Data mining method for design knowledge in mechanical structure symmetry instances

YANG Juncheng, MA Zhiyong, ZHANG Taixia

(1.Faculty of Mechanical Engineering & Mechanics, Ningbo University , Ningbo 315211, China;
2.Zhejiang Provincial Key Lab of Part Rolling Technology, Ningbo 315211, China)

Abstract: Aiming at a lot of design knowledge in mechanical structure symmetry instances,by using the multidimensional fuzzy association rules mining method, design knowledge was mined from structure symmetry instances to guide the scientific application of structure symmetry in mechanical product design. The data to be mined in mechanical structure symmetry instance database was processed by fuzzy method based on the design requirements and the characteristics of the membership function, and the association rules mining algorithm Apriori was used to mine the fuzzy association rules. The mining results include the relationships among the structural symmetry, the mechanical product function, the mechanical product performance and the design restriction. The results indicate that the practicable structural symmetry design knowledge in mechanical systems is established by mining the structure symmetry instances,and the reference of further research on the application rules and application methods of mechanical structure symmetry.

Key words: mechanical structure symmetry; data mining; membership function;fuzzy association rules; design knowledge