적응형 인자 모델을 이용한 개선된 진공펌프 상태진단에 관한 연구 = Study on Vacuum Pump Monitoring Using Adaptive Parameter Model

다국어 초록 (Multilingual Abstract)

This paper introduces statistical features observed from measured batch data from the multiple operation state variables of dry vacuum pumps running in the semiconductor processes. The amplitude distribution characteristics of such state variables as inlet pressures, supply currents of the booster and dry pumps, and exhaust pressures are shown to be divided into two or three distinctive regions. This observation gives an idea of using an adaptive parametric model (APM) chosen to describe their statistical features. This modelling, in comparison to the traditional dynamic time wrapping algorithm, is shown to provide superior performance in computation time and memory resources required in the preprocessing stage of sampled batch data for the diagnosis of running dry vacuum pumps. APM model-based batch data are demonstrated to be very appropriate for monitoring and diagnosing the running conditions of dry vacuum pumps.

참고문헌 (Reference)

1 R. Bahren, 44 : 709-, 1993

2 V. Pravdova, 456 : 77-, 2002

3 R. Bro, 17 : 16-, 2003

4 W. S. Cheung,

5 W. S. Cheung,

6 A. Kassida, 44 : 846-, 1998

7 이상윤, "반도체/디스플레이 공정급 건식진공펌프 개발 개요" 한국진공학회 19 (19): 265-274, 2010

8 W. H. Press, "Numerical recipes in Fortran the art of scientific computing second edition" Cambridge university press 1997

9 R. L. Mason, "Multivariate Statistical Process Control with Industrial Applications" ASA SIAM 2002

10 E. J. Dudewicz, "Modern Mathematical statistics" Wiley 1988

11 성동원, "MPCA 기반의 통계기법을 이용한 진공펌프 상태진단에 관한 연구" 한국진공학회 15 (15): 338-346, 2006

12 J. E. Jackson, "A User’s guide to Principal Component" John Wiley and Sons 1991