Author ORCID Identifier

https://orcid.org/0009-0001-6419-1366

Date Available

9-26-2025

Year of Publication

2025

Document Type

Doctoral Dissertation

Degree Name

Doctor of Philosophy (PhD)

College

Engineering

Department/School/Program

Electrical Engineering and Computer Science

Faculty

Dr. Dan M. Ionel

Abstract

The design and optimization of electric machines face increasing demands for higher efficiency, improved torque density, manufacturability, and effective utilization of materials, particularly in applications such as electric traction and propulsion, where performance, reliability, and cost are critical. This work explores innovative synchronous machine configurations with reluctance rotors and stator-combined excitation, including permanent magnet and DC-excited topologies, to address these challenges through advanced design methodologies, computational modeling, and experimental validation. This research is relevant to address the growing demand for high-performance electric machines that combine high power density with costeffective manufacturing. The conventional limitations in power and torque density, thermal management, and material utilization, particularly when rare-earth magnets are involved, are addressed using innovative topologies, employing advanced techniques for design optimization, as well as comparative studies. Furthermore, the nonlinear behavior and complex saturation effects prevalent in such motor topologies, which greatly affect the performance, are acknowledged and investigated as a potential bottleneck for obtaining optimal designs. The first part of this research, starting in Chapter 2, focuses on permanent magnet (PM) stator-combined excitation topologies, comparing inner and outer rotor configurations. These machines employ toroidal windings for improved slot fill and shortened end turns, while leveraging spoke-type PM arrangements for flux intensification. A detailed examination of their electromagnetic performance, manufacturability, and cooling potential is conducted, with particular attention to the trade-offs between cost, power density, and losses. The study highlights how stator-combined excitation designs enable simplified rotor structures while maintaining competitive torque output, even with non-rare-earth magnets. Chapter 3 explores DC stator-combined excitation configurations as an alternative to PM-based machines, eliminating demagnetization risks while retaining high flux concentration. A novel serpentine winding design is introduced to reduce losses while maintaining the possibility of high power density. These PM-free designs also have several degrees of freedom for motor control, fault-tolerant performance, reduced material cost, as well as the potential for high power density subject to cooling, making them an attractive option for traction and propulsion applications.

Chapter 4 establishes a general theory for stator-combined excitation machines, analyzing the interaction between stator fields and rotor geometry. Generally applicable formulations for Magnetomotive Force (MMF) analysis and airgap flux density are investigated along with explorations of the inductance characteristics that lead to non-salient behavior despite the rotor’s geometric saliency. A systematic approach to establishing the synchronous nature of this machine is also presented. Experimental results validate the analytical and computational models, providing insights into the machine’s nonlinearities, flux-weakening capabilities, and potential for high efficiency operation. Finally, in Chapter 5, a novel hybrid FEA and Machine Learning (ML) for Differential Evolution (DE) optimization of these highly non-linear motor topologies is presented. This final part introduces a machine learning-assisted optimization framework to address the challenges of designing highly nonlinear electric machines. By combining DE with Artificial Neural Network (ANN)-based meta-models, the method significantly reduces computational effort while maintaining accuracy in performance prediction. This approach enables exploration of complex design spaces through nonlinear performance scaling and could potentially facilitate system-level optimization, including, for example, drive cycle analysis for electric vehicle applications. Together, these studies seek to advance the understanding and development of synchronous machines with excitation through stator DC coils or magnets and reluctance rotors, offering solutions for high-performance applications while addressing cost, manufacturability, and material sustainability concerns. The integration of these novel topologies, computationally efficient methods, and experimental validation provides a comprehensive base for futuristic motor design optimization and analysis in the ever-growing electrification of the world. These advancements contribute to the development of high-performance, cost-effective electric machines for transportation and industrial applications, addressing critical needs in material sustainability and manufacturing scalability.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2025.460

Funding Information

–DOE VTO award no. #DEEE000887 (2018 – 2023) –NASA ULI project #80NSSC22M0068 (2023 - 2027) –Other industrial projects sponsored by QM Power, Inc., and supported by Ansys Inc.

Download

Share

COinS