HCMC UNIVERSITY OF TECHNOLOGY AND EDUCATION
FACULTY OF ELECTRICAL AND ELECTRONIC ENGINEERING
Automatic Control Department
Digital Control of Dynamic Systems
Assoc. Prof. Truong Dinh Nhon, PhD
Automatic Control Department
Email: nhontd@[Link]
Cell phone: 0903675119
Website: [Link]
1
CV
Dinh-Nhon Truong received the Ph.D. degree from the Department of
Electrical Engineering, National Cheng Kung University, Tainan, Taiwan. He has
been a lecturer at the Faculty of Electrical and Electronics Engineering,
Hochiminh City University of Technology and Education, Vietnam since 2005.
His research interests are grid-connected wind power systems, dynamic stability
and FACTS devices.
2
Contents
3
Contents
Introduction to control systems; control system examples;
basic principle of feedback control; brief review of control
development history.
Digital control systems design: Time domain
specifications; Dynamic response to unit step and ramp
functions. Stability of discrete time systems; Digital PID
design; Pole placement design.
4
Contents
Digital Control system design: Frequency domain
specifications; Gain and phase margins; Compensator
design with bilinear transformations.
Digital control system design through state space
approach: State space description of discrete systems;
State feedback design via pole placement; State
estimator design; Controller design with state estimator.
5
Reference
[1]. GF Franklin, JD Powell and ML Workman, Digital Control of
Dynamic Systems, 3rd Edition, Addison Wesley, 1998.
[2]. MATLAB as Interface for Intelligent Digital Control of Dynamic
Systems.
6
Introduction
What is a control system ?
1) How to describe the system to be controlled? (Modelling)
2) How to design the controller? (Control)
7
Introduction
8
Modeling of a physical system
A simple mechanical system
9
Modeling of a physical system
A simple mechanical system
1
Modeling of a physical system
A simple mechanical system
1
Modeling of a physical system
A simple mechanical system
(ordinary differential equation)
1
Modeling of a physical system
A simple mechanical system
In frequency domain
1
Modeling of a physical system
Block diagram of a Control system
1
Overview
1
Modeling of a physical system
Block diagram of a Control system
1
Design a digital controller — from continuous to digital
Block diagram of a Control system
1
Design a digital controller — from continuous to digital
Block diagram of a Control system
1
Why digital control?
Control systems in continuous-time setting in the past are
usually implemented using analogue devices such as R, L,
C, and Op-amp together with some necessary mechanical
components. These devices are neither economical nor
durable.
The advances in computer technologies make the
implementation of control systems in discrete-time setting
(i.e., digital controllers) much more efficiently and
economically possible.
1
Why digital control?
Most of control systems nowadays are implemented using
either computers such as PCs or Digital Signal Processes
(DSP), which are specially designed to carry out
computations related to control algorithm realizations.
The advantages of digital controllers using PC or DSP are
obvious — it is fast, reliable, reusable and can be modified
thru simple recoding whenever needed.
2
Design a digital controller — from continuous to digital
2
Design Specifications
Time domain
2
Design Specifications
Steady state accuracy - Continuous
2
Design Specifications
Steady state accuracy - Discrete
2
Design Specifications
Stability - Continuous
2
Design Specifications
Stability - Descrete
2
Second order system - Continuous
2
Second order system - Continuous
2
Second order system - Discrete
2
Second order system - Discrete
3
Second order system - Discrete
3
Second order system - Discrete
3
Time Domain Design
3
Time Domain Design
3
Time Domain Design
3
Time Domain Design
3
Time Domain Design
3
Time Domain Design
3
Time Domain Design
Sampling rate 30 times the bandwidth
3
Digital PID control design via pole placement technique
4
Digital PID control design via pole placement technique
4
Digital PID control design via pole placement technique
4
Digital PID control design via pole placement technique
4
Digital PID control design via pole placement technique
4
Digital PID control design via pole placement technique
4
Digital PID control design via pole placement technique
4
Digital PID control design via pole placement technique
4
Digital PID control design via pole placement technique
4
Digital PI controller
4
Digital PI controller
5
Digital PI controller
5
Digital PI controller
5
Digital PI controller
5
Frequency Domain Design
5
Frequency Domain Design
5
Frequency Domain Design
5
Frequency Domain Design
5
Frequency Domain Design
For discrete systems, the idea is to use a so-called w-transformation (actually it is
a bilinear transformation) to transform the discrete system into a w-domain
system, which is pretty much the same as a continuous system.
Everything we have learnt from the compensation design for continuous systems
can be directly applied to yield a necessary compensator in w-domain, which can
be transformed back to a discrete version through an inverse w-transformation (=
an inverse bilinear transformation)
5
Frequency Domain Design
Lead compensator
5
Frequency Domain Design
Lead compensator
6
Frequency Domain Design
Lead compensator
6
Frequency Domain Design
Lead compensator
6
Frequency Domain Design
Lead compensator
6
Frequency Domain Design
Lead compensator
6
Frequency Domain Design
Lead compensator
6
Frequency Domain Design
Lead compensator Verification in w-domain (pseudo s-domain)
6
Frequency Domain Design
Lead compensator Verification in z-domain
6
Frequency Domain Design
Lead compensator Verification in z-domain
6
State Space Approach
6
State Space Approach
State variable technique is to convert transfer function system
representations into some first order difference equations (in a matrix
form), which many advanced matrix theories and computational tools can
be applied to. Thus, control system design using the state variable
technique can be done in a very systematic fashion and many well-
developed commercial software tools such as MATLAB can be readily
and easily utilized.
7
State Space Approach
7
State Space Approach
7
State Space Approach
7
State Space Approach
7
State Space Approach
7
State Space Approach
Verify for this example by using Matlab
7
State Space Approach
7
State Space Approach
Digital control system design with state feedback
7
State Space Approach
State feedback design via pole placement
7
State Space Approach
State feedback design via pole placement
8
State Space Approach
State feedback design via pole placement
8
State Space Approach
State feedback design via pole placement
8
State Space Approach
State feedback design via pole placement
8
State Space Approach
State feedback design via pole placement
8
Example
Ex_1
8
Example
Ex_2
8
Example
Ex_3
8
Example
Ex_4
8
State estimator
In the state feedback design, we have assume the state variable x of the
plant being available for feedback. Unfortunately, this is usually not the
case in most practical situations.
In real life, usually, we can only access to the system output, i.e., y,
instead. In order to implement the beauty of the state feedback control,
we will have no choice but to estimate the state variables using only the
system output and hope the implementation using such an estimation
works, which turns out indeed the case.
As such, implementing a state feedback control through a state estimator
becomes the standard approach in modern control nowadays.
8
State estimator
9
State estimator
9
State estimator
9
State estimator
9
State estimator
9
Homework:
Group of two
Simulate the results using Matlab
9
Homework 1
9
Homework 2