Aircraft component boundary line grid node distribution and anisotropy parameter prediction method

Abstract

The invention discloses an aircraft part boundary line grid node distribution and anisotropy parameter prediction method, and relates to the field of automation and intellectualization of simulation pretreatment of computational fluid mechanics, and the method comprises the steps: determining geometric data, part semantic tags, calculation conditions and boundary lines of an aircraft CAD model; utilizing the trained multi-task graph neural network model to perform grid region type intelligent classification on the boundary line to obtain an isotropic or anisotropic classification result; then, according to a classification result, predicting a sub-network through differentiated parameters corresponding to the region types in the model, and predicting a grid control parameter set corresponding to each boundary line for each boundary line; and finally, driving a grid generator to automatically generate a high-quality surface grid based on the parameter set. According to the method, full-process intelligent decision making from semantic information to grid generation strategies and parameters is achieved, a traditional method depending on artificial experience is replaced, and the automation degree and efficiency of CFD pretreatment and the consistency of grid quality are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of automation and intelligence in computational fluid dynamics simulation preprocessing, specifically to a method for predicting the mesh node distribution and anisotropic parameters of aircraft component boundary lines. Background Technology

[0002] In the digital design process of complex equipment such as aircraft, numerical simulation based on computational fluid dynamics (CFD) is the core method for evaluating their aerodynamic performance. The accuracy and efficiency of the simulation results are highly dependent on the quality of the computational mesh generated in the preprocessing stage. Among them, generating a high-quality surface mesh (usually a triangular mesh) on the model surface is the foundation and key step in the entire mesh generation process.

[0003] Currently, unstructured mesh generation technology is widely used in industry. In this process, engineers first discretize each boundary line on the surface of the CAD model (i.e., distribute mesh nodes) and set the distribution rules for node sizes. Then, based on these discrete points of the boundary lines, surface triangular meshes are generated using Delaunay triangulation or the leading edge method. To accurately capture key physical phenomena such as boundary layers and shock waves, the mesh needs to exhibit different characteristics in different regions: isotropic meshes (triangles approximately equilateral in shape; isotropic meshes refer to meshes in two- or three-dimensional space where mesh elements have similar or identical dimensional characteristics in all directions. In the context of surface meshes (two-dimensional), this is typically represented by triangles with nearly equilateral shapes) are used in regions with clearly defined strong gradient directions (such as the object surface normal). Anisotropic meshes (long, narrow triangles used for boundary layer advancement; anisotropic meshes refer to meshes in two- or three-dimensional space where mesh elements in a specific direction are much smaller than those in other directions. In surface meshes, this is typically represented by long, narrow triangles, with one side (usually corresponding to the object surface normal) being much shorter than the others). Therefore, the strategic decision-making for surface mesh generation involves two levels:

[0004] (1) Mesh region type judgment: Engineers need to judge, based on experience, whether a certain region on the model surface (specifically its boundary line) should adopt an isotropic or anisotropic meshing strategy.

[0005] (2) Mesh control parameter settings: For boundary lines that are determined to be isotropic, parameters such as the total number of nodes and the node distribution function need to be set manually; for boundary lines that are determined to be anisotropic, in addition to the above parameters, a more complex set of parameters such as the first layer mesh height, growth rate and number of growth layers also need to be set.

[0006] Existing technologies rely primarily on engineers' experience and manual operation, which has the following prominent drawbacks:

[0007] The decision-making process is highly subjective and lacks automation: the determination of grid area types relies entirely on the engineer's personal experience and understanding of flow characteristics, lacking objective and intelligent judgment standards and methods. This high dependence on manual strategic decision-making has become the primary obstacle to achieving full automation of the pre-processing workflow.

[0008] Setting parameters is tedious, inefficient, and error-prone: A complex aircraft model contains hundreds or thousands of boundary lines, and manually setting multiple parameters for each line is time-consuming and laborious. Manual operation makes it difficult to ensure the consistency of parameter settings and is prone to introducing human error, affecting mesh quality and subsequent calculation accuracy.

[0009] Lack of systematic correlation and adaptability: Existing automation or auxiliary tools mostly focus on optimizing single parameters or generating local meshes, failing to establish a systematic and learnable mapping relationship between component functional semantics, flow physics requirements, mesh generation strategies, and specific control parameters. This leads to a disconnect between mesh generation strategies and high-level design intent and global computational conditions, making it unable to adapt to different simulation tasks and design variations.

[0010] In recent years, although some studies have attempted to apply machine learning to grid generation, such as predicting local grid size or optimizing cell quality, most of these methods focus on regression prediction of single, homogeneous parameters, and have not yet solved the core hierarchical decision-making problem of first determining the strategic type and then setting differentiated tactical parameters. How to automate and intelligently complete the entire decision-making chain from grid strategy classification to the coordinated prediction of the whole set of parameters based on high-level semantic information and physical requirements remains a technical challenge that urgently needs to be overcome in this field. Summary of the Invention

[0011] In view of the above-mentioned problems in the background art, the purpose of this invention is to provide a method for predicting the distribution of grid nodes and anisotropic parameters of aircraft component boundary lines, aiming to achieve intelligentization of the entire process from high-level design intent and physical conditions to low-level grid generation instructions. Specific objectives include: realizing intelligent automated decision-making for grid generation strategies, realizing intelligent prediction of differentiated parameters in coordination with strategies, and constructing an end-to-end hierarchical intelligent decision-making framework and system.

[0012] To achieve the above-mentioned objectives, this invention provides a method for predicting the mesh node distribution and anisotropy parameters of aircraft component boundary lines, the method comprising:

[0013] Step 1: Determine the geometric data, component semantic labels, calculation conditions, and boundary lines on the surface of the aircraft CAD model; wherein, the component semantic labels are used to identify the functional category associated with the mesh generation strategy to which each component in the aircraft CAD model belongs;

[0014] Step 2: Based on the geometric data, component semantic labels, and calculation conditions, a trained neural network model is used to classify the mesh region types of each boundary line on the surface of the aircraft CAD model, and the classification results of the mesh region types to which each boundary line belongs are obtained; the mesh region types include isotropic regions and anisotropic regions;

[0015] Step 3: Based on the classification results of the grid region types, use the differential parameter prediction subnetwork in the neural network model corresponding to each grid region type to predict the corresponding set of grid control parameters for each boundary line;

[0016] Step 4: Based on the set of mesh control parameters, drive the mesh generator to generate the surface mesh of the aircraft CAD model.

[0017] This method constructs a complete logical closed loop through four steps: Step 1 clarifies the basis for decision-making (geometry, semantics, working conditions, boundary lines); Step 2 uses an AI model to replace human judgment in determining the grid type; Step 3 intelligently predicts specific parameters based on the strategic decision results; Step 4 generates the grid based on intelligent instructions. This achieves fully automated mapping from semantics to classification to prediction and finally to the grid. It solves the fundamental efficiency and quality problems in traditional CFD preprocessing, where the two core decisions—grid region type judgment and specific parameter setting—rely entirely on manual intervention and are disconnected from each other. It establishes a new automated process framework based on semantics-driven, intelligent decision-making, and automatic execution. It compresses days or even weeks of experience-based manual work into an automated intelligent computing process.

[0018] Preferably, the neural network model is a multi-task graph neural network model; in step 2, the mesh region type is classified for each boundary line on the surface of the aircraft CAD model, specifically as follows:

[0019] Using each boundary line as a node and the topological connections between the boundary lines as edges, construct a boundary line topology graph;

[0020] The boundary line topology map and the feature vectors of each node are input into the multi-task graph neural network model; wherein, the feature vectors of each node are formed based on the geometric features extracted from the geometric data of the corresponding boundary line, the semantic labels of the component in which they are located, and the computing conditions;

[0021] Using the grid region type classification head in the multi-task graph neural network model, each node in the boundary line topology graph is classified into two categories, and the probability that each boundary line belongs to an isotropic region or an anisotropic region is output as the grid region type classification result.

