33 results about "Three dimensional geometry" patented technology

This invention discloses a multi-objective topology optimization method for aircraft design, belonging to the field of aerospace technology. The method includes: establishing a three-dimensional geometric model and defining the optimization region and aerodynamic preservation region; describing the structural boundary using level set functions; applying volume and manufacturing constraints with the objectives of maximizing stiffness and natural frequency; constructing a comprehensive objective function that minimizes the distance between ideal points using a compromise programming method; transforming the constraint problem using the augmented Lagrangian method; calculating shape sensitivity and deriving the normal velocity field; updating the level set functions to drive topology evolution; iterating until convergence to generate a Pareto front solution set; selecting the optimal solution and reconstructing it into a manufacturable three-dimensional model. This invention solves the multi-objective collaborative optimization problem while preserving the aerodynamic shape, resulting in clear boundaries, strong manufacturability, and significantly improved structural performance and engineering applicability.

This invention discloses an additive manufacturing path planning method and system for the head of an ultra-high strength steel deep blind hole shell component. The method includes: acquiring a three-dimensional geometric model of the component to be formed; identifying and extracting the internal cavity surface and edge contour features of the head of the deep blind hole shell component; generating a variable thickness layered slicing scheme based on the internal cavity surface and edge contour features; adopting a multi-level partitioning strategy within each layer based on the variable thickness layered slicing scheme, and designing the main path direction and scanning angle for each region; generating an adaptive spiral progressive filling path that can suppress residual stress and deformation based on the main path direction and scanning angle; and generating a processing instruction file that can directly drive additive manufacturing equipment based on the adaptive spiral progressive filling path. Using this invention, active control of the heat distribution and shrinkage deformation within the internal cavity of the head of the ultra-high strength steel deep blind hole shell can be achieved, improving the forming accuracy, internal quality, and mechanical property consistency of the component.

This invention discloses a method, system, equipment, and medium for calculating pressure loss in perfluorohexanone (PFH) pipeline networks, relating to the field of computer-aided design technology. The method traverses the component tree of the pipeline network building information model using a three-dimensional geometry engine to form a set of pipeline network features; constructs a directed graph topology representing the flow direction of the pipeline network, calculates potential energy loss, converts local resistance into equivalent length, and constructs a weighted directional topology weight matrix; obtains hydraulic balance under single-phase flow conditions through iterative calculation of node simultaneous equations, and generates the corresponding static pressure distribution results of the pipeline network; recursively calculates the dryness and density changes of the gas-liquid mixture using the equation of state, introduces a two-phase flow friction multiplication factor for nonlinear compensation, outputs a three-dimensional dynamic compensation pressure field model, and performs rendering and digital judgment; this solution solves the problem of low accuracy in pressure loss calculation caused by the inability of existing technologies to accurately simulate the flash evaporation effect of PPHH, thus improving the reliability of pipeline network design.

The application discloses a part size precision analysis method and device, electronic equipment and storage medium, comprising: constructing a three-dimensional geometric model of a part to be assembled; determining a part reference and a matching feature based on the three-dimensional geometric model of the part to be assembled; obtaining an assembly process, and determining an assembly sequence and an assembly positioning mode based on at least one of the assembly process, the part reference and the matching feature; performing assembly simulation on the three-dimensional geometric model of the part to be assembled according to the assembly sequence and the assembly positioning mode, to obtain a virtual assembly model of a product; performing size chain analysis on the virtual assembly model, to determine a key size and a tolerance accumulation condition; performing size precision analysis according to the key size and the tolerance accumulation condition, to obtain a size precision evaluation result of the part to be assembled; and performing size optimization design on the part to be assembled according to the size precision evaluation result. The application can effectively improve the accuracy of part size precision analysis, and effectively improve the assembly efficiency through size optimization.

This invention discloses a method and system for optimizing the design of photovoltaic support structures, relating to the field of new energy technology. The method includes: constructing a parametric three-dimensional geometric model; generating multi-condition load combinations considering unsteady wind fields, asymmetric snow accumulation, and ground motion; performing automated finite element simulation; conducting multi-constraint specification verification based on national standards and user-defined thresholds; implementing multi-objective collaborative optimization with steel consumption, manufacturing cost, and construction convenience as objectives using an improved algorithm; and establishing a closed-loop feedback verification mechanism for automatic iteration until convergence. This invention aims to solve the problems of fragmented design processes, disconnect between optimization and verification, and difficulty in balancing safety and economy under multiple conditions in existing technologies. It achieves integrated closed-loop design throughout the entire process, significantly improving structural safety, material economy, and adaptability to complex terrain.

