799 results about "Mesh generation" patented technology

The invention relates to the technical field of crustal stress field data processing, in particular to a crustal stress field three-dimensional dynamic inversion method based on a multi-scale adaptive algorithm. The method comprises the following steps: acquiring a geological data set of a target area; constructing a crustal stress field three-dimensional initial model based on the geological data set, and performing geologic body space division and mesh generation to obtain crustal stress field three-dimensional mesh model data; performing multi-scale region division on the crustal stress field three-dimensional grid model data, and establishing a multi-scale weighting function to obtain multi-scale partition mapping information; and constructing a cross-scale boundary adaptive transmission mechanism, and establishing a stress tensor continuity constraint model at a multi-scale partition boundary to obtain cross-scale stress boundary coupling data. Through a multi-scale adaptive algorithm and dynamic closed-loop optimization, high-precision, dynamic and continuous inversion of a crustal stress field in a complex geologic structure is realized.

The invention provides a 3D model multi-scale lightweight method and system based on topological optimization, and relates to the technical field of model lightweight, and the method comprises the steps: carrying out the mesh generation of a target 3D model, and generating a finite element model; a multi-scale topological optimization model is established, density distribution is adjusted through a density filtering function, and a macroscopic lightweight structure is obtained in combination with minimum size constraint and gray scale gradient projection processing; based on a macroscopic lightweight structure, a skeleton structure is obtained, based on an adaptive polygon segmentation rule, a porous filling unit is constructed, a multi-parameterized model is fused, a parameterized representation method of three variables of density, size and shape is established, and an intermediate scale structure is obtained through a multi-target genetic algorithm in combination with a sensitivity analysis technology; and embedding the intermediate-scale structure into the macroscopic lightweight structure, generating a multi-scale structure in combination with cascade optimization, calculating equivalent material attributes by adopting a homogenization method, iteratively adjusting model parameters, and finally obtaining a multi-scale lightweight result.

The invention relates to the technical field of welding quality evaluation, in particular to a carbon steel pipe welding quality evaluation system which comprises a welding area three-dimensional scanning module, a welding seam dynamic grid generation module, a welding form change analysis module, a welding seam structure diagram construction module, a welding seam quality diagram convolutional network module and a welding defect intelligent recognition module. According to the invention, through real-time adjustment of the dynamic grid generation module, tiny changes of the geometric shape of the welding seam can be captured, new dimensions are provided for welding quality evaluation through generation of welding form characteristic data and construction of the welding seam structure diagram, and the introduction of the welding seam quality diagram convolution network module can improve the accuracy of welding quality evaluation. Node relations and structural features in a weld joint diagram structure are analyzed through a deep learning technology, an intelligent means is provided for welding quality evaluation, an intelligent welding defect identification module optimizes an identification process, various welding defects can be accurately classified, and a powerful technical support is provided for guarantee of welding quality.

The invention discloses a four-dimensional Gaussian model generation method, system and equipment based on Gaussian sputtering, and belongs to the field of computer vision processing. The method comprises the following steps: firstly, generating a target video consistent with a given text prompt by using a pre-trained text-to-video diffusion model, then converting the target video into a three-dimensional grid model frame by frame by using an image-grid generation model, and rendering the three-dimensional grid model in a focal length range under a main view angle, the method comprises the following steps: firstly, calculating a rendering matching focal length, training a static Gaussian model at an initial moment by adopting a Gaussian sputtering method, then training a deformation network, and finally generating a four-dimensional Gaussian model under a given timestamp through the trained static Gaussian model and deformation network. By means of the method, the vivid high-dynamic 4D model which is fine in texture, accurate in geometrical relation and capable of accurately conforming to the semantics of the characters input by the user can be generated within obviously short time.