[0022] This approach involves constructing the entire boundary line system of the model as a graph, with boundary lines as nodes and connections as edges. When input into a graph neural network, the model, in judging a particular boundary line, can simultaneously perceive information from other connected boundary lines, thus making a more co-categorized classification that aligns with the overall geometry and flow logic. This clarifies how intelligent classification works. It solves the problems of single, isolated boundary line judgments lacking context and failing to guarantee the overall rationality and coherence of region division. It endows classification decisions with a global perspective and contextual awareness, resulting in more accurate and consistent classification results.

[0023] Preferably, step 3 specifically involves: the multi-task graph neural network model, based on the grid region type classification result output by the grid region type classification head, routing the boundary line node features belonging to isotropic regions to the first parameter prediction sub-network to predict a first set of grid control parameters including the total number of nodes and the node size distribution function parameter; and routing the boundary line node features belonging to anisotropic regions to the second parameter prediction sub-network to predict a second set of grid control parameters including the total number of nodes, the node size distribution function parameter, the first-layer grid height, the grid growth rate, and the number of growth layers.

[0024] The model incorporates two independent and dedicated prediction sub-networks. The classification results act as routing signals, automatically guiding boundary features to the corresponding sub-networks: those heading towards isotropic networks only predict basic parameters; those heading towards anisotropic networks must predict a complete parameter set including first-layer height, growth rate, and layer number. This mechanism eliminates the possibility of missing key parameters. It also specifies how to differentiate predictions. This addresses the problem of a one-size-fits-all approach or parameter mismatch when predicting parameters for different strategy regions, ensuring that anisotropic regions always obtain the key parameters necessary for advancing the boundary layer. It achieves strong coupling between classification results and parameter prediction, guaranteeing the physical correctness and completeness of the parameter set.

[0025] Preferably, the boundary lines on the surface of the aircraft CAD model are determined as follows:

[0026] The input aircraft CAD model is processed to obtain component segmentation results and key geometric feature lines; wherein, the component segmentation results define the correspondence between each surface of the aircraft CAD model and the semantic labels of the components, and the key geometric feature lines include at least the boundary lines between all components;

[0027] The boundary lines between all components are selected from the key geometric feature lines and used as the boundary lines of the surface of the aircraft CAD model.

[0028] The method described above identifies the boundary lines as originating from key geometric feature lines obtained after intelligent component segmentation of the CAD model, and then filters out the intersection lines between components. This clarifies the automated acquisition source of each boundary line, a key input object, avoiding the bottleneck of still requiring manual extraction. It achieves automated pre-processing from the CAD model to the boundary line list, making the entire process entry point fully automated.

[0029] Preferably, the process of processing the input aircraft CAD model to obtain component segmentation results and key geometric feature lines specifically includes:

[0030] Step S1: Obtain the CAD model of the aircraft to be processed, wherein the CAD model of the aircraft has a boundary representation structure;

[0031] Step S2: Based on the engineering requirements of mesh generation, perform engineering semantic annotation on the faces in the boundary representation structure of the aircraft CAD model to obtain component semantic labels; extract the geometric parameters of each face from the boundary representation structure; generate a multi-view 2D rendering image for each annotated face; associate and store the multi-view 2D rendering image, the geometric parameters, and the component semantic labels to form training samples.

[0032] Step S3: Train a multimodal recognition model using the training samples. The multimodal recognition model is configured to fuse input image features and language semantic information to output semantic labels for components on the surface of the aircraft CAD model.

[0033] Step S4: Input the aircraft CAD model to be identified into the trained multimodal recognition model, identify the surfaces in the aircraft CAD model, and obtain the preliminary surface-part semantic label mapping relationship;

[0034] Step S5: Based on the preliminary face-part semantic label mapping relationship, for faces whose recognition confidence output by the multimodal recognition model is lower than a preset threshold or whose geometric area ratio is less than a set threshold, the geometric rule auxiliary module is activated to perform semantic completion to obtain the final part segmentation result.

[0035] Step S6: Based on the final component segmentation result, extract the key geometric feature lines corresponding to each component from the boundary representation structure of the aircraft CAD model.

[0036] This method, through step S1, limits the processing object to CAD models with boundary representation structures, ensuring the feasibility of all subsequent geometric operations (topology analysis, parameter extraction). By utilizing steps S2-S3 to construct a multimodal dataset oriented towards mesh engineering semantics and training a dedicated model integrating visual, geometric, and linguistic information, the method gains the ability to understand engineering semantics, replacing manual identification by engineers. Steps S4-S5 employ a collaborative mechanism of AI recognition of large components and geometric rule completion of small features, utilizing AI to handle macroscopic patterns while using geometric rules to ensure detailed integrity, overcoming the limitations of single methods on complex engineering geometries. Steps S6-S7 automatically connect the semantic recognition results to the geometric kernel, extracting key geometric feature lines and converting them into a format directly readable by mesh software, thus completing the final step from recognition to application and eliminating manual conversion. This solves the end-to-end automation bottleneck problem in the aircraft CAD mesh generation process, where component recognition relies entirely on manual labor, resulting in extremely low efficiency, inconsistent results, and the inability to directly use the recognition results for subsequent engineering. It achieves a fully automated, high-precision processing flow from the original CAD model to mesh-generated ready engineering data, reducing the time required for manual work, which can take several weeks, to automated processing.

[0037] Preferably, the semantic labels of components are determined as follows: based on the component segmentation results, the semantic labels of each surface of the aircraft CAD model are determined; the semantic labels of components include one or more of the following: wing leading edge, wing upper surface, wing lower surface, fuselage nose and engine nacelle.

[0038] Preferably, the multi-task graph neural network model is obtained through the following offline training method:

[0039] A training dataset is constructed, wherein each sample data in the training dataset corresponds to a boundary line in a historical successful computation case. The sample data includes input features and output labels. The input features include the geometric features, component semantic labels, and computational conditions of the boundary line. The output labels include the optimal grid region type label of the boundary line and the corresponding optimal grid control parameter label. The optimal grid region type label is used to supervise the obtaining of the grid region type classification result, and the optimal grid control parameter label is used to supervise the obtaining of the grid control parameter set.

[0040] Based on the training dataset, the constructed multi-task graph neural network model is trained end-to-end; wherein, the loss function of the multi-task graph neural network model is a joint loss function, which includes a classification loss for optimizing grid region type classification, a first regression loss for optimizing isotropic parameter prediction, and a second regression loss for optimizing anisotropic parameter prediction.

[0041] Preferably, the geometric features include at least one of the length and curvature information of the boundary line; the calculation conditions include at least one of the flight Mach number and the flight angle of attack.

[0042] Preferably, the node size distribution function parameter is used to control the density distribution of grid nodes along the length direction of the boundary line.

[0043] Preferably, in step 4, based on the set of mesh control parameters, the mesh generator is driven to generate the surface mesh of the aircraft CAD model, specifically as follows:

[0044] The predicted set of mesh control parameters for each boundary line is converted into the corresponding control script or API instructions for the target mesh generation software.

[0045] The control script or API instructions are executed to drive the target mesh generation software to complete the discretization of all boundary lines, anisotropic mesh advancement and surface triangulation, and output the final surface mesh file.

[0046] One or more technical solutions provided by this invention have at least the following technical effects or advantages:

[0047] Because this invention creatively adopts a hierarchical integrated technical solution that first intelligently divides the grid region types and then specifically predicts differentiated parameters, it achieves the following outstanding and beneficial effects:

[0048] This invention achieves a breakthrough in intelligent grid generation strategies: By automatically classifying region types based on AI, it is the first to systematically automate strategic decisions (which grid to use) that rely on high-level experience, solving core pain points that existing methods cannot address and achieving true strategy-level intelligence.

[0049] It ensures a high degree of self-consistency between grid parameters and physical requirements: The differentiated prediction architecture driven by classification results ensures that the parameters predicted for anisotropic regions must include the key parameters necessary for boundary layer advancement. The generated grid's density distribution and anisotropic characteristics naturally match the flow physical field, resulting in high physical fidelity.

[0050] This brings about an order-of-magnitude improvement in preprocessing efficiency and quality standardization: compressing the traditional manual decision-making process, which requires several days and relies on repeated trial and error, into automated calculations at the minute level. At the same time, by solidifying best practices in the AI ​​model, it ensures the consistency, repeatability, and high level of grid generation strategies and quality across different projects and personnel.

[0051] A transparent and optimizable standard intelligent process has been constructed: the fuzzy expert decision-making process is decomposed into two clear, learnable, and verifiable steps: classification and differentiated prediction, forming a transparent, explainable, and sustainably optimizable standardized intelligent pre-processing process. Attached Figure Description

[0052] The accompanying drawings, which are provided to further illustrate embodiments of the invention and constitute a part of this invention, are not intended to limit the scope of the invention.

[0053] Figure 1 A schematic diagram of a hierarchical intelligent decision-making model architecture;

[0054] Figure 2 This is a flowchart of the offline training phase.

[0055] Figure 3 Flowchart for the online application phase;

[0056] Figure 4 This is a flowchart illustrating the intelligent component recognition and segmentation method for aircraft CAD models based on three-modal fusion.

[0057] Figure 5 This is a schematic diagram of the architecture of an intelligent component recognition and segmentation system for aircraft CAD models based on three-modal fusion. Detailed Implementation

[0058] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, where there is no conflict, the embodiments of the present invention and the features thereof can be combined with each other.

[0059] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0060] Those skilled in the art should understand that, in the disclosure of this invention, the terms "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the above terms should not be construed as limiting this invention.

[0061] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0062] The core innovation of this invention lies in proposing and implementing a hierarchical intelligent decision-making framework that first classifies and then predicts. This framework first addresses the question of which grid strategy to use, and then addresses the question of what specific parameters to use. The two are tightly coupled, which differs from existing methods that only focus on single-parameter prediction.

[0063] This invention provides a method for intelligent mesh generation on aircraft surfaces based on semantic-driven and artificial intelligence. Please refer to [link / reference]. Figures 1-3 , Figure 1 This is a schematic diagram of a hierarchical intelligent decision-making model architecture, whose complete process includes an offline model training phase and an online intelligent application phase. Figure 2 This is a flowchart of the offline training phase. Figure 3 This is a flowchart for the online application phase.

[0064] 1) Offline training phase:

[0065] Step A: Construct a high-quality training dataset. Extract the input features (geometric, semantic, and operational conditions) and the corresponding optimal output labels (number of nodes, distribution parameters, and anisotropy parameters) for each boundary line from historical successful computation examples.

[0066] Step B: Build and train a deep learning model. Design a multi-task learning model to learn the complex mapping relationship from input features to multiple output targets.

[0067] 2) Online application stage:

[0068] Step C: Perform intelligent prediction on the new model. Input the component semantics, boundary lines, and operating conditions of the new aircraft model, and the trained model will automatically output all control parameters for each boundary line.

[0069] Step D: Drive automated surface mesh generation. Input the output parameters from Step C directly into the mesh generator to generate a high-quality initial surface mesh with one click.

[0070] (1) Detailed introduction of each step:

[0071] 1) Offline training phase;

[0072] Step 11: Construct a high-quality training dataset with region type annotations;

[0073] Implementing entity: Data management platform;

[0074] Execution content: Systematically collect historical successful CFD cases, associate them with CAD models, working conditions and final mesh files, and clearly label each boundary line with its region type label (isotropic / anisotropic), which is the basis for the model to learn classification tasks.

[0075] Target objects: Dispersed historical project files and databases.

[0076] Step 12: Extraction of training sample features and labels;

[0077] Execution entity: Automated feature and tag extraction pipeline;

[0078] Execution content: Extract input feature vectors for each boundary line, including geometric features (such as length, curvature, etc.), component semantic labels (such as wing leading edge), and global calculation conditions (such as Mach number, angle of attack, etc.); extract output labels, including region type labels and mesh parameter labels (number of nodes, distribution function parameters). If it is an anisotropic region, add anisotropic parameter labels (first layer height, growth rate, number of layers).

[0079] Execution target: a single historical data instance.

[0080] Step 13: Construction and training of the graph neural network model;

[0081] Execution entity: Model training server;

[0082] Execution content:

[0083] ① Graph structure input: Construct a graph with all boundary lines of the example as nodes and connections as edges, and input it into the model to enable the model to have context awareness.

[0084] ② Model Architecture: Design a multi-task graph neural network with an explicit hierarchical structure.

[0085] Level 1: Region Type Classification Head. This is a classifier that performs binary classification on each node (boundary line) in the graph, outputting the probability that it is an isotropic or anisotropic region. This step enables intelligent decision-making for the grid strategy.

[0086] Level 2: Differentiated Parameter Prediction Head. This consists of two structurally independent prediction sub-networks with different tasks. Sub-network A is an isotropic parameter prediction head, specifically processing node features classified as isotropic, outputting the node number and size distribution function parameters. Sub-network B is an anisotropic parameter prediction head, specifically processing node features classified as anisotropic, outputting the node number, size distribution function parameters, and complete anisotropic propagation parameters.

[0087] ③ Training Mechanism: Through gating or attention mechanisms, the classification results of the first level are used as routing information to dynamically guide the feature flow to the corresponding prediction sub-network in the subsequent level. A joint loss function (classification loss + isotropic regression loss + anisotropic regression loss) is used for end-to-end training, enabling the model to learn to collaboratively optimize classification and prediction tasks.

[0088] The loss function is as follows:

[0089] ;

[0090] In the formula, , and The weight parameters are initially set to... Adjustments can be made based on subsequent experiments.

[0091] Classification loss :

[0092] The first-level classification head outputs a probability for each node i1. This represents the probability that the boundary line belongs to an anisotropic region. (True label) (0 represents isotropic, 1 represents anisotropic). The classification loss uses binary cross-entropy:

[0093] ;

[0094] In the formula, N is the total number of nodes in the current batch (or graph).

[0095] For real labels The number of nodes in the second-level subnetwork A is predicted to be [number]. The distribution function parameter vector is The actual number of nodes and the distribution parameter vector are respectively and Then the isotropic regression loss function is:

[0096] ;

[0097] In the formula, The total number of isotropic nodes. For the regression loss function, mean squared error (MSE) or smoothed L1 loss (Huber) can be chosen to enhance robustness to outliers. The weight parameters are initially set to... Adjustments can be made based on subsequent experiments.

[0098] For real labels The number of nodes in the second-level subnetwork B is predicted to be [number]. The distribution function parameter vector is The anisotropy parameters are: first floor height ,growth rate Number of floors The actual number of nodes and the distribution parameter vector are respectively and The anisotropy parameters are as follows: , , The isotropic regression loss function is then:

[0099]

[0100] In the formula, The total number of isotropic nodes is initially set to [value]. Adjustments can be made based on subsequent experiments.

[0101] Execution target: The training sample set generated in step 2.

[0102] Step 14: Model Validation and Deployment Preparation;

[0103] Execution Module: Model Evaluation and Deployment;

[0104] Execution content: Evaluate the model's region classification accuracy and parameter prediction precision on the test set, and perform final validation through parameter-driven mesh reconstruction. After successful validation, package the model into a standardized service.

[0105] Target objects: The trained AI model and the test set.

[0106] 2) Online intelligent application stage;

[0107] Step 21: Input and Preprocessing;

[0108] Execution entity: CAD model preprocessing module;

[0109] Execution content: Receive the new aircraft CAD model and operating conditions, automatically perform component segmentation and semantic annotation, and extract the boundary lines of all components.

[0110] Execution target: New CAD model file and user-inputted operating parameters.

