DOI QR코드

DOI QR Code

Performance Comparisons of Wavelet Based T2-Test and Neural Network in Monitoring Process Profiles

공정프로파일 모니터링에서 웨이블릿기 반 T2-검정과 신경회로망의 성능비교

  • 김성준 (강릉대학교 산업시스템공학과) ;
  • 최덕기 (강릉대학교 정밀기계공학과)
  • Published : 2008.12.25

Abstract

Recent developments of process and measurement technology bring much interest to the online monitoring of process operations such as milling, grinding, broaching, etc. The objective of online monitoring systems is to detect process changes as early as possible. This is helpful in protecting facilities against unexpected failures and then preventing unnecessary loss. This paper investigates, when the process monitoring data are obtained as a profile, the monitoring performances of a statistical $T^2$-statistic and a feedforward neural network by using a wavelet transform. Numerical experiments using cutting force data presented by Axinte show that the proposed wavelet based $T^2$-test has an acceptable power in detecting profile changes. However, its operating characteristic is very sensitive to autocorrelation. On the contrary, compared with $T^2$-test, the neural network has more stable performance in the presence of autocorrelation. This indicates that an adaptive feature to analyze noises should be incorporated into the wavelet based $T^2$-test.

최근 공정 및 계측기술이 발전함에 따라 밀링, 그라인딩, 브로칭 등 공정작업의 온라인 모니터링에 대한 관심이 높아지고 있다. 온라인 모니터링 시스템은 공구의 마모나 파손 등과 같은 공정변화를 가급적 조기에 발견함으로써 설비를 보호하고 불필요한 비용의 발생을 억제하는 데 그 목적을 두고 있다 본 논문에서는 온라인 공정관측 데이터가 프로파일로 주어질 때 웨이블릿변환을 이용한 $T^2$-검정과 신경회로망의 모니터링 성능에 대해 고찰한다. 2006년 Axinte가 제시한 절삭력 데이터를 이용하여 수치실험을 수행한 결과, 웨이블릿기반 $T^2$는 양호한 검출력을 나타냈지만 그 검사특성은 자기상관에 매우 민감하게 반응하였다. 반면, 자기상관의 존재 하에서도 신경회로망은 $T^2$-검정에 비해 매우 안정적인 검사특성을 갖는 것으로 나타났다. 이는 웨이블릿기반 $T^2$-검정에 노이즈분석을 위한 적응적인 요소가 필요하다는 점을 시사한다

Keywords

References

  1. D. A. Axinte, 'Approach into the use of probabilistic neural networks for automated classification of tool malfunctions in broaching,' International Journal of Machine Tools and Manufacture, Vol. 46, pp. 1445-1448, 2006 https://doi.org/10.1016/j.ijmachtools.2005.09.017
  2. Y. Ding, E. A. Elsayed, S. Kumara, J. C. Lu, F. Niu and J. Shi, 'Distributed sensing for quality and productivity improvement,' IEEE Transactions on Automation Science and Engineering, Vol. 3, No. 4, pp. 344-359, 2006 https://doi.org/10.1109/TASE.2006.876610
  3. D. C. Montgomery, Introduction to Statistical Quality Control, Fourth Edition, Wiley, New York, 2001
  4. L. Kang and S. Albin, 'On-line monitoring when the process yields a linear profile,' Journal of Quality Technology, Vol. 32, pp. 418-426, 2000 https://doi.org/10.1080/00224065.2000.11980027
  5. M. K. Jeong, J. C. Lu and N. Wang, 'Wavelet-based SPC procedure for complicated functional data,' International Journal of Production Research, Vol. 44, No. 4, pp. 729-744., 2006 https://doi.org/10.1080/00207540500222647
  6. J. Jin and J. Shi, 'Feature preserving data compression of stamping tonnage information using wavelets,' Technometrics, Vol. 41, pp. 327-339, 1999 https://doi.org/10.1080/00401706.1999.10485932
  7. J. Fan, 'Test of significance based on wavelet thresholding and Neyman's truncation,' Journal of American Statistical Association, Vol. 90, pp. 1200-1224, 1996 https://doi.org/10.2307/2291512
  8. 이승훈, 윤동한, 알기 쉬운 웨이브렛 변환, 진한도서, 2002
  9. 배현, 최대원, 천성표, 김성신, 김예진, '시계열데이터마이닝을 이용한 하수처리연속회분식반응기 장비 진단,' 퍼지 및 지능시스템학회 논문지, Vol. 15, No. 4, pp. 431-436, 2005 https://doi.org/10.5391/JKIIS.2005.15.4.431
  10. T. Liao, C. Ting, J. Qu and P. J. Blau, 'A wavelet-based methodology for grinding wheel condition monitoring,' International Journal of Machine Tools and Manufacture, Vol. 47, pp. 580-592, 2007 https://doi.org/10.1016/j.ijmachtools.2006.05.008
  11. Wavelet Toolbox, The Math Works, 2006
  12. M. S. Phadke, Quality Engineering Usig Robust Design, Prentice-Hall, 1989
  13. Neural Network Toolbox, The Math Works, 2006
  14. G. E. P. Box, G. M. Jenkins and G. C. Reinsel, Time Series Analysis: Forecasting and Control, 3rd Ed., Prentice Hall, 1994