The invention belongs to the technical field of electromagnetic modeling of a target in a layered medium, particularly provides a planar multilayer medium electromagnetic scattering modeling method based on a layered medium Green function, and is used for solving the problems that the number of unknown quantities of a system is large and the calculation efficiency is low when electromagnetic scattering characteristics of the target are analyzed in the layered medium. The method comprises the following steps: firstly, carrying out mesh generation on a medium target in a planar multilayer medium; then, a layered medium Green function is calculated based on a space list and an interpolation method, a surface integral equation of the target is solved by adopting a moment method, and a matrix equation of target current and magnetic current is constructed; and finally, solving a matrix equation by applying an improved laminated off-diagonal low-rank matrix, and calculating a target scattering field. According to the method, a direct solving calculation method is adopted, and the problem that an ill-conditioned matrix is difficult to converge in a conventional iteration method is effectively solved; especially for a multi-port excitation problem, repeated calculation is not needed, the solving efficiency of a large-scale problem is improved, and calculation resource consumption is reduced.

The invention provides an airfoil flow field prediction method based on multi-task learning, and the method comprises the steps: firstly building an airfoil flow field prediction model based on multi-task learning, which comprises an encoder, a backbone network and a multi-head decoder; then training the established model by using an airfoil flow field data set and adopting an optimization strategy based on multi-task learning; and finally, performing geometric parameterization and grid generation processing on an airfoil which actually needs flow field prediction to obtain standardized airfoil flow field data, and inputting the airfoil flow field data into the trained airfoil flow field prediction model for prediction to obtain full flow field physical quantity distribution and lift-drag coefficient information. According to the method, a multi-task loss optimization strategy is used in the prediction model training process, so that the problem of conflict between airfoil profile surface loss and volume loss optimization can be effectively solved, and then the construction of an airfoil profile proxy model and the accurate prediction of flow field information and aerodynamic parameters are realized.

The invention provides a customized energy-saving air conditioner control method and device for industrial process requirements, and is applied to the technical field of data processing. According to the method, electromagnetic interference and vibration noise are eliminated through Kalman filtering, and a standardized process-environment-energy consumption correlation sequence is generated; and converting the three-dimensional process load map into a three-dimensional process load map, and establishing an individualized air conditioner dynamic load prediction model in combination with heat and humidity characteristics of a scene through fluid dynamic simulation and self-adaptive grid division. Extracting process priority factors to construct an adjustment matrix, calculating an air conditioner parameter combination under target energy consumption, extracting energy consumption characteristics, dividing standard exceeding risk levels, and constructing a multi-dimensional characteristic matrix; and comparing real-time data with historical data to identify abnormity, and generating an energy consumption optimization correction factor. Users are grouped according to enterprise conditions, key factors are screened by using a gradient boosting tree, a personalized energy-saving control model is constructed by fusing multiple information, target parameters are output, and a real-time adjustment instruction and a time-phased energy-saving strategy are generated in combination with a time sequence.

The invention discloses a curved surface grid generation and self-adaptive re-division method for fluid simulation. The method comprises the following steps: acquiring CAD fluid model data; processing closed surface, short edge and parameter domain distortion and geometric interference problems of the CAD fluid model to obtain a defect-free continuous geometric model; performing curved surface discretization processing on the defect-free continuous geometric model to generate an initial grid; and performing grid re-division processing on the initial grid to generate an optimized anisotropic curved surface grid. The problems of high geometric defect sensitivity, large isotropic grid redundancy and sudden size change of the tail end of a parameter domain in related technologies of fluid simulation are solved.

The invention discloses a curved surface quadrilateral dominant grid generation method, device and equipment for an aircraft and a storage medium, and relates to the technical field of computational fluid mechanics, and the method comprises the steps: creating a surface unstructured grid based on surface grid generation parameters of a CAD file of the aircraft, carrying out the conformal mapping of the surface unstructured grid to a parameter domain plane, and calculating the vertex compression ratio of each grid vertex; selecting the current starting point with the minimum compression ratio and the maximum to-be-compared vertex in the adjacency list, and determining the current to-be-processed vertex with the minimum compression ratio in the adjacent vertexes and the advancing direction; determining to-be-processed discrete points according to the current compression ratio and the coordinate information to generate a current isoline, if the current compression ratio does not exceed a threshold value, iteratively updating the current starting point, and skipping to the step of determining the current to-be-processed vertex again; the discrete points are projected to the outer-layer isolines according to the isoline sequence, and the curved-surface quadrilateral dominant grid is generated based on the distance proportion between each discrete point and the projection point, so that the generation efficiency of the curved-surface quadrilateral grid is improved.