[0111] Step 22: Boundary line feature extraction and graph construction;

[0112] Execution unit: Feature extraction and graph construction module;

[0113] Execution content: Calculate geometric features for each boundary line, combine them with its semantic label and global conditions to form a feature vector. Construct a boundary line graph based on the topological connectivity of the boundary lines.

[0114] Execution target: Preprocessed model data.

[0115] Step 23: Hierarchical intelligent collaborative decision-making;

[0116] Execution entity: AI prediction engine;

[0117] Execution content:

[0118] ① Intelligent region type classification. The model first performs a first-level classification task, outputting a clear region type classification map that clearly identifies the grid strategy type of all boundary lines.

[0119] ② Intelligent prediction of differentiated parameters. Based on the partitioning results, the model automatically routes and activates isotropic and anisotropic parameter prediction heads in parallel, outputting two sets of structured parameter sets respectively. Finally, a structured list containing the region type of each boundary line and its corresponding complete parameter set is output.

[0120] Execution target: The boundary line topology graph constructed in step 2.

[0121] Step 24: Automated mesh generation;

[0122] Execution entity: Grid control script generation and driving module;

[0123] Execution content: The above structured parameter list is automatically converted into control scripts or API instructions that can be recognized by the target mesh generation software (such as NNW-GridStar), driving NNW-GridStar to automatically complete the discretization of all boundary lines, anisotropic mesh advancement and surface triangulation, and finally output a high-quality surface mesh file.

[0124] Target audience: List of prediction results output by the AI ​​model and third-party grid generation software.

[0125] The technical solution of Example 1 solves the two core bottlenecks in the existing technology: the complete reliance on human experience for determining the grid region type and the disconnect between parameter prediction and high-level grid strategy. It realizes full-process intelligentization from high-level design intent to low-level grid instructions. Specifically, it achieves: (1) Automated grid strategy decision-making: Based on component semantics and calculation conditions, it intelligently and automatically determines the grid region type (isotropic or anisotropic) to which each boundary line belongs, replacing human experience judgment. (2) Differentiated parameter intelligent prediction: Based on the region type determination, it differentiates and coordinates the prediction of a complete set of grid control parameters with different structures for different types of regions, ensuring that the parameters are highly matched with the physical strategy. (3) Construction of an end-to-end intelligent system: It establishes a hierarchical decision-making system that integrates intelligent classification and differentiated prediction, realizing full-process automation from CAD model semantic annotation and working condition input to one-click generation of high-quality surface mesh.

[0126] This method proposes an approach for automatically classifying surface mesh regions into isotropic or anisotropic types based on component semantics and computational conditions during CFD preprocessing using an artificial intelligence model. AI is utilized to achieve automatic classification decisions regarding mesh region types.

[0127] This method proposes an intelligent system that integrates region type discrimination and parameter prediction. Based on the region type discrimination result, it automatically selects and executes different parameter prediction logic branches to output parameter sets with different structures for isotropic and anisotropic regions.

[0128] This method constructs a complete, closed-loop framework consisting of three layers: intelligent region partitioning, parameter differential prediction, and automatic mesh generation. It realizes intelligent mapping and execution of the entire chain from high-level semantic information to low-level mesh generation instructions.

[0129] This method is not only applicable to CFD mesh generation for aircraft, but its core idea is to automatically determine the discretization and refinement settings based on the functional semantics and operating conditions of the components. This can be extended to any simulation preprocessing process involving complex geometry and physical field coupling, such as automobiles, ships, and energy equipment, and has broad industry application prospects.

[0130] The hierarchical decision-making framework proposed in this invention, which first performs intelligent classification based on object attributes and then differentiates processing based on the category, has broad applicability. It can be extended to other CAE simulation fields, such as: intelligently allocating different element types and mesh sizes based on material properties and stress states; and dividing contact mesh densities according to the relative motion relationship and importance of contact pairs.

[0131] The specific architecture of the AI ​​model (such as GNN classifier head, CNN classifier) ​​for region classification and the regression network for parameter prediction in this method can be replaced with other similar or related models or networks, and this invention does not impose any corresponding limitations.

[0132] In this method, the definition of region type categories is not limited to binary (isotropic / anisotropic) and can be extended to multi-dimensional classification (such as adding shock wave capture region, wake region, etc.). This invention does not impose any corresponding limitations.

[0133] In this method, the classification model and the prediction model can be a tightly coupled multi-task model or a loosely coupled but collaborative combination of independent models.

[0134] In this method, the third-party software that ultimately drives the generation of the mesh can be selected according to actual needs, and this invention does not impose any restrictions on it.

[0135] Example 2;

[0136] Based on Embodiment 1, Embodiment 2 of the present invention provides a detailed description of the component identification and segmentation method in Embodiment 1, specifically as follows:

[0137] Upon receiving a new aircraft CAD model and operating conditions, the method in Example 2 employs an intelligent component identification and segmentation method based on three-modal fusion for aircraft CAD models to automatically segment and semantically annotate components, extracting all component boundary lines.

[0138] Embodiment 2 of the present invention provides a method for intelligent component recognition and segmentation of aircraft CAD models based on three-modal fusion, the method comprising:

[0139] Step S1: Obtain the CAD model of the aircraft to be processed, wherein the CAD model of the aircraft has a boundary representation structure;

[0140] Step S2: Based on the engineering requirements of mesh generation, perform engineering semantic annotation on the faces in the boundary representation structure of the aircraft CAD model to obtain component semantic labels; extract the geometric parameters of each face from the boundary representation structure; generate a multi-view 2D rendering image for each annotated face; associate and store the multi-view 2D rendering image, the geometric parameters, and the component semantic labels to form training samples.

[0141] Step S3: Train a multimodal recognition model using the training samples. The multimodal recognition model is configured to fuse input image features and language semantic information to output semantic labels for components on the surface of the aircraft CAD model.

[0142] Step S4: Input the aircraft CAD model to be identified into the trained multimodal recognition model, identify the surfaces in the aircraft CAD model, and obtain the preliminary surface-part semantic label mapping relationship;

[0143] Step S5: Based on the preliminary face-part semantic label mapping relationship, for faces whose recognition confidence output by the multimodal recognition model is lower than a preset threshold or whose geometric area ratio is less than a set threshold, the geometric rule auxiliary module is activated to perform semantic completion to obtain the final part segmentation result.

[0144] Step S6: Based on the final component segmentation result, extract the key geometric feature lines corresponding to each component from the boundary representation structure of the aircraft CAD model;

[0145] Step S7: Convert the final component segmentation results and the extracted key geometric feature lines into a structured engineering semantic format for the target mesh generation software to read and utilize, and then output it.

[0146] This method, through step S1, limits the processing object to CAD models with boundary representation structures, ensuring the feasibility of all subsequent geometric operations (topology analysis, parameter extraction). By utilizing steps S2-S3 to construct a multimodal dataset oriented towards mesh engineering semantics and training a specialized model integrating visual, geometric, and linguistic information, the method gains the ability to understand engineering semantics, replacing manual identification by engineers. Steps S4-S5 employ a collaborative mechanism of AI recognition of large components and geometric rule completion of small features, utilizing AI to handle macroscopic patterns while using geometric rules to ensure detailed integrity, overcoming the limitations of single methods on complex engineering geometries. Steps S6-S7 automatically connect the semantic recognition results to the geometric kernel, extracting key geometric feature lines and converting them into a format directly readable by mesh software, thus completing the final step from recognition to application and eliminating manual conversion. This solves the end-to-end automation bottleneck problem in the aircraft CAD mesh generation process, where component recognition relies entirely on manual labor, resulting in extremely low efficiency, inconsistent results, and the inability to directly use the recognition results for subsequent engineering. It achieves a fully automated, high-precision processing flow from the original CAD model to mesh-generated ready engineering data, reducing the time required for manual work, which can take several weeks, to automated processing.