The present application relates to the technical field of process parameter optimization, and provides a forging process parameter optimization method and system based on a deep learning algorithm, which comprises the following steps: obtaining a blank point cloud and other process data, generating a three-dimensional geometric description of the blank, dividing a stress correlation area and a free deformation area, and matching the physical properties of different areas of forging. A deep learning network embedded with forging physical rules is constructed, plastic flow direction and surface heat exchange constraints are introduced in the thermal interaction implicit layer for the two areas, respectively, to avoid physical violation abnormal output that is prone to occur in pure data fitting from the bottom layer. A composite loss composed of parameter reconstruction deviation and physical compliance loss is used for model training, and the final obtained optimization model can output optimal process parameters, which not only reuses historical process experience, but also conforms to the forging physical law, effectively improves the forging forming precision, reduces forming defects, and greatly shortens the process optimization cycle.

A Gaussian space-guided interactive two-handed 3D reconstruction method, belonging to the field of computer vision and 3D geometric modeling technology, is proposed. The method preprocesses the input monocular RGB image and extracts multi-scale 2D visual features, achieving early separation of left and right hand features at the feature layer. A two-handed feature enhancement module is used to structurally enhance and model the relationships between the left and right hand features. A 3D Gaussian geometric field, independently pre-trained from multi-view data, is introduced, and 2D Gaussian spatial features aligned with the visual features are generated through differentiable rendering, serving as a stable spatial geometric prior. The 2D visual features and Gaussian spatial features are fused across modally to obtain a two-handed feature representation containing spatial constraint information. Finally, based on a parametric hand model, the pose and shape parameters of the left and right hands are regressed to generate the corresponding 3D mesh model of both hands. This invention effectively compensates for the lack of spatial information under monocular RGB conditions, significantly improving the structural integrity, spatial consistency, and robustness of the 3D reconstruction results in complex occlusion and two-handed interactive scenarios, demonstrating good versatility and practical application value.

The application provides a model digital generation type design method and device based on a continuous implicit function, and relates to the technical field of artificial intelligence, which comprises the following steps: performing copyright and trademark similarity detection processing on design requirement information to obtain target design requirement information, and sending the target design requirement information to a conditional diffusion model to generate a two-dimensional concept map of a to-be-rendered jewelry; performing implicit three-dimensional reconstruction processing on the two-dimensional concept map to obtain a jewelry three-dimensional geometric model, and performing automatic matching processing on the jewelry three-dimensional geometric model to obtain a jewelry target three-dimensional geometric model containing physical material parameters; performing physical rendering processing and image enhancement processing on the jewelry target three-dimensional geometric model to obtain a multi-angle high-definition rendering map of the jewelry, and sending the jewelry target three-dimensional geometric model and the multi-angle high-definition rendering map as target output to a user end. The application can significantly improve the accuracy and model construction efficiency of jewelry three-dimensional model construction.

This invention belongs to the field of digital deformation technology for ship and marine engineering structures, specifically relating to a method, program, device, and storage medium for body-fitting free deformation of ship and marine structures based on a general convex hexahedron. This invention improves upon traditional free deformation methods by replacing the cuboid control frame with a general convex hexahedron control frame. It innovatively combines coplanarity / convexity checks with trilinear interpolation to achieve the construction and verification of the convex hexahedron control frame and the generation of control points. It combines intelligent optimization methods with Bernstein basis functions and proposes a two-step screening criterion to achieve efficient parametric expression and deformation of three-dimensional geometry. This invention can maintain accurate deformation of complex three-dimensional geometry even when the deformation control frame fits the bow or stern plate, areas with large curvature changes and high tilt angles, and can also reduce the space of design variables to achieve efficient numerical optimization design.

This invention provides an intelligent planning and optimization method and system for injection mold machining paths, belonging to the field of intelligent machining technology. The method includes acquiring a three-dimensional geometric model of the cavity to be machined and process constraints; extracting geometric topology information based on surface features and performing partitioned processing; constructing a multi-objective optimization model based on local curvature distribution characteristics and material removal constraints; dynamically adjusting the connection path generation strategy through adaptive iterative solution of the model to generate a tool trajectory planning scheme; converting the scheme into CNC machining instructions and executing them, while simultaneously collecting actual cutting force feedback information to correct model parameters, thus achieving intelligent machining path optimization. This invention improves machining efficiency and accuracy.

This invention belongs to the field of shield tunneling technology, specifically a method and system for calculating the three-dimensional steady-state seepage field at the excavation face of a shield tunnel under isolated rock conditions. It includes: S1 establishing a three-dimensional geometric model of the seepage field around the tunnel excavation face as a cuboid, with the vertical direction as the z-axis, the direction perpendicular to the tunnel excavation as the x-axis, and the direction along the tunnel excavation as the y-axis; S2, combining the boundary conditions of each region of the seepage field, establishing two-dimensional hydraulic head expressions for the six face regions of the three-dimensional geometric model based on the cuboid three-dimensional geometric model; S3 using the continuity conditions between regions to construct linear partial differential equations and Fourier series expansion equations, solving for the three-dimensional hydraulic head expressions for the six regions; and S4 obtaining the analytical solution for the hydraulic head at any position in the three-dimensional seepage field in front of the excavation face based on the superposition of Fourier series. This invention can accurately calculate the water pressure borne by the excavation face, which also has significant engineering significance for ensuring the safe construction of shield tunnels.

This invention relates to the field of spatial intelligence technology, providing a spatial intelligent hybrid reflection generation system that integrates explicit 3D geometric projection and implicit diffusion generation. The system includes: a spatial reflection data acquisition and preprocessing module; an explicit 3D geometric modeling and mirror parameter analysis module; a virtual viewpoint construction and geometric coarse rendering module; a depth-condition-based bi-branch diffusion generation module; a reflection result geometric consistency constraint module; a phased training and model stabilization module; and an inference deployment and cross-scene adaptation module. Each module achieves hierarchical transmission and bi-directional feedback through spatial geometric information flow and generation constraint information flow, thereby constituting a spatial intelligent hybrid reflection generation system that combines generation capabilities with a spatial consistency guarantee mechanism. This invention effectively overcomes the shortcomings of existing generative reflection methods at the spatial intelligence level. This invention also provides a spatial intelligent hybrid reflection generation method that integrates explicit 3D geometric projection and implicit diffusion generation.

This application relates to a parametric generation method and system for the three-dimensional spatial form of urban buildings, belonging to the field of urban planning technology. The method includes: inputting a target spatial range and building layout parameters; generating building layout areas using a polygon segmentation algorithm; obtaining the two-dimensional base surface of the building to be generated using a maximum inscribed rectangle and orientation algorithm; obtaining building spacing buffers using a buffer algorithm; S5, obtaining the available remaining spatial range using a nine-intersection model algorithm; S6, repeatedly iterating the remaining spatial range to generate a set of all two-dimensional base surfaces of the buildings; performing three-dimensional extrusion on each surface of the set to generate a primary building model at LOD1 level; generating the three-dimensional geometry of the roof of each primary building model to obtain a building model at LOD2 level; applying textures and materials to the building model to obtain a building model at LOD3 level; and exporting the building model into a general three-dimensional model format. This application, through parameter-driven and intelligent iteration, can quickly generate three-dimensional building layouts and models that meet the requirements of the latest planning regulations.

The application discloses a non-rotation type integral inertial particle separator design method, which comprises the following steps: S1, establishing an XY coordinate system and determining a plurality of key nodes on a meridian plane flow passage profile and coordinates of the key nodes; S2, setting a slope value for each key node; S3, calculating a two-degree-of-freedom adjustable curve between two key nodes which need to be connected into a line; S4, calculating the two-degree-of-freedom adjustable curve between all the key nodes which need to be connected, and sequentially connecting to form a meridian plane flow passage profile of the non-rotation type integral inertial particle separator; and S5, rotating the meridian plane flow passage profile by 360 degrees around the x-axis of the coordinate system to generate a solid of revolution, and obtaining a three-dimensional geometric configuration of the non-rotation type integral inertial particle separator. i,1 i,2 The application adopts a two-degree-of-freedom adjustable curve function, and the shape of the curve between any adjacent key nodes can be freely changed by changing free parameters O i,1 i,2 , O i,2 , so that the design and research of the parameterization of the non-rotation type integral inertial particle separator are more flexible and rapid.

