| 학회 | 한국화학공학회 |
| 학술대회 | 2012년 가을 (10/24 ~ 10/26, 부산 BEXCO) |
| 권호 | 18권 2호, p.1701 |
| 발표분야 | 공정시스템 |
| 제목 | Development of fault information methodologies for monitoring and diagnosis sysmtem |
| 초록 | Fault detection and diagnosis are highly developed in academic area. Therefore, there are so many algorithms and these application. However, in industrial world, just univariate sensing data is used. So it is necessary to close this gap. To improve monitoring and diagnosis system in industrial world, multivariate methodology is applied. In this study, PCA, one of the most common multivariate methodologies, is suggested as a monitoring system. When fault is detected by PCA method, three steps methodology for diagnosis are suggested; fault propagation, fault magnitude and its event matrix analysis. This suggested methodology is demonstrated to application of LNG fractionation process. As a result, the application of this methodology has rapid detection and accurate performance compared to using only univariate monitoring and fault diagnosis system. This research was supported by grant from the LNG Plant R&D Center funded by the Ministry of Land, Transportation and Maritime Affairs (MLTM) of the Korean government. |
| 저자 | 편하형, 하대근, 한종훈 |
| 소속 | 서울대 |
| 키워드 | monitoring; diagnosis; Fault information |
| 원문파일 | 초록 보기 |