The invention discloses a porous structure generation method and a streamer simulation method thereof, and relates to the technical field of porous structure generation and simulation, and the porous structure generation method comprises the following steps: defining and determining phase parameters of a porous structure; selecting nodes with a preset proportion from all nodes in the set area, setting the nodes as pore seed nodes, and simulating the growth process of pores by using a result of numerical iteration solution processing based on initial values of phase parameters of the set nodes; and constructing a porous structure model. The porous structure streamer simulation method comprises the following steps: removing pores which are not communicated with the surface of a porous structure to obtain a porous solid medium model; setting parameters of gas, and performing transient solution on the discrete electron diffusion model of streamer development; and according to the transient development process of the streamer, analyzing and obtaining the influence of the pore structure on the development characteristics of the streamer. By applying the method provided by the invention, the mesh generation density in porous structure model simulation is reduced, and the efficiency of streamer simulation is improved.

The invention provides an image processing method, device and equipment for scene reconstruction and rendering. The method comprises the following steps: initializing sparse point clouds corresponding to each scene image to be reconstructed to obtain a plurality of initial Gaussian surface elements; optimizing each initial Gaussian surface element based on an inverse depth map, a rendering scene image and a normal vector map obtained by rendering the initial Gaussian surface elements to obtain a plurality of first target Gaussian surface elements; based on the importance of the first target Gaussian surface elements, determining a second target Gaussian surface element, the importance being determined based on a plurality of sub-models and a target scene image to be reconstructed matched with each sub-model, the plurality of sub-models being obtained by performing mesh generation on the plurality of first target Gaussian surface elements, and the plurality of sub-models being obtained by performing mesh generation on the plurality of first target Gaussian surface elements; the target scene image to be reconstructed matched with each sub-model is determined based on the pose of the scene image to be reconstructed and the rendering result of the sub-model; and determining a third target Gaussian surface element for rendering the reconstructed scene image based on the spatial range of the grid.

The invention provides a three-dimensional object collision detection parcel grid generation method and system, a storage medium and a program product, and the method comprises the steps: extracting all vertex data of a first object according to a sub-Mesh data set after serialization of a three-dimensional model, and determining an initial center point; each vertex of each piece of sub Mesh data is converted from a local coordinate system to a global world coordinate system, and a point cloud data set is generated; constructing an initial convex hull by adopting a matrix hierarchical extremum retrieval algorithm, and generating a final convex hull by adopting a conflict graph and incremental convex hull algorithm; initializing a radius coverage model by taking the geometric center of the model as a sphere center; performing subdivision according to longitude and latitude to generate a hierarchical ball grid; emitting rays, correcting vertex positions layer by layer, approaching the surface of the model, and outputting a collider network. Through the overall point cloud set, the incremental convex hull algorithm and the spherical grid contraction technology, unified topological wrapping of the multiple three-dimensional objects is achieved, the visual fidelity and the collision precision are ensured, and resource consumption is reduced.

The invention provides a boiler combustion state visualization method, which comprises the steps of embedding a plurality of thermocouple temperature sensors in gaps of refractory bricks on the inner wall of a boiler hearth, acquiring temperature data of the hearth at different depths in real time, and transmitting the temperature data to a data acquisition terminal through a wireless communication module; fitting the normalized temperature data matrix by adopting a three-dimensional thin-plate spline interpolation algorithm to generate a hearth three-dimensional real-time temperature distribution model, performing mesh generation on the three-dimensional temperature distribution model by combining hearth geometric structure parameters obtained from computer aided design software, and constructing a hearth temperature finite element model; the hearth temperature finite element model is imported into ANSYS finite element analysis software, hearth wall surface material attributes and combustion heat load boundary condition parameters are set, transient thermal analysis of a hearth temperature field is carried out, and a temperature estimation value of any position of the hearth changing along with time is obtained.

