Search

CN-122020899-A - Intelligent design method and system for machining clamp based on hopeful 3D

CN122020899ACN 122020899 ACN122020899 ACN 122020899ACN-122020899-A

Abstract

The invention discloses an intelligent design method and system for a machining clamp based on hopeful 3D. The method is characterized by constructing a dynamic time-varying voxel field and a reverse differential evolution mechanism, namely calculating interference potential energy of a tool time sequence envelope and a constraint unit in real time in multi-physical field simulation of virtual manufacture, establishing a jacobian matrix of a parameter space and a potential energy space, analyzing the steepest descent direction of an interference gradient, and driving a tool model to perform self-adaptive geometric deformation or pose transformation by utilizing a Newton-Lafson iteration method. The invention realizes the full-flow automation from geometric analysis, rule reasoning, physical field simulation to parameter closed-loop optimization, solves the problems of CAD/CAM data fault and open-loop iteration in the traditional tool design, and ensures the mathematical-level global convergence and processing safety of the process equipment.

Inventors

  • LIAN JUNFENG
  • LIU ZHIJIE
  • ZHANG JUNYI
  • WEI YONGKANG
  • LIU PENG
  • GUO CANHUI
  • GUO KAILE
  • LI XINYUN
  • CHEN JIAWEI
  • WANG YANPEI
  • WANG HONGYI

Assignees

  • 深圳技术大学

Dates

Publication Date
20260512
Application Date
20260131

Claims (8)

  1. 1. An intelligent design method of a machining fixture based on hopeful 3D, which is characterized by being operated on a computer-aided manufacturing terminal comprising a Graphic Processing Unit (GPU) and a Central Processing Unit (CPU), and comprising the following steps: S1, geometrical manifold semantic reconstruction, namely calling a three-dimensional geometrical kernel interface to read boundary representation (B-Rep) model data of an object to be processed, calculating average curvature and Gaussian curvature tensor of each node on the surface of the model by using a discrete differential geometrical algorithm, carrying out region segmentation on the surface based on a curvature continuity principle to construct a geometrical semantic graph containing surface normal vectors, area weights and roughness attributes, screening out characteristic manifold surfaces meeting the constraint of positioning degrees of freedom from the graph according to a six-point positioning principle, and establishing a characteristic adjacent matrix; S2, based on topology initial mapping of a fuzzy neural network, the feature adjacency matrix is used as an input vector and is input into a pre-trained fuzzy neural network reasoning model, the model is used for searching and instantiating corresponding constraint units from a heterogeneous parameterized primitive database based on a maximum membership principle, and the constraint units comprise a positioning element, a clamping element and an auxiliary supporting element; S3, constructing a dynamic Time-varying voxel field (Time-Varying Voxel Field), namely placing a workpiece model and an initial topological assembly set into a virtual multi-physical field simulation environment, analyzing a numerical control machining code generated by a CAM module, extracting a Time sequence track of a tool center point, discretizing a machining space into a three-dimensional voxel grid with resolution better than 0.01mm by adopting an Octree (Octree) space segmentation technology, and constructing a tool scanning envelope dynamically evolving along with Time according to the tool motion track; s4, based on the interference potential evolution of the Minkowski sum, defining a system interference potential function E (t), and executing Boolean intersection operation of a constraint unit entity and a cutter scanning envelope body on each discrete time slice; S5, performing inverse differential driving and multi-target global convergence, namely establishing a jacobian matrix J between a constraint unit driving parameter space P and interference potential energy E, performing inverse differential correction on geometric dimension parameters or pose parameters of the constraint unit along a negative gradient direction by utilizing a Newton-Lafson iteration method, and introducing a rigidity check function to ensure that the elastic deformation of the corrected tool system under the action of the maximum cutting force is smaller than an allowable tolerance until the system meets a global convergence condition; S6, reconstructing heterogeneous data and engineering output, namely locking a final converged topological state, reconstructing a mathematical model into an engineering assembly view through an API interface, and automatically generating an associated bill of materials (BOM) and a processing technology file.
  2. 2. The intelligent design method of the machining fixture based on the hopeful 3D, which is characterized in that in the step S1, the construction of the geometric semantic graph specifically comprises the steps of calculating the principal curvatures of each point on the surface of a model , When (1) Marked as planar features when The feature adjacency matrix describes the topological connection relation among the feature faces through a graph theory algorithm.
  3. 3. The method according to claim 1, wherein in the step S3, each voxel node of the dynamic time-varying voxel field carries a physical attribute tag, and the tag includes a workpiece occupied state, a fixture occupied state, a tool occupied state, and a free space state.
  4. 4. The intelligent design method of the machining fixture based on the hopeful 3D according to claim 1, wherein in the step S5, the inverse differential correction comprises a priority strategy of preferentially adjusting non-functional size parameters of the constraint unit, secondarily adjusting spatial position parameters of the constraint unit, and prohibiting adjustment of reference contact parameters related to positioning accuracy.
  5. 5. A hope 3D based intelligent design system for machining fixtures for implementing the method of any one of claims 1 to 4, comprising: (1) The geometric analysis module is used for executing feature manifold extraction and semantic map construction; (2) The reasoning mapping module is used for generating initial topology assembly based on fuzzy rules; (3) The multi-field simulation module is used for constructing a voxel field and calculating interference potential energy; (4) The iterative optimization controller is used for executing inverse differential correction and rigidity verification; (5) And the data interaction module is used for realizing man-machine interaction and engineering drawing output.
  6. 6. The intelligent design system of the machining fixture based on the hopeful 3D, which is characterized by realizing bidirectional real-time transmission of data streams by deep integration of an Application Program Interface (API) and a hopeful 3D (ZW 3D) geometric modeling kernel.
  7. 7. A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of claims 1 to 4.
  8. 8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 4 when executing the program.

