Face Milling 공정에서 신경회로망을 이용한 공구 마모 감지에 대한 연구
A study on the tool wear monitoring by neural network in the face milling process
- 발행기관 江陵大學校 大學院
- 지도교수 崔德基
- 발행년도 2007
- 학위수여년월 2007. 2
- 학위명 석사
- 학과 및 전공 精密機械工學科
- 원문페이지 vi, 49 p.
- 본문언어 한국어
초록/요약
On-line detection system of abnormal states in a machining process needs to be developed to implement the IMS(Intelligent Manufacturing System). Generally it is difficult to determine the exact point of time for tool change because a tool wear grows progressively on the contrary to other abnormal states. In this article, the shape parameters of cutting force within 1 revolution were proposed as features to detect a tool wear. The feasibility of the shape parameter was discussed and verified through the tool wear experiments. And backpropagation neural network(BPNs) was used for detection of tool wear. Input vector of neural network comprise of variance, skewness and kurtosis thrust force signals.
more목차
1. 서론 = 1
2. 연구개요 = 3
2-1 공구 마모 = 3
2-2 형상 계수 = 7
2-3 신경회로망 = 10
3. 실험 장치 및 장비 = 19
3-1 실험 장치 = 19
3-2 실험 방법 = 24
4. 실험 및 고찰 = 25
4-1 특징 추출 = 25
4-2 형상 지수를 이용한 감지 시스템 = 41
5. 결론 = 46
참고문헌 = 47
감 사 글 = 49
