Public defence in Power Electronics and Electric Drives, M.Sc. Reza Hosseinzadeh

The title of the thesis: Modeling and Control of Bearingless Linear Motors

Thesis defender: Reza Hosseinzadeh
Opponent: Prof. Rafal P. Jastrzebski, University of Turku
Custos: Prof. Marko Hinkkanen, Aalto University School of Electrical Engineering, Department of Electrical Engineering and Automation

Electric motors have been increasingly employed in many applications since their invention in the 19th century. Today, electric motors can be found in various equipment ranging from the simplest home appliances to the most complicated industrial machinery. One of the main components of the motors is the mechanical bearings. They play a crucial role in supporting the moving parts and reducing friction. However, there are some downsides to them, such as the associated wear and tear and the need for regular maintenance and lubrication. In addition, they can generate noise and vibration caused by friction. 

This is where bearingless electric motors come into play. These motor types use magnetic forces to levitate the moving parts, which renders the mechanical bearings redundant. By levitating the moving part, they reduce the wear and tear due to friction, eliminate the need for regular lubrication and maintenance, and lead the way to a smoother and more silent operation. Due to these advantages, they can be employed in applications requiring high speeds, as well as in fields where oil contamination poses challenges, such as the pharmaceutical industry. 

Despite these advantages, the control of bearingless motors is significantly more challenging than those with mechanical bearings. Anyone who has ever used a permanent magnet (PM) knows that the closer a magnet is to a ferromagnetic material, such as steel, the stronger the attraction force becomes. In simple terms, this nonlinear phenomenon is the underlying cause of the challenges in the control of these motor types. This thesis considers model-based control methods for the control of linear bearingless motors. This control design methodology requires an accurate dynamic model of the system. 

In this thesis, a dynamic model is presented for a bearingless linear flux-switching permanent-magnet motor. This dynamic model includes effects such as magnetic saturation and airgap variation. In addition, a dynamic model is also developed for six-degree-of-freedom bearingless linear motor systems. The developed models can be utilized in time-domain simulations for system analysis and real-time control system development. Furthermore, the control of bearingless motors has been covered. For a double-sided bearingless linear motor system, a state-feedback control method is presented. To enhance energy efficiency, a resistive-loss minimization algorithm is also proposed.

Keywords: Linear motor, magnetic levitation, magnetic model, modeling, bearingless, flux-switching permanent-magnet machine, unbalanced magnetic pull, linear actuator, lumped element, artificial neural networks, energy efficiency

Thesis available for public display 10 days prior to the defence at: https://aaltodoc.aalto.fi/doc_public/eonly/riiputus/

Contact:

Doctoral theses in the School of Electrical Engineering: https://aaltodoc.aalto.fi/handle/123456789/53