| 학회 | 한국화학공학회 |
| 학술대회 | 2001년 봄 (04/27 ~ 04/28, 연세대학교) |
| 권호 | 7권 1호, p.533 |
| 발표분야 | 공정시스템 |
| 제목 | 한정적응모델예측제어 |
| 초록 | Constrained model predictive control is widely accepted as a standard advanced control strategies in process industries. However, it is not an easy task to find good models for model predictive control. Moreover, the process dynamics often changes due to contamination and so on. Under this circumstances, the adaptation is quite desired in practice. Recently, combining subspace identification and model predictive control, the so called subspace predictive control strategies are proposed [1]. In subspace identification, a optimal input-output relationship is first obtained and a state space model is found in the following steps. one can design model predictive control using the state space model from subspace identification. However, in the subspace predictive control, the optimal input-output relationship is directly used in model predictive and thus the number of steps required for model predictive control is substantially reduced. In this paper, a couple of adaptation laws are combined with this subspace predictive control and their performance are illustrated with examples. |
| 저자 | 최진훈, 고훈석, 이광순 |
| 소속 | 서강대 |
| 키워드 | Model Predictive Control; Adaptation; Constraints |
| 원문파일 | 초록 보기 |