[0147] In step S2, the surfaces in the boundary representation structure of the aircraft CAD model are annotated with engineering semantics. Specifically, in the mesh preprocessing software environment, semantic labels are manually assigned to the surfaces contained in different components based on aerodynamic characteristics or structural functional differences.

[0148] Preferably, the geometric rule auxiliary module determines and corrects the semantic labels of the surface components based on the geometric topological relationships and parametric features of the aircraft CAD model, thereby forming the final component segmentation result.

[0149] Purely data-driven AI models are prone to low confidence or misjudgment when faced with small-sized, weakly textured, feature-fuzzy, or geometrically rare surfaces in the training data (such as slender wing trailing edges, tiny connecting protrusions, and hatches that smoothly transition into the main body). The above approach addresses this by introducing rule modules based on deterministic geometric knowledge to compensate for AI's shortcomings in such cases. General image segmentation models output pixel-level labels, which are difficult to automatically and error-free associate with the precise boundary representation (B-Rep) structure (faces, edges, vertices, and their topology) in CAD models. The above approach directly limits operations to the geometric and topological levels of the original CAD model, ensuring that any judgments and corrections are precisely applied to specific B-Rep surfaces, guaranteeing absolute geometric accuracy. By filling in gaps and correcting errors in the AI ​​recognition results, the final component segmentation results ensure that all geometric details affecting subsequent mesh generation and simulation analysis are covered, reducing the degradation of mesh quality or simulation accuracy caused by missing key small features. This approach eliminates reliance on a single AI model, creating a hybrid intelligence model that combines data-driven (AI) and knowledge-driven (rule-based) approaches. When faced with new aircraft models or those with unique geometries, even if the AI ​​model performs poorly initially, the method can still ensure the correct identification of basic and critical components through geometric rules, thus enhancing the method's generalization ability and practical stability.

[0150] Preferably, the multi-view 2D rendered image is generated as follows:

[0151] For each surface to be processed, multiple angles are used to wrap around it along its normal vector, and rendering is performed at each viewpoint to generate a series of two-dimensional view images.

[0152] Specifically, for each facet, multi-angle rendering is performed along its normal vector. This simulates the process of viewing a geometric facet from various directions, ensuring that the generated image fully reflects the facet's shape characteristics. This solves the problem of generating richly informative 2D image data from a single, planar aircraft CAD model facet, allowing the visual model to learn from it. It provides the visual model with multi-angle, comprehensive observation data, enhancing the model's robustness in recognizing component faces under different viewing angles and lighting conditions.

[0153] The preferred method for training the multimodal recognition model is as follows:

[0154] A neural network architecture is constructed, comprising a visual encoder, a language understanding module, and a cross-modal attention fusion module. First, the visual encoding module, based on Vision Transformer (ViT), is responsible for extracting features from the input image and outputting a visual token. Second, the language encoding module, based on a pre-trained large model, is responsible for encoding text prompts into text tokens. Finally, the cross-modal fusion module achieves bidirectional deep interaction between the visual and language tokens through a cross-attention mechanism. After multiple rounds of interaction, the features of the two modalities are aligned and fused in a unified semantic space to form a joint feature representation.

[0155] The neural network architecture is trained under supervision by using the multi-view two-dimensional rendered images and corresponding component semantic label texts in the training samples as input.

[0156] The multimodal recognition model is optimized using a combination loss function, which includes the classification cross-entropy loss function, the segmentation cross-union ratio loss function, and the boundary weighted loss function.

[0157] This paper presents a clear technical framework for fusing multimodal information through an architecture consisting of a visual encoder, a language module, and a cross-modal attention fusion module. Supervised learning is performed using a pre-constructed dataset, enabling the model to learn the mapping from images to semantic labels. A combined loss function is employed to simultaneously optimize classification accuracy, segmentation region consistency, and boundary precision, ensuring the quality of the model's output from multiple dimensions. The paper addresses the core technical path of model implementation by solving the problem of how to specifically construct and train a dedicated recognition model capable of understanding image-semantic associations. An optimized and trained intelligent model capable of accurately associating surface images with component semantic labels has been implemented.

[0158] Preferably, step S4 specifically involves identifying large component surfaces in the aircraft CAD model; step S5 specifically involves semantic completion of local feature surfaces, connecting region surfaces, or low-confidence identification surfaces in the aircraft CAD model.

[0159] Steps S4 and S5 are defined as a hierarchical collaborative process: the first layer (AI) is responsible for large components (efficiently handling regularities); the second layer (rules) is responsible for local features, connecting surfaces, and other complex situations (accurately handling special characteristics). This design concept is an efficient way to solve engineering complexity problems. It realizes intelligent division of labor and collaboration in the recognition process, significantly improving the completeness of recognizing small and special features while ensuring the efficiency of large component recognition.

[0160] Preferably, the rules for determining the geometric topological relationships and parametric features of the aircraft CAD model include at least one of the following rules:

[0161] Rule 1: If a face is located at the geometric intersection of two identified large component faces, and the area of ​​the face is less than the area of ​​the adjacent large component face, then the component semantic label of the face is determined to be a connecting feature face or an edge face.

[0162] Rule 2: If a surface is directly adjacent to the main wing and the angle between its normal vector and the normal vector of the main wing reference plane is greater than or equal to 45°, and the surface is located on the outermost side of the wing span and its area is between 0.5% and 5% of the main wing area, then it is classified as a winglet or wingtip.

[0163] Among them, the area proportion rule can be used to identify small faces at the boundary (such as edges). The direction relationship rule can be used to identify small faces that are not parallel to the main flow direction (such as hatches). It provides quantifiable and judgmental rule examples, enabling the system to automatically identify specific types of geometric features such as wing trailing edges and hatches.

[0164] Preferably, the extraction of key geometric feature lines corresponding to each component includes:

[0165] For the components identified by the multimodal recognition model, by analyzing the curvature distribution and boundary topology of its constituent surfaces, a geometric algorithm is used to extract the key geometric feature lines of the components, including its boundary lines and feature contour lines.

[0166] For the feature surfaces identified by the geometric rule assistance module, the corresponding key geometric feature lines are extracted based on their geometric definitions and their relationship with adjacent surfaces.

[0167] For large components identified by AI, geometric algorithms such as curvature and topology analysis are used to automatically extract their boundaries and contours. For small features identified by rules, they are extracted directly based on their geometric definitions (such as the intersection of two faces). Both methods ensure that the output feature lines are accurately associated with the semantic labels of the components. This achieves automated connection from semantic segmentation to geometric feature extraction, outputting geometric data that is directly relied upon for operations such as mesh encryption and block division.

[0168] Preferably, in step S5, before or after forming the final component segmentation result, a post-processing step is also included:

[0169] The semantic labels of components are smoothed based on the adjacency relationship of surfaces to integrate the recognition results of the multimodal recognition model with the supplementary results of the geometric rule auxiliary module, thereby eliminating misjudgments of isolated surfaces; and the final component segmentation results are visualized and exported.

[0170] The label smoothing process is based on the adjacency relationship of faces, performing consistency checks and adjustments on the labels of adjacent faces to eliminate unreasonable abrupt changes or isolated points. The AI ​​results are fused with rule-based results, and visualization and export are supported, forming a complete closed loop from processing to delivery. Post-processing improves the topological consistency and visual smoothness of the segmentation results, and visualization and export functions make the results easy for engineers to verify and use.

[0171] The input to the multimodal recognition model includes natural language instructions; the method allows users to specify the type of part to be recognized or the segmentation requirements by inputting natural language instructions; the visual encoder adopts a Transformer-based architecture, the language understanding module adopts a pre-trained large language model, and the cross-modal attention fusion module adopts a cross-attention mechanism.

