基于ART-2神经网络的设计知识需求研究  

On Acquisition of Product Design Knowledge by an ART-2 Neural Network

作　　者：黄卫东[1] 谢强[1] 丁秋林[1] 

机构地区：[1]南京航空航天大学

出　　处：《机械科学与技术》2008年第3期399-404,共6页Mechanical Science and Technology for Aerospace Engineering

基　　金：国防基础预研项目(S0500A001)资助

摘　　要：通过分析产品设计知识获取和知识共享研究中存在的问题,结合本体理论的运用,构建了产品设计中的概念语义,并利用OWL+RDF+XML形式化语言给出了案例知识的表达,在研究ART-2神经网络竞争学习和自稳机制的基础上,设计了案例的多级组织,实现了基于案例推理(CBR)的检索机制,提升了案例分类和检索的效率。同时,利用案例的聚类和学习机制,研究特征权值的自动获取和优化,并通过应用实验验证了该研究成果的实用性和有效性,从而较好地解决了产品设计中的知识需求。We first analyzed the problems in acquisition of product design knowledge and knowledge sharing. Then we constructed the semantic concepts in product design based on ontology theory, and presented the representation of case knowledge by means of OWL ＋ RDF ＋ XML language. Utilizing the competitive learning capability and selfstabilization mechanism of an ART-2 （adaptive resonance theory） neural network, we designed a case multistage organization, realized the retrival methodology in case-based reasoning （CBR）, and raised the efficiency of case classification and retrival. Based on case clustering and learning mechanism, we also realized automatic acquisition and optimization of characteristic weight. Experiment results show that our method is practical and effective for solving the knowledge demand problem in product design.

关 键 词：知识需求 ART-2 CBR 本体 设计 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]