The application discloses a structured 3D scene and dynamic trajectory generation method for enhancing spatial understanding of a multimodal large language model, and the method comprises the following steps: step 1, in response to high-level scene theme input, a blueprint conforming to human cognition and real logic is constructed for a scene to be generated; step 2, based on an anchor point priority placement strategy and a strict multi-level real-time verification mechanism, an abstract blueprint is instantiated into a three-dimensional geometric layout; step 3, a semantic interest point list is used to generate a camera trajectory; and step 4, rich virtual world information is converted into a highly structured and standardized data set that can be directly used by a machine learning model. The application significantly improves the semantic reality and layout rationality of a generated scene, realizes deep coupling of scene layout and agent trajectory, greatly improves the intrinsic value of training data, and ensures the effectiveness and solvability of all generated scenes for downstream navigation tasks.

The present application relates to a kind of based on limiting hierarchical voxel structure numerical control machining three-dimensional geometry simulation method, belong to numerical control machining simulation technical field, solve the problem of reduced efficiency in the prior art in finding voxel, simulation memory consumption significantly rises and simulation efficiency reduces, including S1: voxel hierarchical structure initialization, based on the three-dimensional geometry data of the blank model to be processed obtained, establish primary voxel, use as the root node of voxel hierarchical structure;S2: determine final stage voxel size, with the voxel between primary voxel and final stage voxel as intermediate level voxel, determine the subdivision granularity of each hierarchical voxel;S3: based on the intersection detection of voxel of tool motion, determine the intermediate level voxel acted by tool at current simulation time;S4: final stage voxel dynamic generation, according to the spatial size of final stage voxel, generate final stage voxel for geometry simulation calculation;S5: based on the geometry simulation calculation of final stage voxel, obtain the three-dimensional geometry simulation result of numerical control machining process.

This specification provides a weld extraction method based on three-dimensional geometric feature analysis. The method includes: acquiring a target three-dimensional model; obtaining a set of contour lines based on the geometric Boolean intersection of each independent entity in the target three-dimensional model; performing a screening process on the contour line set to obtain a weld set, the screening process including a first screening, a segmentation process, and a second screening; determining multiple vertical welds and multiple horizontal welds based on the coordinate differences of each weld in the weld set in a preset direction; determining the horizontal welding parameters of the multiple horizontal welds; and determining the vertical welding parameters based on the horizontal welding parameters, the multiple vertical welds, and the multiple horizontal welds' topological association.

This invention discloses a 3D point cloud upsampling method based on a geometric sequencer paradigm. The method includes the following steps: First, deterministic interpolation is performed on the input sparse point cloud to generate an initial dense point cloud, and the farthest point sampling and k-nearest neighbor algorithm are used to decompose it into a series of local "point blocks". Second, GeoHash spatial indexing technology is innovatively introduced to deterministically sort the disordered point blocks through a space-filling curve, reconstructing them into an ordered one-dimensional sequence that maintains the original 3D geometric proximity. Finally, this ordered sequence is input into a lightweight multilayer perceptron decoder, and through non-parametric feature encoding, including fixed sinusoidal position embedding and relative geometric weighting, a refined displacement vector is predicted, thereby recovering a high-fidelity, high-density point cloud. Applying this invention can achieve high-quality point cloud reconstruction with a small number of model parameters and low memory consumption, improving the deployment flexibility and computational efficiency of the model.

The three-dimensional model viewpoint planning method, device, equipment, storage medium and product disclosed in the present application include representing a three-dimensional geometric model as a two-dimensional Riemann manifold, calculating differential geometry invariants and constructing a multi-scale geometric feature descriptor; constructing a sampling density function based on the feature, adaptively generating seed points on the surface and extracting local geometric context information; generating candidate viewpoints in the normal direction of the seed points, determining the viewpoint pose through multi-objective optimization and constructing a frustum model to calculate the visibility region; constructing a coverage function according to the visible patch set and solving the viewpoint selection problem using a greedy algorithm, and generating the final viewpoint planning result in combination with a termination criterion, multi-scale verification and local refinement strategy. The present application overcomes the shortcomings of traditional methods, such as lack of geometric adaptivity and high computational complexity, and realizes high-coverage and high-efficiency viewpoint planning, which is suitable for application scenarios such as visual detection of complex three-dimensional targets, robot visual servoing, automatic quality control and the like.

