한국 전통 유기 제작에서 결함을 방지하기 위한 기계 학습 기반의 공정 조건 선택 방안

이승철; 한도석; 이혁; 김낙수

상세 보기

한국 전통 유기 제작에서 결함을 방지하기 위한 기계 학습 기반의 공정 조건 선택 방안

Machine Learning-based Process Condition Selection Method to Prevent Defects in Korean Traditional Brass Casting

이승철;
한도석;
이혁;
김낙수

초록

In the present study, in order to prevent the misrun defects that occur during traditional brass casting, a method for selecting the proper casting process conditions is proposed. A learning model was developed and demonstrated to be able to learn the presence or absence of defects according to the casting process conditions and to predict the occurrence of defects depending on the certain pro- cess given. Appropriate process conditions were determined by applying the proposed method, and the determined conditions were verified through a comparison of different simulation results with additional conditions. With this method, it is possible to determine the casting process conditions that will prevent defects in the desired sand model. This technology is expected to contribute to real- ization of smart traditional brass farming workshops

키워드

사형 주조; 미충전 결함; 주조 결함 예측; 기계 학습; 인공신경망; Sand casting; Misrun; Casting defect prediction; Machine learning and Artificial neural network.

제목: 한국 전통 유기 제작에서 결함을 방지하기 위한 기계 학습 기반의 공정 조건 선택 방안

제목 (타언어): Machine Learning-based Process Condition Selection Method to Prevent Defects in Korean Traditional Brass Casting

저자: 이승철; 한도석; 이혁; 김낙수

발행일: 2022-08

저널명: 한국주조공학회지

권: 42

호: 4

페이지: 209 ~ 217