The Induction machine as speed variable drive for automotive traction applications

A detailed loss analysis is indispensable for the design and improvement of high-power-density, highly-efficient traction machines in electric vehicle applications. The losses inside these machines are highly location- and operating-point-dependent. In proximity of the air gap, the iron-loss density is much higher compared to locations in larger distance, such as in the stator or rotor yoke. Due to the skin and proximity effect, the Ohmic losses are not evenly distributed over the winding and the rotor cage. In this thesis, a simulation model which calculates the local loss density for various torque-speed operating points from the solutions of a 2D time-transient finite element simulation is developed. The model combines analytical, static and transient finite element formulations for efficient modeling. The iron-loss model is parameterized by standardized soft magnetic ring core measurements. The total losses are determined by integrating over the local loss density. An algorithm for scaling the losses to different rotational speeds is evolved. The loss-minimal operating points are determined subject to the voltage and current constraints. A single building factor is used in the simulation model. In extensive test-bench measurements, this building factor was determined to 0.93, which is close to the ideal value of 1. Thus, the model incorporates the relevant local loss effects and is applied to improve the machine design. The design parameters studied are the electrical steel grade, the air-gap length, the number of poles, the shapes of the slot openings, and others. The model enables the machine designer to thoroughly understand the local loss effects and the impact of design choices on the local loss distribution and the total losses.