Robust Model Reference Controller Design For Different Classes of Nonlinear Systems

Republic of Iraq
Ministry of Higher Education and Scientific Research

University of Technology
Control and Systems Engineering Department
Control Engineering Branch
Robust Model Reference Controller Design for

Different Classes of Nonlinear Systems
By: Musadaq A. Hadi

Supervised by: Prof. Dr. Hazem I. Ali
1
List of Contents
1. Introduction
2. Problem Statement
3. Aim of Thesis
4. Contributions
5. Control Algorithm Design
6. Study Cases
7. Validations
8. Conclusions
9. List of Publications
2
Introduction.1
Many physical systems are nonlinear to some degree, the

designed linear control systems will behave satisfactorily only
over a limited range of operation. Therefore, a nonlinear
controller should be designed in order to compensate these
systems. In this thesis, a new robust model reference control
algorithm is presented to compensate for the nonlinear
system problems. Many problems are presented and solved
by the proposed control algorithm such as nonlinearity,
uncertainty, instability, bad performance, time-varying, and
external disturbances. In the next slides, the control algorithm
is presented with these problems and the result shows the
effectiveness of its implementation.
3
Problem Statement.2
Nonlinear systems have come with many problems. These

problems are categorized into three groups such as:
• System inherited characteristics problems (nonlinearity,
uncertainty, instability, bad performance, external
disturbances and time-varying).
• System’s structure problems (complex systems, MIMO).
• Analysis/designing problems (model reference design for
MIMO).
4
3.Aim of Thesis
A new robust control algorithm is designed to achieve the
stability, tracking, robustness, and optimality as follow:
• Develop the mathematical model for the nonlinear system.
• Design/build a model reference to specify the desired
performance.
• Analyze the system using Lyapunov stability analysis to obtain
a proper controller that can achieve the stability.
• Use the Most Valuable Player Algorithm (MVPA) to optimize
the parameters of the proposed controller.
• Apply the controller to different nonlinear systems .
• Apply the proposed control algorithm to a real nonlinear
system (MagLev).
5
Contributions.4
The features of the control algorithm is that:
• it produces a controller for every analyzed nonlinear system.
• It uses Lyapunov stability analysis with indefinite time-derivative
function which is a new challenge in the literature of control
theory.
• It uses model reference that is designed depending on the
properties of the system.
• The use of the Most Valuable Player Algorithm (MVPA) to optimize
parameters of the controller and model reference.
• Applying the Variable Transformation Technique (VTT) to simplify
nonlinear systems. However, it is a new technique that did not
applied for many kinds of nonlinear systems.
6
Control Algorithm Design.5
The following is the control algorithm procedure:

Step 1: Develop the mathematical model of different kinds of
nonlinear systems.
Step 2: Use VTT to simplify the complicated systems if necessary.
Step 3: Select or design a proper model reference to specify the
desired performance that the nonlinear system should
achieve.
Step 4: Lyapunov stability analysis is used to analyze the
nonlinear systems.
Step 5: Select a proper cost function (performance index) for the
MVPA optimization method.
7
Step 6: Implement the MVPA optimization method to find the
optimal parameters of the constructed controller and the
reference model.
Step 7: Apply the proposed controller with optimal parameters
to the nonlinear systems (descriptive and physical nonlinear
systems) in order to compensate for the bad terms.
Step 8: Validate the control algorithm by applying it to different
kinds of nonlinear systems (real nonlinear systems (MagLev)).
Step 9: Compare the results with the original works of the
proposed system.
8
Study Cases.6
In this work, the control algorithm is applied to different

nonlinear systems with different problems such as:
1. Strict-Feedback Nonlinear Systems.
a) b)
Problems: 1- Nonlinearity 2- Instability 3- Bad performance 3-
Different Complex structures. 4- Excessive disturbances
(10%-30%).
9
2. Time-Varying Nonlinear System with External Disturbances.
Problems: a) Nonlinearity b) Instability c) Bad performance

d) Time-varying parameters c) External disturbances d)
Complicated structure.
3- Non-Affine Nonlinear System.
Problems: a) Nonlinearity b) Bad performance d) Time-varying

parameters c) External disturbances.
10
4- Control of Mass-Spring System with Uncertain Parameters.
Problems: a) Nonlinearity b) Uncertainty (multiplicative) c) Bad

performance.
5- Control of Single-Link Manipulator (SLM) System.
Problems: a) Nonlinearity b) Bad performance c) Complex

structure with lumped parameters.
11
6- Control of Heartbeat (HB) System.

d) Complex structure.
7- Control of COVID-19 System.

d) Complex structure with functional control action.
12
Validations.7
To validate the proposed control algorithm, It is applied to real

physical system represented by Magnetic Levitation (MagLev)
system as follow:
Problems: a) Nonlinearity b) Instability c) Non-affinity d) Bad

performance.
The control algorithm is applied to the MagLev system using the
previous procedure (Slide 5) which is the same for the other
cases as follow:
13
Step 1: Let the MagLev system be:
Or
Step 2: It is not necessary to apply VTT.
Step 3: Let the reference model be :
Step 4: Let the error vector and its time-derivative is defined as

follow:
14
By using Lyapunov stability analysis such as:
The time-derivative of Lyapunov function yields:
15
By substituting the system and the model reference in Lyapunov
time-derivative function gives:
16
Let the control equation be:
Then
And
Step 5: Select the Integral Square Error (ISE) as a cost function to the
optimization method (MVPA).
Step 6: Use the MVPA to optimize the parameters of the controller (
).
Step 7: Apply control equation ( ) to achieve the stability and
optimum performance of the MagLev system as shown in the next
slide.
17
The responses before and after applying the controller.
18
19
8.Conclusion
In this work, the proposed robust control algorithm has achieved

successively the stability and the performance different kinds
of nonlinear systems. In addition, It can be concluded that:
1. the Variable Transformation Technique (VTT) has
successively simplified different complicated structures of
nonlinear systems.
2. the Lyapunov stability analysis with infinite time-derivative
property has successively applied to the systems and proved
with the results.
20
3. Producing a controller based on the properties of the system
to be controlled gave a unique feature of this controller.
4. The use of the Most Valuable Player Algorithm (MVPA) to
obtain the optimal parameters of the proposed control
algorithm has simplified the design procedure and pushed to
get an optimal controller.
5. A model reference has been designed and optimized using
MVPA that depend on the properties of the nonlinear system
itself.
6. The proposed control algorithm has achieved a more desirable
performance despite the instability, bad performance,
nonlinearity, uncertainty, non-affinity, time-varying, and
disturbance.
21
List of Publications.9
1. Hadi, M, and Ali, H., 2020. Robust Control Algorithm Design for Nonlinear MIMO
Systems using MVPA Optimization. International Journal of Science and
Research. (Accepted).
2. Hadi, M, and Ali, H., 2020. An Applicable Optimal Controller Design for Nonlinear
Time-Delay Systems with Full State Constraints. International Journal of Science
and Research. (Accepted).
3. Hadi, M, and Ali, H., 2020. Optimal Nonlinear Controller Design for Different
Classes of Nonlinear Systems Using Black Hole Optimization Method. Arabian
Journal for Science and Engineering. (Accepted).
4. Hadi, M, and Ali, H., 2020. Optimal Model Reference Control Scheme Design for
Nonlinear Strict-Feedback Systems. Engineering and Technology Journal,
University of Technology. (Accepted).
5. Hadi, M, and Ali, H., 2020. Single Link Manipulator Trajectory Tracking using
Nonlinear Control Algorithm. Al-Nahrain Journal for Engineering Sciences.
(Accepted).
22
6. Hadi, M, and Ali, H., 2020. Control of COVID-19 System using A
Novel Nonlinear Robust Control Algorithm. Biomedical Signal
Processing and Control. (Under Review).
7. Hadi, M, and Ali, H., 2020. Optimal Tracking Controller Design for
Unknown Non-affine Nonlinear Systems with Unknown
Disturbance. Journal of the Franklin Institute. (Second Review).
8. Hadi, M, and Ali, H., 2020. Robust Control Algorithm Design for
Nonlinear MIMO Systems using Black Hole Optimization. Robotics
and Autonomous Systems. (Under Review).
9. Hadi, M, and Ali, H., 2020. Development of an Optimal and Robust
Nonlinear Controller for Magnetic Levitation System. Arabian
Journal for Science and Engineering. (Under Review).
10. Hadi, M, and Ali, H., 2020. Optimal Nonlinear Control Algorithm
Design for Heartbeat System. Arabian Journal for Science and
Engineering. (Under Review).
23
!Thank You
24

Robust Model Reference Controller Design For Different Classes of Nonlinear Systems

Uploaded by

Copyright:

Available Formats

Robust Model Reference Controller Design For Different Classes of Nonlinear Systems

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Robust Model Reference Controller Design For Different Classes of Nonlinear Systems

Uploaded by

Copyright:

Available Formats

Republic of Iraq

Ministry of Higher Education and Scientific Research

Robust Model Reference Controller Design for

By: Musadaq A. Hadi

Many physical systems are nonlinear to some degree, the

Nonlinear systems have come with many problems. These

The following is the control algorithm procedure:

In this work, the control algorithm is applied to different

Problems: a) Nonlinearity b) Instability c) Bad performance

Problems: a) Nonlinearity b) Bad performance d) Time-varying

Problems: a) Nonlinearity b) Uncertainty (multiplicative) c) Bad

5- Control of Single-Link Manipulator (SLM) System.

Problems: a) Nonlinearity b) Bad performance c) Complex

Problems: a) Nonlinearity b) Instability c) Bad performance

Problems: a) Nonlinearity b) Instability c) Bad performance

To validate the proposed control algorithm, It is applied to real

Problems: a) Nonlinearity b) Instability c) Non-affinity d) Bad

Step 4: Let the error vector and its time-derivative is defined as

The time-derivative of Lyapunov function yields:

In this work, the proposed robust control algorithm has achieved

You might also like