Description

Intelligent design method and system for machining clamp based on hopeful 3D Technical Field The invention relates to the technical fields of intelligent manufacturing, computer Aided Design (CAD), computer Aided Manufacturing (CAM), calculation geometry and secondary development of industrial software, in particular to an intelligent design method and system of a machining fixture based on hopeful 3D. Background In the fields of aerospace, automobile engines, precision dies and the like, the geometric shapes of workpieces are increasingly complex (such as blisks, special-shaped boxes, free-form surface blades and the like), and extremely high requirements are put on the design of processing equipment. A Fixture clamp (Fixture) is used as a physical interface for connecting a machine tool and a workpiece, and has the main function of restraining six degrees of freedom (6-DOF) of the workpiece and ensuring positioning accuracy and clamping stability in the machining process. The defects of the prior art are that the current fixture design and manufacturing flow commonly has the following serious technical bottlenecks through searching, so that the process preparation period is long and the trial and error cost is high: (1) Shallow layer nature and dependence of geometric semantic understanding the existing CAD aided design tools are mostly based on simple feature recognition (Feature Recognition) technology, and only can recognize regular features such as holes, grooves, planes and the like. For complex freeform surfaces, software cannot understand the mapping relationship of "positioning constraints" and "geometric manifolds" from the mathematical topology level. Thus, the choice of positioning references is highly dependent on the personal experience of the process personnel. This dependence leads to process scheme quality variations, making it difficult to achieve standardized and optimal solutions. (2) Design and fabrication space-time splitting (Spatio-Temporal Decoupling) in a mainstream industrial software architecture, CAD (design environment) and CAM (manufacturing environment) are two relatively independent, statically decoupled modules. When the engineer designs the fixture in the CAD environment, the engineer cannot sense the Tool Path (Tool Path) planned by the CAM engineer later in real time, whereas when the CAM engineer is programmed, the CAM engineer faces a dead static fixture model. This Open Loop (Open-Loop) mode of operation results in multiple manual "design-import-collision-modification-reintroduction" iterations being necessary before actual machining. According to statistics, the tool design iteration of the complex part needs 3-5 rounds on average, and takes days or even weeks. (3) The lack of adaptation based on physical feedback-Collision Check (collisioncheck) in the prior art is typically "passively error-reporting" -software only highlights the interference area, but does not tell the engineer "how to modify it to be optimal". A mathematical algorithm mechanism is lacking, and CAD model parameters can be reversely driven to automatically correct according to the magnitude (volume) and the direction (gradient) of interference. (4) Fault and island effects of data flow, namely lack of unified data main line between design model, simulation model and manufacturing BOM. After the design is changed, the BOM is often required to be manually re-mapped and updated, and the material version of the production site is very easy to be confused. In summary, the industry urgently needs an intelligent method capable of realizing the full-flow closed loop of the tool layout 'feature perception-rule reasoning-field simulation-gradient evolution-data reconstruction' in a single platform such as a hopeful 3D (three-dimensional) platform by deeply fusing a geometric kernel and a physical field algorithm. Disclosure of Invention The invention mainly aims to overcome the defects of the prior art and provides an intelligent design method and system for a machining clamp based on hopeful 3D. According to the method, the tool design problem is converted into the space topology optimization problem under the multi-constraint condition, a computer algorithm is used for replacing manual trial and error, and mathematical-level global convergence of the tool layout is achieved. In order to achieve the above purpose, the invention adopts the following technical scheme: an intelligent design method of a machining fixture based on hopeful 3D (three-dimensional) is operated on a computer terminal comprising a Graphic Processing Unit (GPU), and comprises the following steps: S1, geometric manifold semantic reconstruction. The method comprises the steps of calling a three-dimensional geometric kernel interface to read boundary representation (B-Rep) model data of an object to be processed, calculating average curvature and Gaussian curvature tensor of each node on the surface of the model by using a discrete diff