Abstract: To meet the classification and coding recognition requirements of corner box parts process knowledge reuse,a part classification and coding recognition method based on facet feature matching was proposed.The shape and machining characteristics of the whole part were analyzed and defined,and the principle of classification and coding based on the machining characteristics was established.The 0-1 feature matrix of the part surface depth was obtained by using the depth difference of the cavity wall edge surface set,which was decomposed and extracted to determine the basic type of the part.The geometrical topological matrix of the surface set and the edge-ring relation matrix of the characteristic surface were designed.To reduce the matching range of typical features and improve the recognition efficiency,the typical features were selected according to the basic types of parts and the attributes of face sets.Some examples were given to illustrate the extraction of typical feature base and its mapping matching with coding type,which verified the effectiveness and feasibility of the classification coding recognition algorithm.

Key words: corner box parts, typical matching, classification and coding, feature recognition, knowledge reuse

摘要： 为满足角盒类零件工艺知识重用的分类编码识别需求,提出一种分面集特征匹配的零件分类编码识别方法。对整体零件形状和加工特征进行了分析定义,建立了基于加工特征的分类编码原则;利用型腔壁边面集深度差异,获取零件面深度0-1特征矩阵,分解提取壁边面集并判断零件基本类型;设计了面集几何拓扑矩阵和特征面边环关系矩阵,根据零件基本类型与面集属性选取典型特征矩阵进行匹配,缩小典型特征匹配范围以提高识别效率。结合实例说明了典型特征库的提取及其与编码类型的映射匹配,验证了分类编码识别算法的有效性与可行性。

关键词: 角盒类零件, 典型特征匹配, 分类编码, 特征识别, 知识重用