Proceedings of International Conference on Leading Edge Manufacturing in 21st century : LEM21
Online ISSN : 2424-3086
ISSN-L : 2424-3086
セッションID: A025
会議情報
A025 Experimental Verification of Ball End-milling Condition Decision Support System Applying Hierarchical and Non-hierarchical Clustering Methods
Hiroyuki KODAMAToshiki HIROGAKIEiichi AOYAMAKeiji OGAWA
著者情報
会議録・要旨集 フリー

詳細
抄録
Unskilled engineers have difficulty determining the appropriate end-mills and end-milling conditions for materials and shapes designed by CAD, even though end-mills are specifically designed for various purpose such as milling at high speed and milling of difficult-to-cut materials. Ball end-mills are usually the most suitable for die and mold milling since they can be easily adapted to workpieces with various complicated shapes. We previously reported an end-milling condition decision support system ("catalog mining system") that applies data-mining methods from square end-mill tool shape parameters listed in a cutting tool catalog. Our aim was to extract new knowledge by applying data-mining techniques to a tool catalog. We used both hierarchical and non-hierarchical clustering methods and principal component regression. We focused on the shape element of catalog data and visually clustered ball end-mills from the viewpoint of tool shape, which here meant the ratio of dimensions, by using the k-means method. Expressions for calculating end-milling conditions were derived using the response surface method. We have now conducted end-milling experiments using ball end-mills and compared the calculated values with the catalog ones to validate end-milling conditions derived from catalog-mining system.
著者関連情報
© 2013 一般社団法人 日本機械学会
前の記事 次の記事
feedback
Top