当前位置: X-MOL 学术CIRP Ann. Manuf. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Where and how to find bio-inspiration?
CIRP Journal of Manufacturing Science and Technology ( IF 4.8 ) Pub Date : 2020-10-26 , DOI: 10.1016/j.cirpj.2020.09.013
Mart Willocx , Amir Ayali , Joost R. Duflou

Biologicalisation calls for the integration of biological knowledge in manufacturing. Although biologists have been cataloguing biological knowledge for centuries, for the non-biologist finding this inspiration as input for bio-inspired design is a major challenge. In this paper, three different methods are used to find bio-inspiration for a case study on the clamping interface of a flexible mobile machining unit: the AskNature database, a natural language processing (NLP) approach, drawing on a large corpus of biological publications, and an informal consultation of an expert biologist as could be organised by a design office. The comparison of the retrieved principles with AskNature as a baseline indicates that the NLP approach allows retrieving publications about relevant biological strategies with a good recall performance, without involving an expert. However, extracting the working principles from the biological articles retrieved with the NLP approach is found to be error-prone and time consuming.



中文翻译:

在哪里以及如何找到生物灵感?

生物化要求将生物学知识整合到制造业中。尽管生物学家对生物学知识进行分类已有数百年历史,但对于非生物学家而言,找到这一灵感作为生物灵感设计的输入是一项重大挑战。在本文中,使用三种不同的方法来寻找生物灵感,以对柔性移动加工单元的夹紧界面进行案例研究:AskNature数据库,一种自然语言处理(NLP)方法,借鉴了大量生物出版物,以及由设计办公室组织的专家生物学家的非正式咨询。将检索到的原则与​​AskNature作为基准进行比较,表明NLP方法可以检索具有良好召回性能的有关生物策略的出版物,没有专家参与。但是,从使用NLP方法检索的生物制品中提取工作原理发现容易出错且耗时。

更新日期:2020-10-30
down
wechat
bug