中国机械工程 ›› 2025, Vol. 36 ›› Issue (05): 1035-1043.DOI: 10.3969/j.issn.1004-132X.2025.05.015

• 智能制造 • 上一篇    下一篇

基于模型定义的三维模型信息完备性检查技术研究

刘泉泉1;方喜峰1*;程德俊1;张胜文1;罗兰珍2;孔俊龙1   

  1. 1.江苏科技大学机械工程学院,镇江,212100
    2.镇江康飞汽车制造股份有限公司,镇江,212132

  • 出版日期:2025-05-25 发布日期:2025-06-26
  • 作者简介:刘泉泉,男,2000年生,硕士研究生。研究方向为数字化设计与制造、智能制造。E-mail:3294632957@qq.com。
  • 基金资助:
    :国家自然科学基金(52305060);国防基础科研项目(A0720133010);江苏科技大学2023年度研究生教育教学改革研究课题(YJG2023Z_01)

Research on Information Completeness Checking Technique of 3D Model for MBD

LIU Quanquan1;FANG Xifeng1*;CHENG Dejun1;ZHANG Shengwen1;LUO Lanzhen2;KONG Junlong1   

  1. 1.School of Mechanical Engineering,Jiangsu University of Science and Technology,
    Zhenjiang,Jiangsu,212100
    2.Zhenjiang Kangfei Automobile Manufacturing Co.,Ltd.,Zhenjiang,Jiangsu,212132

  • Online:2025-05-25 Published:2025-06-26

摘要: 针对三维模型信息标注不规范、结构差异难诊断、尺寸冗余缺失以及人工检测效率低等问题,提出了整体的完备性检查方法。通过调用三维计算机辅助设计(CAD)软件应用程序接口(API)函数,对三维CAD软件的检查功能模块进行二次开发,完成对基于模型定义(MBD)技术中三维模型信息的检查与纠正;通过二次开发CAD软件实现对MBD模型基本方向视图的自动捕获,再通过Python调用OpenCV库,将新捕获的视图与数据库中已有模型视图利用图像结构相似性(SSIM)指标进行对比,得到最相似模型视图图像,将两者通过图像减法运算对结构差异区域进行提取并高亮显示;最后对三维模型尺寸按照分类规则分类,通过深度优先搜索(DFS)算法结合尺寸冗余缺失检查规则,完成尺寸的完备性检查。以某厢舱汽车产品为对象验证了所提方法的可行性。

关键词: 基于模型定义, 完备性检查, 二次开发, 结构相似性指标, 深度优先搜索算法

Abstract: Aiming at the problems of non-standard information labeling, difficult diagnosis of structural differences, lack of dimensional redundancy and low efficiency of manual detection for 3D model information, a holistic completeness check method was proposed. By calling the application programming interface(API)functions of the 3D computer-aided design(CAD) software, secondary development of the inspection modules was carried out to complete the inspection and correction of 3D model information for MBD technology. Automatic capture of the basic direction view of the MBD model was realized through secondary development of CAD software, and then OpenCV library was called through Python to compare the newly captured view with the existing model view in the database by using the image SSIM, and the most similar model view image was obtained. The structural difference areas between two images were extracted and highlighted through image subtraction operation. Finally, the sizes of the 3D model were classified according to the classification rules, and the size completeness check was completed by DFS algorithm combined with the size redundancy missing check rule. The feasibility of the proposed method was verified by taking a cabin car product as an object. 

Key words: model-based definition(MBD), completeness check, secondary development, structural similarity index metric(SSIM), depth-first search(DFS) algorithm

中图分类号: