Date of Award


Publication Type

Master Thesis

Degree Name



Electrical and Computer Engineering

First Advisor

Kar, Narayan




Research conducted previously has shown that a battery electric vehicle (BEV) motor design incorporating drive-cycle optimization can lead to achievement of a higher torque density motor that consumes less energy over the drive-cycle in comparison to a conventionally designed motor. Such a motor indirectly extends the driving range of the BEV. Firstly, in this thesis, a vehicle dynamics model for a direct-drive machine and its associated vehicle parameters is implemented for the urban dynamometer driving schedule (UDDS) to derive loading data in terms of torque, speed, power, and energy. K-means clustering and Gaussian mixture modeling (GMM) are two clustering techniques used to reduce the number of machine operating points of the drive-cycle while preserving the characteristics of the entire cycle. These methods offer high computational efficiency and low computational time cost while optimizing an electric machine. Differential evolution (DE) is employed to optimize the baseline fractional slot concentrated winding (FSCW) surface permanent magnet synchronous machine (SPMSM). A computationally efficient finite element analysis (CEFEA) technique is developed to evaluate the machine at the representative drive-cycle points elicited from the clustering approaches. In addition, a steady-state thermal model is established to assess the electric motor temperature variation between optimization design candidates. In an alternative application, the drive-cycle cluster points are utilized for a computationally efficient drive-cycle system simulation that examines the effects of inverter time harmonics on motor performance. The motor is parameterized and modeled in a PSIM motor-inverter simulation that determines the current excitation harmonics that are injected into the machine during drive-cycle operation. These current excitations are inserted into the finite element analysis motor simulation for accurate analysis of the harmonic effects. The analysis summarizes the benefits of high-frequency devices such as gallium nitride (GaN) in comparison to insulated gate bipolar transistors (IGBT) in terms of torque ripple and motor efficiency on a drive-cycle.