded meshes are arranged in layout order, maintaining relative spatial logic. Reference direction arrows are added (e.g., sun direction, paving start edge, area number, and strategy labeling).

[0148] Step S43: generating drawing elements according to the surface expansion data to obtain drawing element data;

[0149] In one embodiment, the system uses each unfolded paving unit (such as a triangle, rectangle, or other polygonal patch) as a basic drawing object, extracts its 2D boundary contour, and supplements it with engineering information related to the paving process. This generates the following types of drawing elements: geometric boundary line elements, which represent the spatial contour of each paving unit in the drawing, in the form of a closed polyline or closed polygon structure; paving start mark elements, which identify the paving start point of each paving block, often presented as a starting dot or starting number arrow; unit number and battery model annotation elements, which generate a unique number mark for each paving unit and append the selected battery type information for that unit, such as "B-158×158-72", to guide battery assembly configuration; spatial reference information elements, which indicate the mapping relationship of the paving unit in the original 3D structure, including the original structure ID and paving grid position number; and auxiliary layer elements, which display additional attributes of the paving area, such as simulated stress distribution contour maps, occlusion area identifiers, and reserved area outlines. The system introduces a layer management mechanism, which organizes and manages different types of drawing elements into multiple layers. Preferably, the following layer structure is set up: Layer 0: stores structural outline elements as a basic framework reference; Layer 1: stores laying path information and numbering marks to guide layout and installation operations; Layer 2: is used to add text annotations, technical instructions, and other textual identification content; Layer 3: reserves special areas used in process compilation, such as routing space and buffer areas. The system encapsulates each complete drawing unit (consisting of boundary lines, numbers, type annotations, etc.) as a drawing element block (Block). Each block is saved as an independent object and supports operations such as copying, adjustment, and renumbering, facilitating drawing version iteration or modular application.

[0150] Step S44: Export the drawing format according to the drawing element data to obtain the solar wing battery selection data.

[0151] In one embodiment, the system supports exporting drawings in multiple common CAD formats, including .DXF, .DWG, .SVG, and .PDF. Drawing and drawing generation can be performed through CAD development interfaces, such as the AutoCAD.NET API or OpenCascade development tools. During the export process, the system automatically sets the drawing's view scale, print area, text annotation dimensions, and frame structure. It also generates metadata fields below the drawing, including information such as the drawing number, draftsman, reviewer, and version number, to meet the requirements of standard engineering drawings. Furthermore, the system generates structured battery selection information based on the geometric dimensions and layout density of each battery placement area recorded in the drawing elements. Specifically, this includes fields such as the area identifier (area ID), which identifies the individual placement areas on the solar wing; the number of cells, which indicates the total number of battery cells that can be accommodated in the area; matching battery models, which automatically recommends matching standard battery models based on the area shape and cell specifications; power estimates, which calculate an estimated total power output for the area based on the rated power and number of each cell; and electrical connection recommendations, which recommend series, parallel, or mixed connection methods based on the layout topology for reference during electrical system design.

[0152] Preferably, the present application further provides a solar wing battery selection system based on a three-dimensional layout, which is used to execute the solar wing battery selection method based on a three-dimensional layout as described above. The solar wing battery selection system based on a three-dimensional layout includes:

[0153] A three-dimensional substrate modeling module is used to obtain three-dimensional data of the solar wing and construct a three-dimensional substrate model based on the three-dimensional data of the solar wing to obtain a three-dimensional model of the solar wing;

[0154] A layout simulation generation module is used to simulate the structure-driven deployment strategy of the solar wing three-dimensional model to obtain fixed simulation data, and to generate a perturbation layout scheme by performing fixed strategy simulation on the solar wing three-dimensional model to obtain random simulation data;

[0155] The structural region analysis and optimal sorting module is used to perform regional structural domain analysis on fixed simulation data and random simulation data to obtain regional structural domain data, and to perform optimal layout sorting on the regional structural domain data to obtain layout sorting data;

[0156] The drawing generation and matching output module is used to export the layout sorting data into two-dimensional drawings and obtain the solar wing battery matching data.

[0157] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, the scope of the present invention is limited by the attached application documents rather than the above description, and it is intended that all changes that fall within the meaning and scope of equivalent elements of the application documents are included in the present invention.

[0158] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A solar wing battery selection method based on three-dimensional layout, characterized in that: The following steps are involved: Step S1: Acquire three-dimensional data of the solar wing, and construct a three-dimensional substrate model based on the three-dimensional data of the solar wing to obtain a three-dimensional model of the solar wing; Step S2: performing a structure-driven deployment strategy simulation on the solar wing three-dimensional model to obtain fixed simulation data, and performing a perturbation layout scheme generation on the fixed strategy simulation on the solar wing three-dimensional model to obtain random simulation data; Step S3: performing regional structure domain analysis on the fixed simulation data and the random simulation data to obtain regional structure domain data, and performing optimal layout sorting on the regional structure domain data to obtain layout sorting data; Step S4: Export the layout sorting data into a two-dimensional drawing to obtain solar wing battery selection data.

