| 학회 | 한국화학공학회 |
| 학술대회 | 2002년 봄 (04/26 ~ 04/27, 강원대학교) |
| 권호 | 8권 1호, p.349 |
| 발표분야 | 공정시스템 |
| 제목 | SMB 프로세스에 대한 식별 |
| 초록 | This work is concerned with the model identification for a simulated moving bed (SMB) process. Currently, the main applications to continuous chromatographic separation can be divided in two groups: first, the large-scale industrial production of relatively cheap speciality products and second the separation for high value products in small amounts, which very often exhibit separation factors near unity or which are thermally unstable. The separation costs in both cases are very high compare to the process costs. An optimal design and operation might therefore be the only possibility to exploit the economic potential of the process and to make its application feasible. A few rigorous design and optimisation procedures for SMB reactors based on dynamic process models are described in the literature but some drawbacks of the design methods for SMB separation process such as high computational load and difficulty in controller design make it necessary to introduce a more simple identified model. It is difficult to characterize the dynamic behaviour of an SMB process because the process is a hybrid one consisting of continuous and discrete components. Hence, the optimization of SMB processes is confined to finding the optimal steady state operating conditions using several guide rules such as the triangle theory but the operation of SMB process based on the guide rules is not sufficient to deal with various control problems such as disturbance rejection or setpoint tracking. In this study we identify an SMB process using the subspace identification method. The internal flow rate ratios are chosen as input variables and the averaged concentrations of rich components in raffinate and extrat are selected as output variables. |
| 저자 | 송인협1, 이현구2 |
| 소속 | 1서울대, 2화학공정신기술(연) |
| 키워드 | Identification; SMB unit |
| VOD | VOD 보기 |
| 원문파일 | 초록 보기 |