[0172] The model's input can include natural language commands, which are then translated into internal tasks that drive the model and geometry engine, achieving intelligent conversion from natural language to concrete operations. This enhances the system's usability and flexibility, allowing users to drive the system to perform specific tasks using natural language (such as segmenting all wing surfaces). The key modules in the architecture are concretized into corresponding selections, ensuring the feasibility of the solution: a visual encoder (ViT or other Transformer architectures), a language understanding module (QwenLM or other large models), and a fusion module (CrossAttn), ensuring the powerful capabilities of each module.

[0173] Preferably, in step S7, the structured engineering semantic format includes: a set of component geometric groups for the mesh generation software to recognize and call, a set of feature lines describing the key geometric features of the components, and optional mesh generation strategy parameters associated with preset components. By explicitly defining the output format as the structured engineering semantic format required by industrial software, including the component group set, feature line set, and mesh strategy parameters, the output of this method is no longer merely a display result, but rather instructions or data that can be directly read and applied by CAE software, thus achieving a closed loop of engineering value.

[0174] One or more technical solutions provided in Embodiment 2 of the present invention have at least the following technical effects or advantages:

[0175] (1) Full automation of component identification has been achieved, greatly improving the efficiency of mesh preprocessing;

[0176] Because this invention employs a visual-geometric-linguistic three-modal fusion recognition and hierarchical collaborative processing mechanism, it can automatically and accurately complete the identification and semantic segmentation of various aircraft components. The effect is that it shortens the process of identifying medium-sized passenger aircraft components, which originally relied entirely on manual identification and grouping by engineers and took 1-2 weeks, to automated processing, thus solving the primary efficiency bottleneck in the mesh generation process.

[0177] (2) Outputs engineering semantics and geometric features that can directly drive mesh generation;

[0178] Because the output of this invention is not a general image segmentation mask, but rather a face-part label mapping table and key geometric feature lines that are precisely associated with the original B-Rep model, the effect is that the recognition results can be directly converted into components, sets, or layers that mesh generation software can recognize, and key feature lines (such as wing leading / trailing edge lines) used for mesh partitioning and densification control are automatically extracted, achieving recognition-ready functionality and eliminating the large amount of manual conversion and interpretation work required in traditional methods.

[0179] (3) Improved the completeness of identification and engineering practicality of small-sized and weak-featured parts;

[0180] Because this invention employs a layered strategy of AI recognition for large components combined with geometric completion for small features, it intelligently completes small or inconspicuous components such as wing trailing edges and access hatches using geometric rules. The effect is a significant improvement in the overall segmentation integrity, ensuring that subsequent mesh generation covers all necessary geometric details and making the generated mesh more consistent with the geometric integrity requirements of high-fidelity simulation.

[0181] (4) Improve grid quality by supporting smart grid strategy presets through semantic understanding;

[0182] Because the model training of this invention incorporates knowledge of mesh engineering, it possesses engineering semantic understanding capabilities. The effect is that the system can automatically recommend or apply preset mesh generation strategies based on component semantics (such as the leading edge of an airfoil), thereby generating higher-quality meshes more suitable for computation in key aerodynamic regions.

[0183] (5) Possesses strong domain generalization ability and knowledge accumulation value;

[0184] Because the core framework of this invention is not dependent on any specific aircraft model and supports incremental learning, the benefits are significant: only a small amount of data from new aircraft models needs to be labeled and fine-tuned to quickly adapt to new aircraft models, greatly reducing technical maintenance and expansion costs. Simultaneously, the system can distill the mesh generation experience of senior engineers into reusable models and rules, forming core knowledge assets.

[0185] Please refer to Figure 4 , Figure 4This invention provides a method for intelligent component recognition and segmentation of aircraft CAD models based on three-modal fusion, comprising:

[0186] Step S1 - Model Acquisition and Format Confirmation: Acquire the CAD model of the aircraft to be processed, wherein the CAD model of the aircraft has a boundary representation structure;

[0187] Step S2 - Construction of a multimodal dataset oriented towards engineering semantics: Based on the engineering requirements of mesh partitioning, engineering semantic annotation is performed on the faces in the boundary representation structure of the aircraft CAD model to obtain component semantic labels; geometric parameters of each face are extracted from the boundary representation structure; for each labeled face, a multi-view 2D rendering image is generated; the multi-view 2D rendering image, the geometric parameters, and the component semantic labels are associated and stored to form training samples;

[0188] Step S3 - Multimodal Fusion Recognition Model Training: The multimodal recognition model is trained using the training samples. The multimodal recognition model is configured to fuse input image features and language semantic information to output semantic labels for components on the surface of the aircraft CAD model.

[0189] Step S4 - Initial identification of macroscopic components based on AI: Input the CAD model of the aircraft to be identified into the trained multimodal recognition model, identify the surfaces in the aircraft CAD model, and obtain the preliminary surface-component semantic label mapping relationship;

[0190] Step S5 - Collaborative completion of small features based on geometric rules: Based on the preliminary face-part semantic label mapping relationship, for faces whose recognition confidence output by the multimodal recognition model is lower than a preset threshold or whose geometric area ratio is less than a set threshold, the geometric rule auxiliary module is activated to perform semantic completion to obtain the final part segmentation result.

[0191] Step S6 - Automatic extraction of key geometric feature lines: Based on the final component segmentation results, extract the key geometric feature lines corresponding to each component from the boundary representation structure of the aircraft CAD model;

[0192] Step S7 - Structured Engineering Semantic Output: The final component segmentation results and extracted key geometric feature lines are converted into a structured engineering semantic format for the target mesh generation software to read and utilize, and then output.

[0193] This invention provides a fully automated and high-precision component identification and geometric feature extraction capability for mesh generation in computational fluid dynamics and finite element analysis, and is applicable to the digital simulation and manufacturing of complex equipment such as aviation, aerospace, and ships.

[0194] This invention provides a method for intelligent recognition, semantic segmentation, and geometric feature extraction of aircraft components based on a three-modal fusion of vision, geometry, and language. This invention addresses the inefficiencies, inconsistencies, and automation breakpoints caused by reliance on manual component recognition in the aircraft mesh generation process. Specifically, it includes:

[0195] (1) Eliminate the manual dependence in the component identification process and realize the automatic and accurate identification and semantic grouping of each component in the aircraft CAD model;

[0196] (2) Provide engineering semantic output that can be directly used for mesh generation, including component face sets and key feature lines (such as boundary lines and densification lines).

[0197] (3) Supports preset differentiated grid strategies for different components, enabling intelligent grid planning that is configured upon identification;

[0198] (4) Improve the automation level and processing efficiency of the entire mesh generation process, and provide reliable pre-processing tools for digital simulation.

[0199] This invention provides a joint modeling and processing method for a visual-geometric-linguistic three-modal approach. Its overall architecture comprises four parts: data construction, model training, inference and recognition, geometric processing, and mesh strategy generation. The specific steps are as follows:

[0200] (1) Data Construction: A Semantic Annotation System Oriented to Grid Requirements

[0201] 1) B-Rep Surface-Level Semantic Annotation: Using mesh generation software such as NNW-GridStar, and combining experience in aerodynamic and structural mesh generation, surface-level annotations are performed on aircraft CAD models represented in B-Rep format. This includes, but is not limited to: nose, forward fuselage, mid-fuselage, aft fuselage, upper wing surface, lower wing surface, vertical tail, horizontal tail, engine nacelle, etc. After annotation, each component is visualized in a different color for easy verification.

[0202] 2) Multi-view image generation: For each B-Rep face, perform 360-degree multi-view rendering around its normal vector to generate a series of main view images. Scaling, rotation, and lighting adjustments are supported to enhance view coverage and lighting robustness.

[0203] 3) Dataset Augmentation and Construction: Data diversity is enhanced using geometric transformations (rotation, translation, scaling), color perturbation, and noise injection. A structured multimodal dataset is constructed, with each data point including: B-Rep facet ID, multi-view image set, semantic labels, and original geometric parameters.

