Search

CN-121997500-A - Intelligent die-casting manufacturability analysis method and device

CN121997500ACN 121997500 ACN121997500 ACN 121997500ACN-121997500-A

Abstract

The invention provides an intelligent die-casting manufacturability analysis method and device. The method converts the casting model into a triangular mesh and calculates structural features. And solving a candidate set of the drawing mode direction through a triangular grid. For each drawing direction, a mold-openable region, an under-mold-openable region and a mold-unclampable region are determined. And determining a candidate region of each slide block to be arranged and a loose core direction based on the under-die-opening region and the non-die-opening region. And constructing various manufacturability indexes of the added sliders based on the content, analyzing various manufacturability index results of core pulling of each slider by utilizing a scoring model, generating an analysis report and outputting the analysis report. The method comprehensively considers the material properties, geometric structure characteristics, drawing scheme and specific die casting forming process conditions of the casting, takes the drawing scheme design as a core, and realizes evaluation and optimization of the drawing scheme. Thereby greatly improving the degree of automation of the die-casting product drawing scheme design flow, obviously reducing the scheme design time consumption and effectively reducing the operation difficulty.

Inventors

  • MA TONGQING
  • LI ZHONGLIN
  • LIU FENGYUAN

Assignees

  • 深圳适创腾扬科技有限公司

Dates

Publication Date
20260508
Application Date
20260409

Claims (10)

