基于语义分析的工艺知识库构建方法及应用  

A Semantic Analysis Based Method for Process Knowledge Base Construction and Application

作　　者：梁蕾 宋庆明 沈思远 吴敏 Liang Lei;Song Qingming;Shen Siyuan;Wu Min(COMAC Shanghai Aircraft Manufacturing Co.Ltd,Manufanturing Engineering Technology Center,Shanghai 201324)

机构地区：[1]上海飞机制造有限公司制造工程技术中心,上海201324

出　　处：《现代计算机》2022年第7期59-64,共6页Modern Computer

摘　　要：工艺知识库是实现智能工艺设计的核心组件。工艺知识的获取即工艺知识库构建一直是关键问题。现有工艺知识库的构建方法为人工构建,对构建人员的专业能力及信息技术能力要求极高,难以保证无冲突、概念理解一致、知识覆盖充分等要求。面向该问题,本文基于深度学习方法提出了基于语义分析的工艺知识库构建方法,能够自动地从历史工艺数据中归纳出工艺知识,降低了人工投入,能够保证工艺知识库的覆盖度,并且知识与历史应用情境关联,实现在相同情境匹配下知识的推送重用。在某型号支线客机航电与电气专业上构建了工艺知识库并实现了工艺知识推送软件。Process knowledge base is the core component of intelligent process design.The acquisition of process knowledge,that is,the construction of process knowledge base,has always been the key issue.The existing construction method of process knowledge base is manual method,which requires high professional ability and information technology ability,and it is difficult to guarantee the requirements of no-conflict,consistent of concept understanding and sufficient of knowledge coverage.For this prob⁃lem,based on the deep learning method,this paper proposes a process knowledge base construction method based on semantic analysis,which can automatically summarize the process knowledge from the historical process data,reduce the human labor,en⁃sure the coverage of the process knowledge base,and associate the knowledge with the historical application situation,so as to real⁃ize the recommendation and reuse of knowledge under the same situation matching.The process knowledge base and the process knowledge recommendation software are built in the avionics and electrical specialty of a certain aircraft.

关 键 词：工艺知识库 语义分析 深度学习 

分 类 号：TP391.72[自动化与计算机技术—计算机应用技术]