OPPONENT REVIEW OF THE Ph.D. THESIS
Branch of study: Electrical machines, apparatus and drives
Title: Applications of Particle Swarm Optimization Algorithms in Control of Induction Motor Drives
Author: Ing. Ho Dang Sang
Supervisor: Prof. Ing. Pavel Brandštetter, CSc.
Oponent: Doc. Ing. Petr Palacky, Ph.D..
Department of Electronics
Faculty of Electrical Engineering and Computer Science VSB- Technical University of Ostrava
17. listopadu 15/2172, 708 33 Ostrava-Poruba Tel. +420 597 325 930, e-mail: petr.palacky@vsb.cz
The presented PhD thesis deals within the range of 90 pages with the problematics of application of PSO optimization of induction motor control with respect to energy optimization and on-line identification of stator resistance.
Achieved results are documented by the means of simulations and also experimentally in the form of waveforms of important physical quantities.
The content of the thesis is divided into six chapters followed by references, author's publications and projects.
According to recommendations for preparation of the opponents' opinions, I evaluate this work from the following viewpoints:
a) Comments on relevance to the practice and development of the science branch
The topic of the PhD thesis is very present as it is focused on control algorithms for the field of modern control methods of induction motors.
The sensors are difficult to mount in harsh environment and high speed drives. Hence, it is very necessary to have suitable schemes that don’t use these sensors. The sensorless control methods of the IM drives get a high attention because of their basic advantages such as reduction in hardware complexity and cost, increasing mechanical robustness, higher reliability, lower maintenance demands and cost, a capability of working in hostile environments, etc.. However, sensorless techniques require precise knowledge of motor parameters that change during operation. Therefore, their constant identification is necessary.
For these reasons, the importance of such research in the field of electric drive control is growing significantly and is absolutely necessary in the field of electrical machines, devices and drives.
b Evaluation how the dissertation fulfilled the stated goals
In my opinion, the chosen solution procedure fully complies with both the general practices and mentioned objectives in the Chapter 1.2.
The actual solution is contained in chapters 4 and 5. The work is logically divided from the introduction to control methods and mathematical models of individual parts of the drive, to the concept of using the PSO algorithm for control and finally to its integration into specific solutions for sensor and sensorless control.
Designed and implemented solutions are subsequently verified experimentally. This also fulfills the goals of the PhD thesis.
c) Comments on problem solving procedure. Comments on the results of the dissertation work with the concrete contribution of the thesis author
The work focuses mainly on the design of Particle Swarm Algorithm structures and its applications in the induction motor control. The introductory part of the thesis describes the mathematical model of induction motor which is followed by a very detailed description of the method of vector control of AC machines and mathematical modeling of induction motor with iron loss consideration.
The author further focuses on the theoretical analysis of the particle swarm optimization and describes the principle of PSO algorithm.
In the following chapters, this algorithm was used to optimize the energy balance of the induction motor control and also to identify the stator resistance online.
I consider this part to be the main contribution of the author to scientific field.
Proposed methods are subsequently simulated using the Matlab Simulink environment, where the PI and PSO Stator Resistance Estimation methods are compared. The waveforms of important physical quantities from experimental verification are documented in the next chapter.
Comments to the dissertation and questions to the author are attached in the appendix. This does not reduce the level of this work in any way and can be used during the defense discussion.
d) Comments on formal level of the thesis
The dissertation has a very good formal level. Individual chapters of the thesis create logical continuity of solved problems, enabling a good understanding of the given problem. There are very few misprints or grammatical mistakes, graphic processing also has a high level.
e) Comments on dissertation's publications
A list of publications of the author is available on pages 88 and 89. It consists of 16 titles in total, 10 of which are indexed in the Web of Science or Scopus database. He also participated in SGC projects. The main part of the work has been published sufficiently.
f) Conclusion
The PhD thesis contains a number of new insights to the sensorless control of AC machines using evolutionary algorithms.
Due to high professional level, the significant contribution to the field and also to the precise processing, the PhD thesis fulfils the requirements of the law.
I recommend this PhD thesis for public presentation, after the successful defense I also recommend to award the PhD degree to Mr. Ing. Ho Dang Sang.
In Ostrava, 5. 9. 2020
doc. Ing. Petr Palacký, Ph.D.
Questions to discuss in the defense:
1) The list of abbreviations and symbols does not contain all abbreviations used in the text.
2) What is the difference between the symbols R and r (Eq 4.10, 4.12, 4.13)?
3) Could the author explain in detail how equation (4.32) can be obtained?
4) Can you explain in detail the PSO algorithm used in Fig. 4-29? What equations (3.1) – (3.5) will look like for the stator resistance estimation?
5) How the actual motor torque was measured or obtained?
6) Can you explain the differnece between the reference and actual motor torque (Fig. 5-14, Fig. 5-22 etc.)?
