Events at Department of Energy Technology

Industrial/PhD Course: Model Predictive Control of Power Electronic Converters

Time

31.01.2018 kl. 08.30 - 02.02.2018 kl. 16.00

Description

Organizer

Associate Professor Tomislav Dragičević

Lecturers

Associate Professor Tomislav Dragičević, Department of Energy Technology, mail: tdr@et.aau.dk
Associate Professor Cristian Lascu, Department of Energy Technology, cla@et.aau.dk
Associate Professor Ulrik Nyman, Department of Computer Science, mail: ulrik@cs.aau.dk
Assistant Professor Yongheng Yang, Department of Energy Technology, mail: yoy@et.aau.dk

ECTS: 3
 

Description

Model Predictive Control (MPC) is a conceptually simple yet powerful methodology to control power converters and electric drives. It has many advantages over traditional linear controllers including (i) faster response, (ii) high robustness to parameter variation (iii) explicit multivariable control accounting for the process and actuator constraints. The advances in processing power of digital signal processors have recently promoted MPC into the first commercial applications, which opened a door towards improved performance and efficiency of power electronic converters and drives demanded by the evolving industry applications. This course aims to provide the fundamentals required to understand, design and implement MPC to power electronic converters used for a variety of applications including grid-connected converters, drives and microgrids. The motivation is to facilitate wider and faster exploitation of MPC by bridging the gap between theory and successful industrial implementation through cooperation and exchange of experience between academic/research and industrial communities. It is envisioned for both PhD students and practicing engineers.

Some of the course contents are based on recently obtained research results.

The main topics are as follows:

  • Introduction to Model Predictive Control (MPC) for Power Electronic Systems and Drives
  • Power Converter Modelling Fundamentals and Discretization
  • Finite-Control-Set Model Predictive Control  (FCS-MPC) Principle
  • Periodic and Dead-Beat Control Principles
  • Quantitative Performance Evaluation of the FCS-MPC
  • Application Example: FCS-MPC in 2-level, 3 phase Voltage Source Converter for AC microgrids
  • Laboratory Exercises (Simulation + Laboratory Demonstration)

Prerequisites

General knowledge in electrical AC circuits and electrical power engineering, preferably background at the graduate level in power electronics. Matlab/Simulink knowledge is recommended for the exercises.

Form of evaluation

Individual evaluation of the course participant will be performed on a basis of:

Attendance rate (5%)
Mini-project (95%)

Course lecturers will design three mini-projects for the Ph.D. course. Each student will be assigned with or select a specific mini-project within the lectured topics, where the students should model the system, design the controllers, and perform simulations. Students are required to finalize the mini-projects within three weeks after the course by submitting a formal technical report with simulation results, which will be assessed by the lecturers in two weeks.

Registration

https://phd.moodle.aau.dk

Price

6000 DKK for PhD students outside of Denmark and 1500 DKK for PhD students in Denmark, who are not from AAU. 8000 DKK for the Industry.

Host

Department of Energy Technology

Address

Aalborg University, Pontoppidanstræde 101, room 19, 9220 Aalborg East