Logo: to the web site of Uppsala University

Electromobility solutions are central to the current decarbonization of transportation. So far, no other solution has contributed significantly to reducing carbon emissions other than adopting or using electric vehicles. However, a significant percentage of these new vehicles have been designed to use Rare-Earth Elements (REEs), mainly in Permanent Magnets (PMs) of their electric motor, but also in other components of their powertrain. Currently, REEs are classified by the European Union and other governing agencies as "critical raw materials" due to their fragile supply chain and their importance in strategic industries. Alternatives to REE-based PMs have been studied for some time now, but no alternative is widely used today in electromobility applications.

This thesis explores the use of ferrites PMs in electric machines; in particular, it studies the design and optimization of Spoke Type Permanent Magnets Synchronous Machines (Spoke) using this type of PMs. Studying in detail the implications of designing a Spoke machine with such magnets is important since they aim to substitute PMs with substantially different characteristics. Ferrite PMs have a much smaller remanence, making their pairing with a Spoke topology advantageous since it allows for higher levels of magnetic flux density on the rotor. Also, ferrite PMs present a much more significant risk of demagnetization than the REE-based PMs. Thus, designing and optimizing electric machines with the former implies testing previous design paradigms and proposing new reflections on the design and optimization process. For that, this thesis includes the development of an empirical demagnetization model for ferrite PMs, explores optimization goals and constraints enabled through the use of meta-modeling techniques, and applies the conclusions of these optimization studies to design and experimentally verify a Spoke machine with ferrite PMs.

The results of this thesis show that Spoke machines with ferrite PMs have the capability of achieving similar performance to the current electric machine solutions in electromobility applications. This is shown through the experimental results and the test of a Spoke machine prototype. The research also contributed to more detailed modeling of ferrite PMs, highlighting the current lack of detailed models for PM demagnetization on a macroscopic scale. The development of meta-models showed an acceleration in the optimization process and good agreement with both FEM results and experimental measurements. The results also highlight a new paradigm in the design of electric machines when using ferrite PMs; the demagnetization of ferrite PMs should not be explicitly minimized during the design process of these machines. Instead, the performance metrics after demagnetization should be maximized.