  1. 1. An intelligent die casting manufacturability analysis method, said method comprising: converting the parameterized geometric surface of the casting model into corresponding uniform triangular grids, and calculating structural characteristics of each parameterized geometric surface of the casting model based on the uniform triangular grids corresponding to each parameterized geometric surface; Calculating an inertia principal axis of the casting model and a principal axis of a bounding box of the casting model according to uniform triangular grids corresponding to each parameterized geometric surface, and carrying out rotary transformation according to the inertia principal axis and the principal axis of the bounding box to construct a candidate set of a drawing direction; For any drawing direction in the candidate set, defining the drawing direction as a movable die opening direction, defining the opposite direction of the drawing direction as a static die opening direction, and carrying out ray detection based on the movable die opening direction and the static die opening direction to obtain a die-openable surface piece set; Determining a die-openable area and an underdie-openable area according to any one of the die-openable surface patches and the die-openable direction in the die-openable surface patch set, wherein the die-openable direction is the moving die-openable direction or the static die-openable direction; determining a candidate region of a slide block to be arranged and a candidate direction set of the slide block according to the underdie opening region and the non-die opening region, wherein the non-die opening region is a region which cannot be contacted by the moving die opening direction and the static die opening direction; Selecting a target sliding block candidate direction from the sliding block candidate direction set, and taking the target sliding block candidate direction as a core pulling direction of a sliding block corresponding to a candidate region of the sliding block to be arranged, wherein the target sliding block candidate direction is a sliding block candidate direction which enables the total area of the underdie region or the non-die region to be minimum and the minimum die drawing angle value to be maximum; Constructing a feature vector of the drawing direction based on material properties of the casting model, structural features of each parameterized geometric surface, the mold-openable region, the undermold-openable region, the candidate regions of the sliders to be arranged, and core pulling directions of the sliders corresponding to the candidate regions of the sliders to be arranged; And respectively inputting the feature vectors of each pattern drawing direction into a scoring model to obtain a plurality of pattern drawing manufacturability index results of each pattern drawing direction, generating an analysis report based on the plurality of pattern drawing manufacturability index results, and outputting the analysis report.
  2. 2. The method of claim 1, wherein the calculating structural features of each parameterized geometric surface of the casting model based on the corresponding uniform triangular mesh of each parameterized geometric surface comprises: Acquiring point cloud data and topology connection information in a uniform triangular grid corresponding to each parameterized geometric surface; and inputting all the point cloud data and the topological connection information into the identification model, and outputting the structural characteristics of each parameterized geometric surface.
  3. 3. The method of claim 1, wherein performing a rotational transformation based on the principal axes of inertia and the bounding box to construct a candidate set of draft directions comprises: uniformly rotating and transforming around the principal axes of the inertial principal axis and the bounding box according to a preset angle to obtain each axial direction; And constructing a candidate set of the drawing direction according to the principal axes of the inertia and the bounding box and each axial direction.
  4. 4. The method of claim 1, wherein the performing radiation detection based on the moving die opening direction and the static die opening direction to obtain a set of die-openable patches comprises: Transmitting parallel rays to the inside of the casting model along the moving die opening direction and the static die opening direction; Calculating the intersection points between each parallel ray and the triangular meshes of each parameterized geometric surface, and marking the visible side of the triangular patch in the triangular mesh where each intersection point is positioned as a die-openable patch; and determining a set of the die-openable patches corresponding to the die-extracting direction based on all the die-openable patches.
  5. 5. The method of claim 4, wherein said determining a die-openable region and an under-die region from any of the set of die-openable panels and a die-opening direction comprises: Calculating a drawing angle according to the normal directions and the opening directions of three vertexes of the openable surface piece aiming at any openable surface piece in the openable surface piece set; When the draft angle is larger than a preset threshold value, marking the die-openable surface patch where the intersection point is located as a die-openable area; And when the draft angle is not greater than a preset threshold value, marking the die-openable panel where the intersection point is positioned as an underdie area.
  6. 6. The method of claim 1, wherein the determining a candidate region and a set of slider candidate directions for a slider to be arranged from the under-molded region and the non-molded region comprises: extracting a communication region of the under-mold-opening region and the non-mold-opening region based on the adjacent relation between the under-mold-opening region and the non-mold-opening region on the triangular mesh; marking each communication area as a candidate area of the slide block to be arranged; And aiming at any candidate area of the slide block to be arranged, taking the normal directions of all triangular patches contained in the candidate area of the slide block to be arranged as slide block candidate directions to obtain a slide block candidate direction set.
  7. 7. The method of claim 1, wherein the training process of the scoring model comprises: acquiring historical casting sample data, wherein the historical casting sample data comprises sample feature vectors; initializing a deep learning model to obtain an initial scoring model; inputting the sample feature vector into the initial scoring model to obtain manufacturability scores, draft angles and risk grades; Calculating a manufacturability score, an actual draft angle and an actual risk level, wherein the manufacturability score, the draft angle and the risk level correspond to the sample feature vector; If the loss function converges, marking the initial scoring model as a scoring model; And if the loss function is not converged, modifying parameters of the initial scoring model, and returning to the step of inputting the sample feature vector into the initial scoring model to obtain manufacturability scores, draft angles and risk grades.
  8. 8. An intelligent die casting manufacturability analysis device, said device comprising: The conversion model representation unit is used for converting the parameterized geometric surface of the casting model into corresponding uniform triangular grids and calculating the structural characteristics of each parameterized geometric surface of the casting model based on the uniform triangular grids corresponding to each parameterized geometric surface; The first construction unit is used for calculating an inertia main shaft of the casting model and a main shaft of a bounding box of the casting model according to uniform triangular grids corresponding to each parameterized geometric surface, and carrying out rotary transformation according to the inertia main shaft and the main shaft of the bounding box so as to construct a candidate set of the drawing direction; The definition unit is used for defining the drawing direction as a movable die opening direction and defining the opposite direction of the drawing direction as a static die opening direction aiming at any drawing direction in the candidate set, and carrying out ray detection based on the movable die opening direction and the static die opening direction to obtain a die-openable surface piece set; The first determining unit is used for determining a die-openable area and an underdie-openable area according to any one of the die-openable surface pieces and the die-openable direction in the die-openable surface piece set, wherein the die-openable direction is the moving die-openable direction or the static die-openable direction; The second determining unit is used for determining a candidate region of the sliding block to be arranged and a candidate direction set of the sliding block according to the under-die-opening region and the non-die-opening region, wherein the non-die-opening region is a region which cannot be contacted by the moving die in the die-opening direction and the static die in the die-opening direction; The device comprises a selection unit, a selection unit and a selection unit, wherein the selection unit is used for selecting a target sliding block candidate direction from the sliding block candidate direction set, and taking the target sliding block candidate direction as a core pulling direction of a sliding block corresponding to a candidate region of the sliding block to be arranged; The second construction unit is used for constructing a feature vector of the drawing direction based on the material property of the casting model, the structural features of each parameterized geometric surface, the mold-openable region, the mold-unclamping incapable region, the candidate regions of the sliding blocks to be arranged and the core pulling directions of the sliding blocks corresponding to the candidate regions of the sliding blocks to be arranged; And the scoring unit is used for respectively inputting the feature vectors of each drawing direction into a scoring model to obtain a plurality of drawing manufacturability index results of each drawing direction, and generating and outputting an analysis report based on the plurality of drawing manufacturability index results.
  9. 9. The apparatus of claim 8, wherein the transformation model representation unit comprises: the acquisition module is used for acquiring point cloud data and topology connection information in the uniform triangular grids corresponding to the parameterized geometric surfaces; and the identification module is used for inputting all the point cloud data and the topological connection information into the identification model and outputting the structural characteristics of each parameterized geometric surface.
  10. 10. The apparatus of claim 8, wherein the first building element comprises: The deflection module is used for uniformly rotating and transforming around the principal axes of the inertial principal axis and the bounding box according to a preset angle to obtain each axial direction; And the construction module is used for constructing a candidate set of the drawing direction according to the principal axes of the inertia and the bounding box and each axial direction.

Description

Intelligent die-casting manufacturability analysis method and device Technical Field The invention relates to the technical field of casting engineering, in particular to an intelligent die casting manufacturability analysis method and device. Background With the development of modern manufacturing, software tools such as computer aided design and computer aided manufacturing play an important role in the casting process. Casting processes are widely used in the manufacture of metal and plastic articles, but face challenges in improving production efficiency and product quality. The casting process involves pouring a liquid material into a mold and extracting the molded product after it has solidified, wherein the extraction is a critical step that directly affects product quality and mold life. At present, the drawing feasibility assessment mainly depends on analyzing the angle between the normal direction of the casting surface and the drawing direction, but the prior art has some defects. First, selection of the draft direction depends on engineer experience, and existing tools provide limited choices, lack of intelligent evaluation mechanisms based on geometric features and process constraints, may result in omission of the optimal draft direction, and thus increase manufacturing risk. Secondly, the existing tool can only rely on manual identification of a basic back-off region when identifying a slide block core pulling region, so that design efficiency is reduced. Finally, the drawing analysis is mainly focused on geometric characteristics, and material characteristics and process rules are not fully considered, so that reasonable drawing feasibility scoring is difficult to perform, and the optimization of the structure is affected. In summary, it is imperative to develop an integrated and intelligent analysis tool for manufacturability of casting to improve the practical application effect of the casting process. Disclosure of Invention In view of the above, the embodiments of the present invention provide an intelligent analysis method and device for manufacturability of die casting to solve the problem of difficult and incomplete analysis of manufacturability of die casting. In order to achieve the above object, the embodiment of the present invention provides the following technical solutions: The first aspect of the invention discloses an intelligent die casting manufacturability analysis method, which comprises the following steps: converting the parameterized geometric surface of the casting model into corresponding uniform triangular grids, and calculating structural characteristics of each parameterized geometric surface of the casting model based on the uniform triangular grids corresponding to each parameterized geometric surface; Calculating an inertia principal axis of the casting model and a principal axis of a bounding box of the casting model according to uniform triangular grids corresponding to each parameterized geometric surface, and carrying out rotary transformation according to the inertia principal axis and the principal axis of the bounding box to construct a candidate set of a drawing direction; For any drawing direction in the candidate set, defining the drawing direction as a movable die opening direction, defining the opposite direction of the drawing direction as a static die opening direction, and carrying out ray detection based on the movable die opening direction and the static die opening direction to obtain a die-openable surface piece set; Determining a die-openable area and an underdie-openable area according to any one of the die-openable surface patches and the die-openable direction in the die-openable surface patch set, wherein the die-openable direction is the moving die-openable direction or the static die-openable direction; determining a candidate region of a slide block to be arranged and a candidate direction set of the slide block according to the underdie opening region and the non-die opening region, wherein the non-die opening region is a region which cannot be contacted by the moving die opening direction and the static die opening direction; Selecting a target sliding block candidate direction from the sliding block candidate direction set, and taking the target sliding block candidate direction as a core pulling direction of a sliding block corresponding to a candidate region of the sliding block to be arranged, wherein the target sliding block candidate direction is a sliding block candidate direction which enables the total area of the underdie region or the non-die region to be minimum and the minimum die drawing angle value to be maximum; Constructing a feature vector of the drawing direction based on material properties of the casting model, structural features of each parameterized geometric surface, the mold-openable region, the undermold-openable region, the candidate regions of the sliders to be arranged, and core pulling directions of th