The invention provides a drawing simulation method for an anisotropic material, and belongs to the technical field of drawing simulation, and the drawing simulation method comprises the following steps: firstly, measuring a microstructure structure of the anisotropic material, and constructing a constitutive model of a dynamic self-adaptive material; importing the constitutive model into finite element analysis software to form a finite element model of multi-physical field interaction behaviors; importing the finite element model into an adaptive mesh generation algorithm to form a finite element simulation result; performing real-time analysis on a finite element simulation result by using a deep learning algorithm, and outputting an analysis report in a visual chart form; and the die fillet radius, the drawing gap, the blank holder force, the drawing speed, the friction coefficient and the electromagnetic regulation and control parameters are synergistically and dynamically optimized, and the lug rate of the cup-shaped part is obtained according to drawing simulation. According to the deep drawing simulation method for the anisotropic material, the convex lug rate of the cup-shaped part is obtained by constructing the constitutive model and forming the finite element result of the finite element model, so that the convex lug problem of deep drawing of the cup-shaped part is predicted.

Building on U.S. Provisional Patent Application No. 63 / 693,803, paragraphs for 2D-to-3D conversion and [0023]-[0024] for secure content verification, the Matrix 3D AI Polygon Mesh Hash Key System revolutionizes transforming 2D content into secure, interactive 3D AR / VR assets. Integrating AI and LiDAR, it enables real-time 3D mesh generation with precise geospatial anchoring. Users can reshape 3D content instantly via natural language commands, enhancing interactivity. Quantum-resistant encryption, using AES-256 and CRYSTALS-Kyber, ensures robust asset security. The system excels in gaming, surveillance, content verification, military operations, space exploration, deepfake detection, and pharmaceutical research, offering unmatched precision and scalability. Its multi-stage rendering pipeline, powered by distributed AI-driven bots, supports efficient real-time 3D mesh generation, achieving 95% accuracy and 1 cm resolution. This AR / VR technology advancement transforms industries with scalable, secure solutions for interactive 3D content creation and management, setting a new standard for precision and versatility.

The invention belongs to the technical field of simulation analysis, and particularly relates to a medium-high voltage switch cabinet internal humidity inversion calculation method based on simulation analysis, which comprises the following steps: establishing a geometric model of a switch cabinet, importing the geometric model into finite element simulation software to obtain a simulation analysis model, adding each physical field and computational domain and boundary conditions thereof, and calculating the internal humidity of a medium-high voltage switch cabinet. All the physical fields are coupled, and mesh generation is carried out; performing numerical solution on the simulation analysis model by using a transient solver to obtain a temperature and humidity distribution condition in the switch cabinet; the simulation analysis model is optimized, a simulation data set is constructed, and a PINN is trained; based on the environment temperature and humidity and the operation current collected in the actual operation of the switch cabinet, the trained PINN is used for prediction, and the steady-state relative humidity of the current transformer is obtained. According to the invention, based on the limited boundary monitoring data, efficient and accurate reconstruction of the humidity field of the key area in the switch cabinet is realized, and a feasible path is provided for equipment state sensing and condensation risk early warning.

The invention discloses a non-uniform and conformal grid-oriented electromagnetic particle adaptive simulation method and system, and relates to the technical field of electromagnetic particle simulation. The method comprises the following steps: performing non-uniform right-angle mesh generation based on a charged particle average free path, and identifying conformal mesh units; carrying out initialization by adopting macro particles; executing electromagnetic field propulsion, field interpolation, particle propulsion, charge current backfilling and collision simulation in each time step; and accelerated calculation is realized through a CPU and GPU hybrid parallel architecture. Through the method, the numerical simulation precision can be improved, the adaptability of complex geometry is enhanced, the coupling stability of the electromagnetic field and the particles is ensured, the information interaction between the field and the particles is improved, and the large-scale calculation efficiency is improved.

