Efficiency optimized control of switched reluctance machines

Kojima, Tetsuya; de Doncker, Rik W. (Thesis advisor); Hameyer, Kay (Thesis advisor)

Aachen : buchdruck.de (2018)
Book, Dissertation / PhD Thesis

In: Aachener Beiträge des ISEA ; Band 109 109
Page(s)/Article-Nr.: 1 Online-Ressource (x, 142 Seiten) : Illustrationen, Diagramme

Dissertation, RWTH Aachen University, 2017

Abstract

This thesis studies efficiency optimization to minimize the losses of switched reluctance machines for average torque control (ATC) and PWM based predictive direct instantaneous torque control (PWM-DITC). Generally, ATC is used at medium to high speed to achieve high efficiency, and PWM-DITC is used at low speed to reduce torque ripple. For ATC, conventional loss minimization methods cannot utilize zero-volt loops (ZVLs) that are additionally inserted between the turn-on and turn-off angles to improve the efficiency. In this work, a fast optimization method is proposed. The proposed method calculates the current reference for a torque reference. This current calculation method removes a time-consuming nonlinear constraint regarding a torque reference, and thus accelerates the optimization. Experimental results demonstrate that the iron loss is largely reduced by exploiting ZVLs and the loss reduction is especially large at medium speed and partial load. The loss reduction is also investigated for a city driving cycle. For PWM-DITC, this thesis proposes an optimal torque sharing function (OTSF) to minimize the copper loss through approximating the corresponding optimal torque waveform. While dividing a period of torque sharing into three sections, the OTSF uses a quadratic, a linear, or another quadratic function in one section. Consequently, the OTSF reduces the on-line calculation time compared to conventional TSFs as well as achieving more accurate approximation. These characteristics of the OTSF are verified by experiment.