Scribd
Upload
79 views

Advanced Control, Lecture 1, Introduction

The document outlines the topics to be covered in an advanced control course, including analysis of stability, controllability, and observability using state-space models, as well as the design of state feedback controllers, observers, linear quadratic regulators, and Kalman filters. Students will apply these concepts to a mechanical system case study and complete a research proposal, literature review, and controller design project over the course of the semester.

Uploaded by

sepehr
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
79 views

Advanced Control, Lecture 1, Introduction

The document outlines the topics to be covered in an advanced control course, including analysis of stability, controllability, and observability using state-space models, as well as the design of state feedback controllers, observers, linear quadratic regulators, and Kalman filters. Students will apply these concepts to a mechanical system case study and complete a research proposal, literature review, and controller design project over the course of the semester.

Uploaded by

sepehr
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 18

Advanced Control

Saeed Behbahani
behbahani@iut.ac.ir

Department of Mechanical Engineering 　


Isfahan University of Technology
Refrences
• Chi-Tsong Chen, Linear System Theory and
Design, Oxford University Press, Third Edition,
1999.
• Belanger, Control Engineering, Sounders
College Publishing, 1995
• Ashish Tewari, Modern Control Design with
MATLAB and SIMULINK, John Wiley & Sons,
LTD., 2002.
• G.F. Franklin, J.D. Powel, M. Woekman, Digital
Control of Dynamic Systems, Addison Wesley
Longman, Inc., Third Eddition, 1998.
Evaluation
• Exam 60
• Midterm 30%
• Final 30%
• Research Project 25%
• Homework, etc. 15%
Project
1- Select a mechanical plant as the case 10/9/99
study and prepare a research proposal for
modern control of the selected system

2- Perform literature review for the selected 30/9/99


case study and write a literature review report

3- Design a controller for the selected case 10/11/99


study (Modeling, controller design, simulation,
comparison with literature, discussion, etc.)
1st week lectures
• Overview of the course
• Model descriptions
• Examples
• Linearization
• Chapter 1 & 2 in Chen’s book
– One may skip reading
• page 16, Section 2.3.1
• page 27 (after Example 2.11) - page 29
Automatic control design
Mechanical
System ID
Real system Electrical
Physical law Modeling
Chemical
etc.
Math model Implementation

Analysis
Design Controller
Stability Analysis
Stabilization
Controllability Simulation
Regulation
Observability
Disturbance rejection
Minimality etc.
Estimation etc.
Goal of the course
• Control theory with state-space models
• Systematic tools for analysis and design of linear
control systems
– Analysis
• Stability, controllability and observability, realization theory
and minimality
– Design
• State feedback, observer, Linear Quadratic (LQ) regulator,
Kalman filter, Linear Quadratic Gaussian (LQG) control
– Matlab implementation
• Foundations for advanced control courses
State-space model
Continuous-time Discrete-time

x : state vector u
: input vector y :
output vector

System System
u y u y
Example: Mass-Spring-Damper

Input-Output Differential
Equation
State Space Model
Input-Output Differential
Equation

For Linear Systems:


Mathematical Representations
Input-output Differential Equation:
Example:

Initial Condition:

State Space Model:


Classic Control Vs. Modern Control
Based on Transfer function model Based on State Space model

Applicable mostly for SISO systems Applicable for MIMO systems

Output variables are fed beck State vector is fed back

Not good for a large order system Applicable for large order system

Only few of closed loop poles can be All the closed loop poles can be
varied by varying a limited number of controlled.
controller parameters, not all of them
Not necessarily optimum Controller can be designed optimally
Advantages of methods using
State-Space models
• Multivariable systems
• Time-varying systems
• Nonlinear systems
• Computationally reliable
Analysis

Stability
Q. Does the state x0
converge to the origin?
or remain bounded?
How can we know the
state trajectory
property?
Analysis (cont’d)

Controllability Observability
Q. Is there an input u that Q. Can we determine the
transfers x0 to the origin initial state uniquely
in a finite time? from the observations?

x0
x0
u y

System
Design
State feedback Observer

u x u y
v system system

K observer

Q. What is achievable with Q. Can we make an


a state feedback? observer to estimate x?

Q. How to combine these?


Observer-based control
Optimal design
Linear quadratic (LQ) Kalman filter

u x v
system u system y

KF w
Q. What is the optimal u?
Q. How to obtain a “best”
estimate x in noisy case?

Q. How to combine these?

Linear quadratic Gaussian (LQG)


Advanced control courses
• Robust control
• Nonlinear control
• Adaptive control
• Digital and sampled-data control
• Optimal control
• System identification etc.

You might also like