The invention relates to a lung region electrical impedance tomography (EIT) image reconstruction method based on a graph structure, and establishes an EIT depth imaging model in which double branch channels of a high-frequency channel graph attention network and a low-frequency channel graph attention network are parallel, and the model is named as HLGAT. The method comprises the following steps: injecting current excitation into a finite element simulation model of a lung region to obtain boundary voltage data required by EIT image reconstruction, and performing non-uniform mesh generation on the lung region needing image reconstruction to obtain nodes required by a graph neural network and an adjacency relationship between the nodes, so as to obtain a lung region model; storing the data in a graph structure form suitable for network input; the HLGAT aggregates the low-frequency characteristic information and the high-frequency characteristic information through the two channels, node output characteristic information of the two channels is fused in a learnable mode, a new adjacency matrix is reconstructed according to the node characteristic information output by the network, then a new Laplace matrix is constructed, more accurate conductivity distribution can be obtained, and the accuracy of the conductivity distribution is improved. And the physical process of electrical impedance imaging can be described more accurately, so that the accuracy of model imaging is improved.

The invention discloses a three-dimensional curved surface pattern layering processing method based on parameterized segmentation, and aims to solve the problem of high-precision pattern processing on the surface of a curved surface by using a laser scanning galvanometer. The method comprises the following steps: firstly, importing a three-dimensional model, and carrying out parameterization processing to obtain a two-dimensional parameter domain; positioning a to-be-processed plane pattern in the two-dimensional parameter domain to form a complete two-dimensional geometric grid covering a pattern area, generating a three-dimensional pattern grid attached to a curved surface, and calculating a center normal vector of the three-dimensional pattern grid; then, generating an equidistant plane group on the surface of the curved surface, and calculating an intersection line of the plane and the three-dimensional pattern grid as a laser scanning track; and the five-axis linkage platform-galvanometer system is started for machining. Through the parameterization domain, distortion of curved surface pattern machining is effectively solved, machining uniformity is improved, and the method is particularly suitable for laser machining of large-scale or high-complexity curved surface patterns.

A multi-scale coupling analysis method for a high-temperature liquid metal and jet water interaction system mainly comprises the following steps as shown in an attached drawing: 1, carrying out geometric modeling and grid division on a liquid metal pool type space, jet water and other pipeline systems, and setting a valve action sequence; 2, setting a coupling boundary, and establishing an IPC communication pipeline; 3, performing steady-state calculation, and mutually providing initial boundaries for overall initialization; 4, a system program is imported into the boundary of the two-dimensional fluid domain, the fluid state in the pipeline under the decompression heating condition is calculated, and boundary information in a result is written into the IPC pipeline; 5, the two-dimensional fluid program reads source items from the system program, thermodynamic parameter distribution of the molten pool is calculated, and boundary information is provided; and 6, judging the action condition of the component at the current moment and repeating the step 4 and the step 5 until the calculation time reaches the end moment. The method provided by the invention can simulate the whole interaction process of liquid metal and jet water and embody the influence of flash evaporation in a critical flow jet boundary and an injection pipeline on jet.

Certain aspects of the present disclosure provide techniques and apparatus for improved machine learning. In an example method, a multiview latent tensor is generated based on processing a textual input using a diffusion machine learning model, where the multiview latent tensor corresponds to a plurality of orthographic projections corresponding to the textual input. A triplane latent tensor is generated based on the multiview latent tensor using a conversion machine learning model, and a three-dimensional mesh is generated based on processing the triplane latent tensor using a decoder machine learning model.

