Bearingless Induction and Permanent Magnet Machine Test Bench

Additively Manufactured Motor Housing with Integrated Cooling Channels

3 Pole AMB Stator

Electrostatic levitation testing

SiC Power Pole

AMDC

AC Homopolar Motor

Bearingless Motor Stator

Combined Winding

Bearingless Machine Testing

Research

Severson Group Research

We are creating new electromechanical systems to realize a smarter, more sustainable future. My team investigates the science needed to design and actuate these machines to their full potential.

Our core research focus areas include the following:

  • Design and control of electric machines and power electronics
  • Bearingless motors and magnetic bearings
  • Off-highway vehicle electrification and flywheel energy storage

Our research is inherently multidisciplinary and relies on finite element modeling (electromagnetic, structural, thermal), circuit simulation, design optimization, control theory, and experimentation. We have a particular soft spot for magnetic levitation technology.

Community Resources

We have created several tools to design and actuate our electromechanical systems. We are sharing these tools as open source resources via GitHub.  These tools include:

AMDC: A control platform to operate advanced electric machines dubbed the “Advanced Motor Drive Controller”. This specialized platform combines a large number of configurable user I/O oriented toward electric machinery with the Xilinx Zynq SoC, an open source and extensible firmware architecture, and practical model based control examples to create a versatile sandbox for developing extreme performance electric machine control systems.

AMDS: A sensing platform for feedback control of electric machines. The “Advanced Motor Drive Sensing System” can be used directly with the AMDC or with a third party control solution. This project is fully open-source, including the hardware design, firmware, and a how-to guide. To get started with the AMDS, refer to the AMDC documentation website articles, available here.

eMach: An open-source electric machine modeling, evaluation, and optimization framework developed in Python. The codebase both interfaces to professional FEA tools and includes rapid analytic multi-physics models of electric machinery.