[0204] (2) Model training:

[0205] 1) Model Architecture Design: ViT (Vision Transformer) is used as the visual encoder to extract image features. CrossAttn (a cross-attention mechanism) is used to align visual features with linguistic features (from QwenLM). QwenLM serves as the language understanding and generation module, responsible for semantic mapping and prompt response.

[0206] 2) Supervised fine-tuning strategy: The input is the main view image of the face, the corresponding semantic label, and the structured prompt words (such as the face belongs to [category]).

[0207] 3) The loss function adopts a ternary combination: cross-entropy loss (optimizes classification accuracy) + IoU loss (improves the overlap of segmented regions) + boundary weighted loss (enhances the accuracy of boundary pixel recognition).

[0208] loss function for:

[0209] ;

[0210] in, For semantic cross-entropy loss, For IoU loss, For boundary-weighted loss, the weighting coefficients are initially set. , , Adjustments can be made based on subsequent experiments.

[0211] Semantic cross-entropy loss:

[0212] ;

[0213] In the formula, N is the total number of pixels, i.e., the total number of faces in the B-Rep model, C is the number of semantic categories of the component, and y i2,c For the true label of the i2th pixel belonging to category c, p i2,c This represents the probability predicted by the model that the i2th pixel belongs to class c.

[0214] The IoU loss is calculated by averaging the IoU for each part category:

[0215] ;

[0216] Boundary loss ( The initial setting is 0.3, which can be adjusted based on experiments. Boundary binary classification loss:

[0217] ;

[0218] In the formula, This represents the true boundary label of the i2th pixel. This indicates that the pixel is located inside the component. This indicates that the pixel is located on the geometric boundary of two different components; This represents the probability that a pixel predicted by the model is a boundary. Based on the topological relationships of the B-Rep model, when two adjacent faces belong to different components, their shared edge is marked as a boundary.

[0219] Loss for distance transformation:

[0220] ;

[0221] In the formula, The total number of boundary pixels, This represents the distance from pixel i2 to the nearest real geometric boundary. This represents the distance from pixel i2 predicted by the model to the boundary. This forces the predicted boundary lines to be closer to the actual geometric boundaries, improving the geometric usability of the segmentation results in CAD models.

[0222] 4) The training employs progressive fine-tuning, first pre-training on large-scale general image-text data, and then fine-tuning on domain image-text pairs labeled with semantic tags for each component of the aircraft. This approach can utilize prior knowledge of general visual language while enabling the model to accurately grasp the fine-grained semantics of the aircraft's unique components, thus improving domain adaptability and recognition accuracy.

[0223] (3) Reasoning and recognition:

[0224] 1) Input: The CAD model of the aircraft to be processed (B-Rep) and natural language prompts (e.g., please identify all fuselage surfaces).

[0225] 2) Layered recognition processing flow:

[0226] The first layer is the overall component recognition: the model extracts visual features face by face, combines them with the semantics of prompt words to make inferences, and prioritizes the identification of the main structural components, i.e., large components. The main structural components refer to the components with independent functions on the aircraft, such as the entire wing, the entire fuselage, and the tail. For simplicity, they will be referred to as large components in the following descriptions. The primary semantic labels of each B-Rep face are output to form a face-component mapping table.

[0227] The second layer is small component / feature geometric completion: For small surfaces (such as surfaces belonging to the trailing edge of an wing or local protrusions) whose confidence level is lower than the set threshold or ignored by the model in the previous round of recognition, the geometric rule assistance module is activated instead of relying directly on the AI ​​model for secondary recognition.

[0228] Widgets / features mainly include the following types:

[0229] Connecting feature surfaces or edge surfaces: Surfaces located at the junction of two large components (such as the upper and lower surfaces of an wing), such as the trailing edge and leading edge of an wing.

[0230] Featured surfaces with opening, closing, or maintenance functions: surfaces surrounded by large components and whose orientation differs significantly from the main body, such as hatches and maintenance access covers.

[0231] Localized raised or recessed surfaces: such as rivet heads, sensor mounting bases, oil filler caps, and other small geometric features.

[0232] Low-confidence recognition surfaces: Surfaces that AI models cannot recognize with high confidence due to reasons such as blurred geometric features and insufficient training data coverage.

[0233] 3) Working principle of the geometric rule auxiliary module:

[0234] Input: The set of large components identified by AI (such as the upper surface of the left wing and the lower surface of the left wing) and all the B-Rep surfaces contained therein.

[0235] Processing: Based on the geometric topology and parametric information of the CAD model, preset geometric rules are executed. For example, Rule 1: If a surface belongs to the upper or lower surface of the wing, or is located at the boundary between the two, at the leading edge of the wing, has high curvature, and its spanwise length is close to the entire wing span, it is classified as the leading edge of the wing. Rule 2: If a surface is located at the boundary between the upper and lower surfaces of the wing, and its area is less than 1% of the overall wing area, it is classified as the trailing edge of the wing. Rule 3: If a surface is directly adjacent to the main wing but its normal is significantly deviated from the main wing plane, and it is located at the outermost point of the wing spanwise, with an area between 0.5% and 5% of the main wing area, it is classified as a winglet or wingtip. Rule 4: If a surface is surrounded by a large fuselage component, and its normal is approximately perpendicular to the main fuselage direction, and its area is small, it may be an access hatch or door, and further determination is made based on the geometric characteristics of its adjacent surfaces. Rule 5: If a surface belongs to a large component of the engine nacelle, is located at the foremost end of the nacelle, and has a closed annular boundary and a smooth Gaussian positive curvature transition, then it is classified as an air intake lip.

[0236] Output: Add annotations to the corresponding part categories for facets that conform to the geometric rules, or create new subcategory labels for them.

[0237] 4) Post-processing: Label smoothing is performed based on face adjacency relationships, integrating AI recognition results with geometric rule supplementary results to avoid misjudgment of isolated faces. Component-level result visualization and export are supported.

[0238] (4) Geometric treatment:

[0239] 1) Key Geometric Feature Line Extraction: Based on the complete identification of all components (including large components identified by AI and small features supplemented by geometric rules), boundary lines, curvature extrema lines, and feature contour lines of each component are extracted using B-Rep topology and geometric information. For large components identified by AI, geometric algorithms (such as curvature-based edge detection and topological connectivity analysis) are used for automated extraction. For small features identified by the geometric rule-assisted module, feature lines are extracted directly using their geometric definitions (such as the trailing edge line, i.e., the intersection of the upper and lower wing surfaces), resulting in higher accuracy.

[0240] 2) Output Interface: Feature lines are output in a format supported by NNW-GridStar. Seamless integration with simulation workflows is supported, improving CAE analysis efficiency.

[0241] The visual encoder (ViT) can be replaced by other mainstream networks (such as ConvNeXt, ResNet), and this embodiment of the invention does not impose any restrictions on it. The language model (QwenLM) can be replaced by other large language models (such as GLM, LLaMA), and this embodiment of the invention does not impose any restrictions on it. The specific judgment conditions and parameters of the geometric rules can be adjusted according to the characteristics of different components. The output interface can be adapted to different target mesh generation software (such as ANSA, HyperMesh).

[0242] Example 3;

[0243] Based on Example 2, please refer to Figure 5 , Figure 5 This is a schematic diagram of the architecture of an intelligent component recognition and segmentation system for aircraft CAD models based on three-modal fusion. Embodiment 3 of this invention provides an intelligent component recognition and segmentation system for aircraft CAD models based on three-modal fusion. The system adopts a layered architecture and aims to serve the automated mesh generation in aircraft digital simulation, including:

[0244] The data construction layer, configured to build a multimodal training dataset oriented towards grid engineering semantics, includes:

[0245] Semantic annotation unit is used to perform engineering semantic annotation on the faces in the boundary representation structure of the aircraft CAD model and generate component semantic tags;

[0246] A geometric parameter extraction unit is used to extract the geometric parameters of each face from the boundary representation structure;

[0247] The multi-view rendering unit is used to generate multi-view 2D rendering surface images for each labeled surface;

[0248] The data management unit is used to associate and store the surface image, the corresponding geometric parameters, and the semantic labels of the components to form structured training samples;

[0249] A model training layer, connected to the data construction layer, is configured to train a multimodal recognition model using the multimodal training dataset. This layer includes:

[0250] The model architecture management unit is used to build and maintain a neural network architecture that includes a visual encoder, a language understanding module, and a cross-modal fusion module.

