Computer Integrated Manufacturing System ›› 2022, Vol. 28 ›› Issue (9): 2683-2699.DOI: 10.13196/j.cims.2022.09.004

Previous Articles     Next Articles

Structural solution generation method of fuzzy front-end requirements based on knowledge graph

MO Zhenchong1,GONG Lin1,2+,YE Fan1,ZHU Degang1,XIE Jian1   

  1. 1.School of Mechanical Engineering,Beijing Institute of Technology
    2.Yangtze Delta Region Academy of Beijing Institute of Technology
  • Online:2022-09-30 Published:2022-10-23
  • Supported by:
    Project supported by the National Key Research and Development Program,China(No.2018YFB1700802).

基于知识图谱的模糊前端需求结构解生成方法

莫振冲1,宫琳1,2+,叶帆1,祝德刚1,谢剑1   

  1. 1.北京理工大学机械与车辆学院
    2.北京理工大学长三角研究院(嘉兴)
  • 基金资助:
    国家重点研发计划资助项目(2018YFB1700802)。

Abstract: In the fuzzy front-end stage of product conceptual design,there are a large number of expected requirements for product performance represented by natural language.This type of requirement is not only an important measure of user satisfaction with products,but also a source of motivation to promote product innovation and development.However,this type of requirement is often the one with the most significant characteristics of fuzziness,incompleteness and subjective limitations,and lacks accuracy in the representation and transformation of product conceptual design.By quoting the general engineering parameters in the classical TRIZ theory,the Product (patents/requirements)—General engineering parameters—Structure (PGS) mapping model for the above design problems was constructed,which laid a foundation for the mapping transformation of fuzzy front-end requirements to the design process at the theoretical level.Under the guidance of the mapping model,the domain knowledge graph was constructed based on the patent data,which realized the identification and extraction of performance parameter entities and enhanced the availability of the knowledge graph supporting conceptual design.Combined with the PGS mapping model and domain design knowledge graph,the structural representation and completion of the fuzzy front-end requirements in the form of engineering parameters were realized,and the structural solution space of conceptual design was further constructed.The knowledge graph was constructed in the field of mechanical equipment and verified by an example.

Key words: concept design, fuzzy requirements, product design, knowledge graph, engineering parameters

摘要: 产品概念设计模糊前端阶段存在大量自然语言表征的产品性能期望型需求,这一类需求不仅是用户对产品满意度的重要衡量标准,还是推动产品创新发展的源动力。然而,这一类需求也往往是模糊性、非完备性和主观局限性特征最为显著的一类需求,在面向产品概念设计的表征和转换等阶段的精确性不足。针对上述设计问题,首先通过引用经典TRIZ理论中的通用工程参数,构建了产品(专利/需求)—通用工程参数—结构(PGS)映射模型,在理论层面上为模糊前端需求面向设计过程的映射转换奠定了基础;其次,在所述映射模型的引导下,基于专利数据构建了领域知识图谱,实现了性能参数型实体的识别与提取,增强了支持概念设计的知识图谱可用性;再次,结合PGS映射模型及领域设计知识图谱,实现了工程参数形式的模糊前端需求结构化表征与补全,并进一步构建了概念设计结构解空间;最后,在机械设备领域构建了知识图谱并完成了实例验证。

关键词: 概念设计, 模糊需求, 产品设计, 知识图谱, 工程参数

CLC Number: