FEM Analysis Applied to Electric Machines for Electric Vehicles

Detta är en avhandling från Uppsala : Acta Universitatis Uppsaliensis

Sammanfattning: Electric vehicle technology is an interdisciplinary field in continuous development. It appears to be a margin for improvements. The Division for Electricity at Uppsala University is doing significant research in the field. The present thesis investigates electric machines for vehicular applications, both in the driveline and in the traction motor.Section 1 presents a driveline with two galvanically isolated voltage levels. A low power side is operated at the optimum voltage of the batteries, while a high power side is operated at a higher voltage leading to higher efficiencies in the traction motor. Both sides are coupled through a flywheel that stabilizes the power transients inherent to a drive cycle.A review of electric machine topologies for electric vehicles is presented in Section 2. The permanent magnet excited machine is the most suitable technology for an electric driveline.Section 3 is devoted to numerical models applied to electric machines. The equivalent circuit of a motor/generator with two sets of windings is first presented. This machine couples both sides of the driveline and drives the rotor of the flywheel. The electric parameters are calculated with custom FEM models. A discussion on slotless machines concludes with a simple model to analyze the magnetic field from one static 3D simulation. The tooth ripple losses in solid salient poles are also analyzed with a novel FEM approach. A complete description of the losses in electric machines gives a proper background for further discussion on efficiency.Section 4 presents the experimental work constructed to validate the theoretical models. The experiments include an axial flux, single wounded prototype, an axial flux, double wound prototype and a planed radial flux coreless prototype.Section 5 focuses on traction motors for electric vehicles. A simulated prototype illustrates a design and calculation process. The loss theory and the numerical methods presented in Section 3 are applied.