[0251] A training execution unit is used to perform supervised training on the neural network architecture by inputting the training samples;

[0252] An optimization unit is used to optimize the multimodal recognition model using a combined loss function;

[0253] The inference and recognition layer, connected to the model training layer, is configured to load the trained multimodal recognition model and perform intelligent recognition and segmentation on the input aircraft CAD model. This layer includes:

[0254] The first recognition sublayer is used to run the multimodal recognition model, perform preliminary recognition of the surfaces of the CAD model, and output preliminary surface-part semantic label mapping relationships;

[0255] The rule-assisted sublayer, as a geometric rule-assisted module, is used to determine, complete, and correct semantic labels for low-confidence or unrecognized surfaces output by the first recognition sublayer based on predefined geometric topology and parameterized rules.

[0256] The post-processing sub-layer is used to smooth and integrate the recognition results based on the face adjacency relationship, forming the final complete component segmentation result;

[0257] A geometry processing and output layer, connected to the inference and recognition layer, is configured to perform engineering processing and output of the recognition results. This layer includes:

[0258] The feature line extraction unit is used to automatically extract the key geometric feature lines corresponding to each component from the final component segmentation result based on the boundary representation structure of the CAD model;

[0259] The format conversion and output unit is used to convert the component segmentation results and key geometric feature lines into a structured engineering semantic format that can be read by the target mesh generation software and output it.

[0260] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0261] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for predicting a grid node distribution and anisotropy parameter of an aircraft component boundary line, characterized in that, The method comprises: Step 1: determining the geometric data of an aircraft CAD model, component semantic labels, a calculation working condition, and each boundary line on the surface of the aircraft CAD model; wherein the component semantic labels are used to identify the functional categories to which each component in the aircraft CAD model belongs, which are associated with a mesh division strategy; Step 2: based on the geometric data, the component semantic labels, and the calculation working condition, using a trained neural network model to classify the mesh region types of each boundary line on the surface of the aircraft CAD model, to obtain a mesh region type classification result to which each boundary line belongs; the mesh region types include isotropic regions and anisotropic regions; Step 3: according to the mesh region type classification result, using a differentiated parameter prediction subnetwork corresponding to each mesh region type in the neural network model to predict a corresponding mesh control parameter set for each boundary line; Step 4: based on the mesh control parameter set, driving a mesh generator to generate a surface mesh of the aircraft CAD model.

2. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of claim 1, wherein, The neural network model is a multi-task graph neural network model; in step 2, the mesh region types of each boundary line on the surface of the aircraft CAD model are classified as follows: A boundary line topology graph is constructed with the boundary lines as nodes and the topological connection relationship between the boundary lines as edges; The boundary line topology graph and the feature vectors of each node are input into the multi-task graph neural network model; wherein the feature vectors of each node are formed based on the geometric features of the corresponding boundary line extracted from the geometric data, the semantic labels of the component, and the calculation working condition; Using a mesh region type classification head in the multi-task graph neural network model to perform binary classification on each node in the boundary line topology graph, and outputting the probability that each boundary line belongs to an isotropic region or an anisotropic region as the mesh region type classification result.

3. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of claim 2, wherein, Step 3 is specifically as follows: according to the mesh region type classification result output by the mesh region type classification head, the multi-task graph neural network model routes the boundary line node features belonging to the isotropic region to a first parameter prediction subnetwork to predict a first mesh control parameter set containing the node total number and node size distribution function parameters; and routes the boundary line node features belonging to the anisotropic region to a second parameter prediction subnetwork to predict a second mesh control parameter set containing the node total number, node size distribution function parameters, first layer mesh height, mesh growth rate, and growth layer number.

4. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of claim 1, wherein, The determination of each boundary line on the surface of the aircraft CAD model is as follows: An input aircraft CAD model is processed to obtain a component segmentation result and key geometric feature lines; wherein the component segmentation result defines the correspondence between each surface of the aircraft CAD model and the component semantic labels, and the key geometric feature lines at least include the boundary lines between all components; All boundary lines between components are selected from the key geometric feature lines as each boundary line on the surface of the aircraft CAD model.

5. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of claim 4, wherein, The process of processing the input aircraft CAD model to obtain component segmentation results and key geometric feature lines specifically includes: Step S1: Obtain the CAD model of the aircraft to be processed, wherein the CAD model of the aircraft has a boundary representation structure; Step S2: Based on the engineering requirements of mesh generation, perform engineering semantic annotation on the faces in the boundary representation structure of the aircraft CAD model to obtain component semantic labels; extract the geometric parameters of each face from the boundary representation structure; generate a multi-view 2D rendering image for each annotated face; associate and store the multi-view 2D rendering image, the geometric parameters, and the component semantic labels to form training samples. Step S3: Train a multimodal recognition model using the training samples. The multimodal recognition model is configured to fuse input image features and language semantic information to output semantic labels for components on the surface of the aircraft CAD model. Step S4: Input the aircraft CAD model to be identified into the trained multimodal recognition model, identify the surfaces in the aircraft CAD model, and obtain the preliminary surface-part semantic label mapping relationship; Step S5: Based on the preliminary face-part semantic label mapping relationship, for faces whose recognition confidence output by the multimodal recognition model is lower than a preset threshold or whose geometric area ratio is less than a set threshold, the geometric rule auxiliary module is activated to perform semantic completion to obtain the final part segmentation result. Step S6: Based on the final component segmentation result, extract the key geometric feature lines corresponding to each component from the boundary representation structure of the aircraft CAD model.

6. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of claim 4, wherein, The semantic labels of components are determined as follows: based on the component segmentation results, the semantic labels of each surface of the aircraft CAD model are determined; the semantic labels of components include one or more of the following: wing leading edge, wing upper surface, wing lower surface, fuselage nose and engine nacelle.

7. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of claim 2, wherein, The multi-task graph neural network model was obtained through the following offline training method: A training dataset is constructed, wherein each sample data in the training dataset corresponds to a boundary line in a historical successful computation case. The sample data includes input features and output labels. The input features include the geometric features, component semantic labels, and computational conditions of the boundary line. The output labels include the optimal grid region type label of the boundary line and the corresponding optimal grid control parameter label. The optimal grid region type label is used to supervise the obtaining of the grid region type classification result, and the optimal grid control parameter label is used to supervise the obtaining of the grid control parameter set. Based on the training dataset, the constructed multi-task graph neural network model is trained end-to-end; wherein, the loss function of the multi-task graph neural network model is a joint loss function, which includes a classification loss for optimizing grid region type classification, a first regression loss for optimizing isotropic parameter prediction, and a second regression loss for optimizing anisotropic parameter prediction.

8. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of claim 2, wherein, The geometric features include at least one of the length and curvature information of the boundary line; the calculation conditions include at least one of the flight Mach number and the flight angle of attack.

9. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of claim 3, wherein, The node size distribution function parameters are used to control the density distribution of grid nodes along the length of the boundary line.

10. The aircraft component boundary line mesh node distribution and anisotropy parameter prediction method of Claim 1, further characterized by, In step 4, based on the set of mesh control parameters, the mesh generator is driven to generate the surface mesh of the aircraft CAD model, specifically as follows: The predicted set of mesh control parameters for each boundary line is converted into the corresponding control script or API instructions for the target mesh generation software. The control script or API instructions are executed to drive the target mesh generation software to complete the discretization of all boundary lines, anisotropic mesh advancement and surface triangulation, and output the final surface mesh file.

Images