Catalog data:
Control engineering: basic theory. State space description of linear dynamic systems.
Stability theory, frequency domain analysis. Controllability, observability. Loop shaping
for dynamic response. Pole assignment, state feedback. Linear observers, Kalman filter.
Design and separation principle. LQ regulator and LQG theory. LQ control system design,
dynamic compensation. Tracking control, servomechanism design. |
Course
year:
MSc 1st year
Semester: 1A
Hours p/w: 4
Other hours: --
Assessment: See remarks.
Assessm.period(s): See remarks.
(see academic calendar) |
Prerequisites:
wb2310 |
Follow
up: wb2305 |
Detailed
description of topics:
Control engineering: basic theory. State space description of linear dynamic systems.
Realization of transfer function models by state space models. Controllability,
observability, minimal order. Parallel and series connection, pole-zero cancellation,
relationship with controllability and observability. Controllability and observability
canonical forms. Jordan canonical form. Stability theory, frequency domain analysis.
Dynamic response, relationship with pole and zero locations in the complex domain. Loop
shaping for dynamic response, robustness indicators. Multi-input and multi-output systems.
Pole assignment, design of state feedback. Linear observers, Kalman filter. Design of
observer. Control design and separation principle. LQ regulator and LQG theory. Algebraic
Riccati equation, choice of performance criteria. Asymptotic analysis, LQ control system
design, dynamic compensation. Disturbances and reference signals, modelling of exogenous
variables. Internal model principle, design of tracking control systems, servomechanism
design. |
Course
material:
ISBN: [0-07-022441-2] Friedland,B. Control System
Design: An Introduction to State-Space Methods. McGraw-Hill Book Co., New York, NY, 1986.
|
References
from literature:
-
[1] ISBN: [0-13-589763-7] Brogan,W.L., Modern
Control Theory. 3rd Edition, Prentice Hall, Inc., Englewood Cliffs, NJ, 1991.
-
[2] ISBN: [4-8337-0191-X] Chen,Chi-Tsong, Linear System
Theory and Design, Holt,Rinehart and Winston, Inc., New York, NY, 1984.
-
[3] ISBN: [0-13-638560-5] Anderson,B.D.O. Moore,J.B.,
Optimal Control. Linear Quadratic Methods, Prentice Hall, Inc., Englewood Cliffs, NJ,
1990.
-
[4] ISBN: [0-12-527780-6] O'Reilly,J., Observers for
Linear Systems, Academic Press, London, 1983.
-
[5] ISBN: [0-13-638122-7] Anderson,B.D.O. Moore,J.B.,
Optimal Filtering, Prentice Hall, Inc., Englewood Cliffs, NJ, 1979.
|
Remarks
(specific information about assesment, entry requirements, etc.):
Assessment: computer aided control system design
exercise, where the student is required to apply the various approaches discussed in the
course. Exercise is done individually using Matlab computational environment and Matlab
Control System Toolbox or similar. The exercise can be executed throughout the year. The
succesful completion of the design exercise is a prerequisite for the participation in the
written examination.
|
Goals:
The course serves as an introduction to the
concepts and techniques currently used in basic modern control theory. The course requires
the development of the technical skills involved in state space system theory. It also
extends the notions of control system design towards time-domain techniques based on pole
placement and linear optimal control using quadratic performance criteria. The exercises
in the course stress the use of a computational linear-algebra environment (Matlab or
similar) for linear control system design. The exercises familiarize the student with
model-based control design, supported by modern computational tools for dynamic analysis,
simulation and control performance assessment.
|
Computer
use:
The computer will be used regularily in demos and
examples, and constitutes the main tool for the software environment (Matlab and
Toolboxes) used in the exercise.
|
Laboratory
project(s): |
Design
content:
The goal is control system design. The presentation
of theoretical issues is done with the purpose to use the notions involved immediately in
a design context.
|
Percentage
of design: 80% |