Spring 2002 Course Announcement
CHFEN 7203: Advanced Process Control
Class Number 12925, 3 Credit Hours

Offered on Mondays, Wednesday and Fridays from 9:40 to 10:30 in JTB 110 (James Talmage Building on Presidents Circle).

DESCRIPTION:

The focus of this course is on robust multivariable control. We will emphasize the ability to solve control problems rather than mathematical theory. Nevertheless, this course will require a solid background in classical and state-space control methods and may be quite challenging to some students.

COVERAGE:

1.

We will quickly review/introduce fundamentals of robust multivariable design - definitions, analysis and performance of multivariable linear time invariant systems, interactions, directionality and robustness of MLTI systems. This part of the course will take roughly one-third of the semester.

2.

We will then consider the shaping of sensitivity and complementary sensitivity of MLTI systems, specification of model and/or disturbance uncertainty and desired performance in terms of the generalized plant (linear fractional transformation framework).

3.

We will use the second half of the semester to study controller design methods for MLTI systems. This includes LQR/LQG methods, frequency-weighted LQR/LQG design, minimax design as it relates to $H_\infty$ and H₂ design methods and, finally, H₂ and $H_\infty$ synthesis. Time permitting, we will also consider $\mu$-synthesis methods.

CLASS OBJECTIVE:

At the end of the class, you should be able to independently formulate a robust control problem for MLTI systems and solve it using the Robust Control toolbox in Matlab.

TEXTBOOKS:

We will use one or both of the following books:

1.

Sigurd Skogestad and Ian Postlethwaite, Multivariable Feedback Control. Wiley, 1996.

2.

Kemin Zhou with John C. Doyle, Essentials of Robust Control. Prentice Hall, 1998.

SOFTWARE:

Matlab with Robust Control toolbox.

PREREQUISITES:

No formal prerequisites. However, students are assumed to have a ``working'' knowledge of classical SISO control, linear algebra, matrix analysis. Knowledge of state-space methods will be very helpful. If you want to take this course as a first graduate control class, you may have trouble keeping up.

HOMEWORK:

I hope to have 8 or more sets of homework problems. Most (if not all) assignments will require computer simulation. A class project involving a literature review, computer simulation, a written report and an oral presentation at the end of the semester will be required.

EXAMINATIONS:

All examinations will be of the take-home variety.

INSTRUCTOR:

Mikhail Skliar
Office: 3290G MEB
Phone: 581-6918
E-mail: Mikhail.Skliar@utah.edu