Location: JHE 326H
Robust analysis mathematical techniques are currently available to assess the stability and performance of closed loop systems in the presence of model error. Two techniques will be reviewed: a -Structured Singular Value method and a Quadratic Lyapunov function-based technique. The usefulness of these techniques is illustrated through their application to two problems that have been traditionally very difficult to solve since they require intensive computations. The use of robust analysis formulations is shown to provide approximate solutions to these problems with reasonable computational effort.
The problems to be discussed are as follows:
i- Design of a distributed Model Predictive Control strategy in the presence of model error: The idea behind this control strategy is that the inputs and outputs of a process are grouped into subsystems. Then, for each subsystem, a lower dimensional predictive controller can be formulated and the resulting set of predictive controllers is coordinated based on a predefined performance index. Both the unconstrained and constrained cases will be discussed.
ii- Simultaneous process design and control design of a chemical plant: In this simultaneous approach, the controller parameters are optimized together with the system design parameters to determine the optimal design and operating conditions of a chemical process. The solution to this optimization problem must address the trade-offs between conflicting design and control objectives.
Return to list of all seminars