The invention discloses an electromagnetic field simulation grid adaptive generation method based on a neural network. The method comprises the following steps: performing mesh generation on a semiconductor device simulation model to obtain original generation data, and constructing node features; an undirected graph is constructed, graph nodes of the undirected graph adopt node features of grid points, and connecting edges adopt Euclidean distances between the grid points and adjacent grid points; inputting the undirected graph into a double-branch neural network, and outputting a predicted node position; and correcting each node of the original subdivision data by using the predicted node position obtained by the double-branch neural network to form new subdivision data for electromagnetic field simulation. According to the method, physical field information can be fused to efficiently adjust the node density, and the geometric boundary and topological structure characteristics of the device can be strictly kept.

The invention belongs to the technical field of hydraulic structure engineering, and relates to a dam foundation curtain anti-seepage performance evaluation method and system under the action of seepage corrosion, and the method comprises the steps: drawing a hydropower station dam engineering riverbed cross section geometric model, and carrying out the mesh generation of the geometric model to obtain a calculation model; constructing a dam foundation anti-seepage curtain corrosion degradation analysis initial model; coupling a seepage field, a calcium ion transport field and a corrosion field in the dam foundation anti-seepage curtain corrosion degradation analysis initial model by utilizing the calculation model to obtain a permeability coefficient, porosity and a diffusion coefficient; updating parameters to obtain a dam foundation anti-seepage curtain corrosion degradation analysis updating model; whether the permeability coefficient, the porosity and the diffusion coefficient converge or not is judged, and if yes, the anti-seepage curtain structure calcium ion concentration, the solid-phase calcium concentration and the water head corresponding to the parameters are output; evaluating the anti-seepage performance of the dam foundation curtain; and otherwise, carrying out coupling solution on the analysis updating model by utilizing a calculation model. The technical problem of inaccurate calculation results in the prior art is solved.

The invention relates to the technical field of face verification, and discloses a face identity verification data processing method based on dynamic feature extraction, which comprises the following steps: acquiring a continuous face video frame sequence and calculating a full-pixel instantaneous velocity vector to generate an original dense optical flow field, selecting rigid region anchor points to calculate a rigid affine transformation matrix and construct a theoretical rigid motion field, performing differential stripping on the theoretical rigid motion field from the original dense optical flow field, and extracting a non-rigid micro-motion residual field; mapping the non-rigid micro-motion residual field to a facial muscle topological grid to generate a time sequence feature tensor, and calculating a geodesic line distance between a covariance matrix of the time sequence feature tensor and a reference dynamic feature in a Riemannian manifold space; when the geodesic distance is smaller than a threshold value, verification is passed, through a rigid-non-rigid orthogonal decomposition mechanism, the special viscoelastic micro-motion and cooperation law of biological soft tissue is captured by utilizing a residual field, and the high-simulation mask is effectively defended.

The invention discloses a 3D Gaussian-based dressing human body geometry and texture parameterization reconstruction method. The method comprises the steps of generating human body and clothing grids of each frame of an RGB video based on a human body parameterization model SMPL and a clothing parameterization model TailorNet; generating a subdivision grid based on the human body and clothing grid, and generating an initial 3D Gaussian ball by taking the subdivision grid as a reference; training a Gaussian estimation network, and rendering by using a linear skin and 3D Gaussian splashing method; a motion sequence is given, and a dressed human body under the motion sequence is rendered by using a Gaussian estimation network. The invention provides the dressing human body reconstruction method which is more fit with the clothing and the human body model, the dressing human body rendering result under the given motion sequence can be output after the network training is completed, and the real and natural clothing wrinkle geometry and texture can be reconstructed.

