|
Abstract: |
As representatives of high-precision parts, aerospace products are characterized by complex structures, long production cycles and high precision requirements. After the design of aerospace products is completed, a manufacturability assessment needs to be conducted based on 3D model features in terms of modeling quality and process design, otherwise the cost of design changes will be increased. Due to the poor structure and reusability of product manufacturing feature information and assessment knowledge in the current aerospace product manufacturability assessment process, it is difficult to realize automated manufacturability assessment. To address these issues, this paper first establishes a domain ontology model for aerospace product manufacturability assessment. On this basis, a structured representation method of manufacturability assessment knowledge and a knowledge graph data layer construction method are proposed. Based on the semantic information and association information expressed by the knowledge graph, a rule matching method based on subgraph matching is proposed to improve the precision and recall. Finally, applications and experiments based on the software platform verify the effectiveness of the proposed knowledge graph construction and rule matching method. |
Key words: knowledge graph aerospace product manufacturability assessment rule matching |
DOI:10.11916/j.issn1005-9113.23072 |
Clc Number:TP3 |
Fund: |
|
Descriptions in Chinese: |
航天产品作为高精度零件的代表,具有结构复杂、生产周期长、精度要求高等特点。航天产品在设计完成后,需要基于三维模型特征,从建模质量和工艺设计等方面进行可制造性审查,否则会增加设计变更成本。由于目前航天产品可制造性审查过程中产品制造特征信息和审查知识结构性差、可复用性差,导致难以实现可制造性自动化审查。针对该问题,本文首先建立了航天产品可制造性审查的领域本体模型。在此基础上,提出了可制造性审查知识的结构化表示方法与知识图谱数据层构建方法。基于知识图谱表达的语义信息与关联信息,提出了一种基于子图匹配的规则匹配方法,以提高审查规则匹配的准确率和召回率。最后基于软件平台的应用和实验验证了所提出的知识图谱构建和规则匹配方法的有效性。 |