摘要
由于机匣加工特征多、类型复杂,目前主要采用手动式、交互式等加工特征识别方法来实现机匣工艺设计过程中的特征提取,但是这些方法操作繁琐,智能化程度不高,导致工艺设计周期长。为了提升机匣加工特征识别效率,实现特征识别自动化,提出了一种基于图和规则加工特征自动识别方法。首先,根据机匣几何结构复杂程度,对机匣加工特征进行了归类。然后采用边界表示法表征机匣三维模型,定义了面属性、边属性、角度属性及面与面之间的拓扑关系码,提出了基于加权属性邻接矩阵的机匣三维模型的数据结构,通过对加权属性邻接矩阵的遍历和行列运算,建立了机匣加工特征的识别和抑制规则,构建了特征识别和抑制算法,并与预定义规则库进行匹配。最后在MATLAB平台上搭建了特征识别的仿真环境,选择了三个典型机匣案例,测试了机匣加工特征识别效果。结果表明,该特征识别方法具备较高的识别精度和效率,识别62阶矩阵仅用时0.171s。
Due to the manifold processing features and complex types of casing,manual and interactive processing feature recognition methods were mainly used to realize feature extraction in the technological design process of casing.However,these methods were cumbersome to operate and not highly intelligent,leading to a long process design cycle.In order to improve the efficiency of feature recognition in casing machining and realize feature recognition automation,an automatic feature recognition method based on graph and rule was proposed.First,according to the complexity of the geometric structure of the casing,the machining features of the casing were classified.Then using the boundary representation characterization for the 3d model,defining the attribute of surface,edge and angle,as well as the topological relationship between each surface,the 3d model data structure based on the weighted attribute adjacency matrix was proposed,establishing the recognition and suppression rules for the processing features of the casing by the traversal row and column operations of the weighted attribute adjacency matrix,feature recognition and suppression algorithm was constructed and then match with predefined rules library.Finally,a simulation environment for feature recognition was built on the MATLAB platform.Three typical casing cases were selected to test the processing feature recognition effect of the casing.The results show that the feature recognition method has high recognition accuracy and efficiency,and it only takes 0.171 seconds to recognize the 62-order matrix.
作者
郭亮
杨滔
李湉
周明
GUO Liang;YANG Tao;LI Tian;ZHOU Ming(School of Mechanical Engineering,Southwest Petroleum University,Sichuan Chengdu 610500,China;Aecc Chengdu Engine Co.,Ltd.,Sichuan Chengdu 610065,China)
出处
《机械设计与制造》
北大核心
2023年第6期212-218,共7页
Machinery Design & Manufacture
基金
国家自然科学基金(51705438)
四川省科技计划项目(2018JY0366)
中国航发自主创新专项基金项目(ZZCX-2017-039)
西南石油大学青年科技创新团队(2019CXTD02)。
关键词
机匣
特征识别
图和规则
加权属性邻接矩阵
Casing
Feature Recognition
Diagrams and Rules
Weighted Attribute Adjacency Matrix
