Abstract

There exists an immense demand for efficient, robust, reliable, and cost-effective motors for propulsion systems in hybrid electric vehicles (HEVs) and plug-in hybrid electric vehicles (PHEVs). Due to the absence of windings on the rotor and excellent extended speed features, much attention has been recently drawn towards switched reluctance machine (SRM) based vehicle propulsion system. Although these advantages are beneficial, the inherently large torque ripple and highly nonlinear characteristics of the SRM presents significant challenges from the controller design standpoint. Hence, appropriate design of a dedicated control scheme is critical, so as to enhance the overall performance. In order to achieve satisfactory control performance, the derivation of an accurate inductance model is essential. In this thesis, a "current saturation" approach is utilized, and the SRM magnetization characteristics are obtained from experimental results. Thereafter, the accuracy of the proposed method is discussed and verified by finite element analysis (FEA). In order to deal with SRM nonlinearities, this thesis introduces a finite approximation approach (Newton method) iteration computing algorithm. Based on the precise magnetic model, the optimal bus voltages are calculated, and are eventually stored in a meticulous look-up table. In order to engage high-speed switching control, a field programmable gate array (FPGA) is employed, and suitable simulation test results are analyzed in detail. It was found that the overall output torque ripple was minimized considerably, using the proposed novel FPGA based "voltage-profile" control approach.