2. The method according to claim 1, characterized in that Step S1 is specifically as follows: Obtain three-dimensional data of solar panels; Perform heterogeneous 3D structure perception based on solar wing 3D data to obtain heterogeneous 3D structure data; Perform layout adaptation domain perception on heterogeneous three-dimensional structure data to obtain layout adaptation domain data; The base plate model is generated according to the layout adaptation domain data to obtain a three-dimensional model of the solar wing.

3. The method according to claim 2, characterized in that The specific heterogeneous three-dimensional structure perception is: Perform heterogeneous data analysis based on the three-dimensional data of the solar wing to obtain heterogeneous data; Perform structural semantic mapping on heterogeneous data to obtain structural semantic data; A three-dimensional structure expression body is generated according to the structural semantic data to obtain heterogeneous three-dimensional structure data.

4. The method according to claim 2, characterized in that The deployment of adaptive domain perception is specifically as follows: Calculate the surface geometric paving ability based on heterogeneous 3D structural data to obtain preliminary geometric paving domain data; Eliminate the structural functional area of the preliminary geometrically paved domain data to obtain functional mask data; Perform semantic mask fusion based on functional mask data and heterogeneous 3D structure data to obtain semantically enhanced layout domain data; The semantically enhanced layout domain data is screened for regional connectivity to obtain the layout adaptation domain data.

5. The method according to claim 1, wherein The structure-driven spreading strategy simulation is as follows: Constructing a structural constraint diagram for the three-dimensional model of the solar wing to obtain structural constraint diagram data; Performing paving behavior graph scheduling on the structural constraint graph data to obtain paving behavior graph data, wherein the paving behavior graph scheduling includes a horizontal filling strategy, a vertical filling strategy, a central symmetric filling strategy, and an S-shaped filling strategy; The legitimacy of the laying behavior diagram data is verified to obtain fixed simulation data.

6. The method according to claim 1, characterized in that The perturbation layout scheme is generated as follows: Performing local graph partitioning on the structural constraint graph data to obtain local graph data; Performing perturbation graph selection on the local graph data to obtain perturbation graph data; Performing disturbance space selection on the paving behavior graph data according to the disturbance graph data to obtain disturbance space data; Perform disturbance processing according to the disturbance space data to obtain disturbance result data; The legitimacy of the disturbance result data is verified to obtain random simulation data.

7. The method according to claim 1, characterized in that The regional structure analysis is as follows: Constructing structure domain response tensors for fixed simulation data and random simulation data to obtain structure domain response tensor data; Calculate the coverage rate of the routable area based on the structural domain response tensor data to obtain the coverage rate data of the routable area; Calculating the local symmetry offset according to the structural domain response tensor data to obtain the local symmetry offset data; Perform region label inference based on the deployable region coverage data and local symmetry offset data to obtain region label data; According to the regional label data, the fixed simulation data and the random simulation data are sorted by domains to construct candidate maps, and the regional structure domain data is obtained.

8. The method according to claim 7, characterized in that The preferred layout order is as follows: The regional weight tensor is fused and constructed on the regional structure domain data to obtain the regional weight tensor data; Perform nonlinear mapping and sorting on the regional weight tensor data to obtain preliminary layout sorting data; The preliminary layout ranking data is enhanced with semantic annotation of candidate solutions to obtain layout ranking data.

9. The method according to claim 1, characterized in that Step S4 is specifically as follows: Performing three-dimensional layout projection according to the layout sorting data to obtain three-dimensional layout projection data; Perform surface expansion according to the three-dimensional layout projection data to obtain surface expansion data; Generate drawing elements according to the surface expansion data to obtain drawing element data; The drawing format is exported according to the drawing element data to obtain the solar wing battery selection data.

10. A solar wing battery selection system based on three-dimensional layout, characterized in that: For executing the solar wing battery selection method based on three-dimensional layout according to claim 1, the solar wing battery selection system based on three-dimensional layout comprises: A three-dimensional substrate modeling module is used to obtain three-dimensional data of the solar wing and construct a three-dimensional substrate model based on the three-dimensional data of the solar wing to obtain a three-dimensional model of the solar wing; A layout simulation generation module is used to simulate the structure-driven deployment strategy of the solar wing three-dimensional model to obtain fixed simulation data, and to generate a perturbation layout scheme by performing fixed strategy simulation on the solar wing three-dimensional model to obtain random simulation data; The structural region analysis and optimal sorting module is used to perform regional structural domain analysis on fixed simulation data and random simulation data to obtain regional structural domain data, and to perform optimal layout sorting on the regional structural domain data to obtain layout sorting data; The drawing generation and matching output module is used to export the layout sorting data into two-dimensional drawings and obtain the solar wing battery matching data.

Images