This invention discloses a method for determining the force boundary of a tooling flag used in large wing assembly, comprising: Step 1, establishing a three-dimensional geometric model of the tooling flag in the large wing assembly tooling, and obtaining a simplified model of the tooling flag through simplification processing, and determining the center of mass of the wing model in the simplified model; Step 2, setting mechanical boundary conditions for the simplified model of the tooling flag; Step 3, establishing constraint equations based on the simplified model of the tooling flag and the center of mass of the wing model; Step 4, applying a concentrated load to the center of mass of the wing model in the simplified model of the tooling flag, and solving for the displacement field and stress field results of the simplified model of the tooling flag under the concentrated load, in order to determine whether the design stiffness of the simplified model of the tooling flag meets the requirements; Step 5, solving for the stress field results of the simplified model of the tooling flag under the concentrated load on the simplified model of the tooling flag obtained in Step 4 that meets the stiffness design requirements, and calculating the force boundary of each tooling flag in the simplified model of the tooling flag. The technical solution provided by the embodiments of the present invention solves the problem that the traditional design method for the stress boundary of the tooling flag does not take into account the actual spatial distribution of the tooling flag, resulting in a large difference from the actual situation. This leads to inaccurate stress on the designed tooling flag, unreasonable distribution position and quantity of the tooling flag, and consequently, excessive stress on local tooling flags, causing deformation of the entire wing tooling.

This invention discloses a method, system, and device for the collaborative design of wave energy buoys and power output devices. The method involves: inputting a subset of buoy geometric parameters from a set of design variables into a parameterized geometric model to generate a three-dimensional buoy geometry; inputting the three-dimensional buoy geometry and a subset of power output device parameters into a frequency domain hydrodynamic analysis model to calculate an annual power generation assessment value; calculating a cost assessment value based on the buoy geometry; inputting the annual power generation assessment value and the cost assessment value into a comprehensive objective function to calculate a comprehensive evaluation value; and generating a new generation of design variable sets based on the comprehensive evaluation value, iterating until convergence and outputting the optimal design variable set; and resolving the optimal buoy geometric parameter subset and the optimal power output device parameter subset from the optimal design variable set for collaborative design of wave energy buoys and power output devices. Therefore, this invention can efficiently find the globally optimal combination of buoy shape and power output device parameters in a high-dimensional design space.

In an aspect, a composite material system comprises: a structure having an architected three-dimensional geometry; wherein said three-dimensional geometry is monolithic and deterministic; and a matrix phase; wherein said matrix phase at least partially infiltrates said structure. In some embodiments, the three-dimensional geometry is a nano- or micro-architected three-dimensional geometry.

The application relates to a cooler fault detection method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring real-time temperature data of each region of the cooler under real-time working conditions to construct a real-time temperature distribution database, and acquiring real-time oil flow flow rate data to construct a real-time oil flow flow rate database; comparing the real-time temperature distribution database with a preset standard temperature distribution database to determine whether there is a temperature fault region, and comparing the real-time oil flow flow rate database with a preset standard oil flow flow rate database to determine whether there is an oil flow flow rate fault region; constructing a three-dimensional geometric model of the cooler; when the temperature fault region exists, performing digital twin display on the temperature fault region based on the three-dimensional geometric model of the cooler, and when the oil flow flow rate fault region exists, performing digital twin display on the oil flow flow rate fault region based on the three-dimensional geometric model of the cooler. The method can improve detection accuracy.

This disclosure provides a method for 3D reconstruction of a volleyball court and motion capture of athletes, applicable to the analysis of athlete-court interaction in volleyball scenarios. First, court reconstruction and camera estimation are performed: the 3D geometry of the court is reconstructed from the input volleyball video sequence, obtaining a 3D point cloud representation of the court and camera extrinsic parameter information. Then, athlete pose estimation is performed: through human body region detection and cropping in video frames, feature extraction, human keypoint localization and coordinate decoding, human pose parameter regression, and temporal consistency constraints, the 3D pose of the volleyball player is estimated. Finally, the volleyball court geometry and its global motion trajectory consistent with its spatial structure are output. By jointly modeling the volleyball court and athletes in 3D, a stable, continuous, and intuitively observable 3D motion process and scene structure can be obtained under a single video input condition, facilitating the analysis of volleyball techniques and their spatial relationships.

This application relates to the field of building construction technology, specifically a BIM-based construction method for high-slope irregular-shaped reinforced concrete structures. The method involves: surveying the construction site and collecting basic data; establishing a three-dimensional geometric model using BIM software; simulating the mechanical properties of the three-dimensional model and optimizing the reinforcement layout; preparing for on-site construction based on the optimized plan; using a high-precision laser positioning system to accurately locate key nodes; constructing a modular adjustable formwork support system; performing reinforcement binding operations; pouring the irregular-shaped structure using self-compacting concrete; conducting quality inspection and surface treatment of the formed structure; and feeding the inspection results back to the BIM model to form a closed-loop optimization. This invention, through the combination of BIM technology and intelligent equipment, achieves precise control of the entire construction process for high-slope irregular-shaped structures, improving construction efficiency and project quality.

This invention relates to a method and system for planning grinding paths. The method includes: acquiring three-dimensional geometric shape data and surface physical property data of a target workpiece, and generating a physical feature fusion dataset through spatiotemporal registration; based on the physical feature fusion dataset, dividing the surface of the target workpiece to be ground into multiple processing sub-regions, and dividing at least one attitude-stabilized layer into each processing sub-region; using a preset path planning mechanism, generating an initial grinding path for each attitude-stabilized layer, wherein the initial grinding path consists of multiple path segments, forming an initial path segment set; based on each path segment in the initial path segment set, generating an optimal path segment execution sequence using a preset global temporal optimization mechanism; generating the final grinding operation instruction according to the optimal path segment execution sequence, and controlling the grinding equipment to execute it. This improves batch consistency, process stability, and overall efficiency of the processing quality.

This invention discloses a method for predicting the thickness of the solidified shell in an argon-blown continuous casting slab crystallizer, relating to the field of continuous casting production process simulation technology. The method includes: establishing a three-dimensional geometric model based on the geometric parameters of the continuous casting crystallizer and its submerged entry nozzle; extracting the computational domain from the three-dimensional geometric model based on boundary delimitation and Boolean operations, and then performing structured mesh generation to obtain a mesh model corresponding to the computational domain; importing the mesh model into simulation software, selecting the computational model and solution method, defining physical property parameters and boundary conditions, performing initialization operations, and then iteratively solving to obtain the distribution results of the solidified shell in the continuous casting crystallizer; post-processing the distribution results, determining the solidification front position based on a liquid fraction threshold, and obtaining the thickness distribution of the solidified shell in the continuous casting crystallizer at different heights and under different operating conditions. This invention has the advantages of accurately predicting the thickness of the solidified shell, improving reliability, and enhancing engineering applicability.

The present application relates to a kind of slot processing feature identification method and system.The present application includes input with slot feature parts three-dimensional geometric model;Extract the edge adjacency relationship of each face in the three-dimensional geometric model, construct adjacency graph, generate the attribute adjacency matrix of model based on the adjacency graph;Slot database is constructed;Based on the slot database, the face edge adjacency relationship of each slot feature template is extracted, and the corresponding pre-matching subgraph matrix is constructed;Based on the pre-matching subgraph matrix, from the attribute adjacency matrix of the model, the candidate slot feature that is consistent with each feature subgraph topological structure is screened out;The candidate slot feature is sequentially screened and classified according to the preset order;According to the preset face and edge geometric rule, the geometric constraint verification is carried out to candidate slot feature;Based on the final slot feature set, the slot processing feature in three-dimensional geometric model is output.The present application effectively solves the technical problem that the effect of conventional method is poor in common slot feature identification.

The application discloses a feature-driven point cloud completion method and system for a CAD parameterized model, relates to the technical field of three-dimensional geometry processing and computer vision, and comprises the following steps: constructing a multi-source feature line collection and standardization system, obtaining feature line candidates through multiple paths such as analysis of CAD information, mesh point cloud geometric clues and interactive editing rules, and outputting a unified format feature representation after consistency evaluation, cleaning and repairing; designing a local feature extraction method based on adaptive weights of multiple relationships, dynamically adjusting weights according to the relationship between target points and neighbor points in three-dimensional space and high-dimensional feature space in a DGCNN framework, and quantifying the contribution of each neighbor point to the local feature of the target point; and proposing a feature line deep information injection technology, fusing feature line geometric information into a point cloud completion deep learning model through an encoder, using a feature line weighted chamfer distance loss function to strengthen the fidelity of boundary and acute angle regions, and effectively improving the accuracy and geometric detail retention capability of point cloud completion.