The invention discloses a Chinese character structure and geometric space interactive writing method, which relates to the technical field of interactive writing methods, and comprises the following steps: carrying out geometric classification processing on Chinese characters by adopting a structural analysis method to obtain a Chinese character type set comprising a left-right structure, an up-down structure and a surrounding structure; carrying out geometric rule definition processing on the obtained Chinese character type set to obtain a geometric constraint rule base containing component proportion, gravity center alignment and stroke topological relation; performing structured storage processing on the obtained geometric constraint rule base by adopting a database construction method to obtain a dynamically-called Chinese character writing template database; the writing process of the user is monitored in real time, and a writing feature data set containing stroke coordinates, writing speed and pressure data is obtained; performing matching processing on the obtained writing feature data by adopting a dynamic grid generation method to obtain a self-adaptive reference grid corresponding to the currently written Chinese character; and carrying out deviation detection processing on the obtained self-adaptive reference grid by adopting a computer vision method.

The invention relates to the technical field of digital twinning, particularly provides a digital twinning construction method based on multi-source point cloud data fusion, and solves the problems of low multi-source point cloud data fusion precision, low calculation efficiency and poor adaptability to complex scenes in the prior art. The method comprises the following steps: S1, preprocessing multi-source point cloud data: carrying out data denoising and coordinate unified processing on the point cloud data of each source; s2, designing a point cloud data fusion algorithm: performing feature extraction and description, feature matching and fusion decision and weighted fusion calculation on the preprocessed point cloud data to obtain fused point cloud data; s3, constructing and optimizing a digital twinborn model: performing triangular mesh generation, model optimization processing, texture mapping and attribute addition on the fused point cloud data, and constructing the digital twinborn model; the digital twinning construction method based on multi-source point cloud data fusion can effectively solve the problems that multi-source point cloud data fusion is low in precision, low in calculation efficiency and poor in adaptability to complex scenes.

The invention discloses a computational fluid mechanics-oriented parallel boundary layer grid generation method and device. The method comprises the following steps of: firstly, generating an initial background grid with boundary layer units according to an original CAD (Computer Aided Design) geometric model; then, according to the initial background grid and the topological connection relation thereof, obtaining an object plane boundary of the background grid and grid units connected with the object plane boundary so as to construct a fixed closed area; performing regional decomposition on the initial background grid by using a graph division tool; secondly, encrypting curved surface grids of the sub-regions; carrying out parallel calculation on normal vectors of curved surface grid points and a stratified progress long distance according to the boundary layer unit, and generating a triangular prism unit by applying a stratified progress method; and finally, carrying out quality detection and intersection judgment on all boundary layer units to obtain a complete boundary layer grid. According to the method, the defect of lack of a boundary layer grid parallel generation method in large-scale grid generation can be overcome, rapid parallel generation of large-scale boundary layer grids is realized, and the performance bottleneck in the aspect of generation efficiency is overcome.

The invention relates to the field of biological modeling, and discloses a shoulder joint stress simulation method and system based on a three-dimensional finite element. The method comprises the following steps: acquiring three-dimensional geometric data of a shoulder joint through scanning equipment, and generating an initial grid model; dividing the preliminary calculation grid to determine an initial stress value, simulating stress transfer distribution, and generating a preliminary stress distribution field; calculating the stress gradient change rate of each grid unit and performing local subdivision to generate a stress gradient change rate distribution diagram; performing density subdivision on the grid units exceeding the change threshold to generate a refined grid model; the calculation grid is divided again, the stress value of each grid unit is calculated, and an updated stress field is generated; calculating a difference vector, judging whether the difference vector meets a convergence condition or not, and obtaining a convergence judgment result; and if not, iteratively subdividing the local grid to generate a final stress distribution field. Through adaptive grid subdivision and iterative convergence optimization, the problem of insufficient simulation precision is solved, and the accuracy is improved.

A system for generating 3D indoor scenes from text input is provided. It includes a user interface, a text processing module, a scene code generator, a layout generation module, an appearance generation module, a NeRF module, and a PeRF module. The user inputs text describing a room, which is processed into scene codes. These codes guide the generation of a 3D layout using oriented bounding boxes, ensuring spatial integrity. The appearance module then creates a visual representation with a panoramic image. The NeRF module constructs a base 3D model, which is refined by the PERF module for enhanced visual coherence.