Adaptive Model Predictive Control /
The petro-chemical industry is characterized as having very dynamic and unpredictable marketplace conditions. The design of advanced control systems for chemical processes is quite a challenging task since it requires the satisfaction of many physical constraints on the values of the controlled vari...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
2000.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=728332461&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | The petro-chemical industry is characterized as having very dynamic and unpredictable marketplace conditions. The design of advanced control systems for chemical processes is quite a challenging task since it requires the satisfaction of many physical constraints on the values of the controlled variables. Conventional control theory is not capable of incorporating such constraints directly into the control system design. Instead, constraints are handled in a mostly ad hoc fashion, once the design has been carried out. Another important characteristic of conventional automatic control theory is that a considerable amount of effort is expended in first stabilizing an open loop unstable plant. However, for most process control applications, the plant is already open loop stable to start with thereby making the initial stabilization step unnecessary. What is therefore desirable for process control applications is a control scheme which directly handles constraints and also does not destabilize a plant which is stable to start with. Model predictive control schemes [1] apparently possess both these characteristics and this is what accounts for their immense popularity in process control applications. The purpose of adaptive MPC design is to try to give both stability and robustness to the controlled process even when the internal model used as a basis for the MPC design is unavailable or changes with time. In this dissertation, the theoretically provable guarantees of stability are brought into adaptive MPC designs by employing a unified design procedure for various MPC algorithms. Simulation examples are also presented in each case. An important contribution of the work reported here is that it bridges the traditionally diverse areas of adaptive system theory and industrial process control. |
|---|---|
| Item Description: | Vita. "Major Subject: Electrical Engineering". |
| Physical Description: | ix, 98 leaves : illustrations ; 28 cm. Issued also on microfiche from University Microfilm Inc. |
| Bibliography: | Includes bibliographical references (leaves 89-96). |