| 학회 | 한국화학공학회 |
| 학술대회 | 1998년 봄 (04/24 ~ 04/25, KOEX) |
| 권호 | 4권 1호, p.405 |
| 발표분야 | 공정시스템 |
| 제목 | 불안정 공정에 있어서 향상된 DMC의 강건성 분석 |
| 초록 | Model Predictive Control (MPC) has emerged as a powerful practical control technique during the last decade. There are several representative MPCs such asDynamic Matrix Control (DMC), Model Algorithmic Control (MAC), Internal ModelControl (IMC), Quadratic Programming Solution of Dynamic Matrix Control (QDMC),etc.. The above-mentioned control algorithms have been applied successfully in manyindustrial chemical processes. These control techniques use the discrete convolutionmodel to represent the process output, such as step response coefficients and impulseresponse coefficients of the reference models or real data models, and control theprocess by several optimization tools using the process outputs over some finite futuretime interval. In the algorithms of MPCs an explicit dynamic model of the processesare usually used to predict the effect of future actions of the manipulated variables onthe outputs. We can determine the future moves of the manipulated variables using theoptimization with the objective of minimizing the prediction error subject to operationconstraints.We already proposed the enhanced strategy of DMC for the open loopunstable processes in the last spring symposium. In this article, we studied therobustness of the enhanced strategy of DMC against the model mismatch and the noiseeffect of process outputs for the open loop unstable processes such as integratingprocesses and the process with Right-half Plan(RHP) poles. Gernerally the dynamicmatrix control is hardly applicable to the noisy processes and poorly defined model. |
| 저자 | 구도균, 손세훈, 이동권, 정재학 |
| 소속 | 영남대 |
| 키워드 | MPC; DMC; Robustness |
| 원문파일 | 초록 보기 |
