Machine Tool Chatter Diagnosis Algorithm Research using MFCC Method

Title
Machine Tool Chatter Diagnosis Algorithm Research using MFCC Method
Other Titles
MFCC를 이용한 공작 기계 채터 진단 알고리즘 연구
Author
박준홍
Keywords
Chatter; Speaker Recognition; Diagnosis; 채터; 화자 인식; 진단
Issue Date
2016-10
Publisher
한국소음진동공학회
Citation
한국소음진동공학회 2016년도 추계 학술대회 초록논문집, Page. 110-110
Abstract
Whenever chatter signal appears in the process of manufacturing workpiece by machinery tool, unusual noise and vibration signals inevitably occur. A variety of faults are significantly correlated with this phenomenon. Changing manufacturing condition such as rotational speed and depth of machine tools blade, the signal of chatter occurrence was measured using vibrational accelerometers. In this research, the chatter diagnosis algorithm using Mel Frequency Cepstral Coefficient (MFCC) and Deep blief network (DBN), one of the Deep learning algorithm is proposed. Deep belief network is effective to classify complicated and various signals as it mimics the hierarchical cognitive process of a human brain using neurons. MFCCs, widely used for natural language recognition are coefficients that collectively make up a Mel Frequency Cepstrum (MFC) which is acquired by adapting Mel filter bank to the FFT of a vibration signal. In this paper, to acquire features from the distincive noise and vibration between normal and chatter state MFCCs is used. Chatter diagnosis algorithm, DBN using MFCCs as input features, is suggested.
URI
http://www.dbpia.co.kr/Journal/ArticleDetail/NODE07041481http://repository.hanyang.ac.kr/handle/20.500.11754/81334
Appears in Collections:
COLLEGE OF ENGINEERING[S](공과대학) > MECHANICAL ENGINEERING(기계공학부) > Articles
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE