A novel strategy for obtaining nonlinear flux linkage model of switched reluctance motor

Abstract

This paper proposes a novel strategy for obtaining nonlinear flux linkage model of switched reluctance motor (SRM) without rotor locking devices. The strategy involves two parts, namely, measurement and reconstruction. In the measuring part, the flux linkage curves of aligned and unaligned positions are obtained firstly. For the middle positions, the flux linkage curves are measured under current chopper control (CCC) mode, which avoids the magnetic coupling effect. In the reconstructing part, a back-propagation neural network (BPNN) is adopted to obtain the entire flux linkage characteristics with the measured samples. Finally, in order to verify the accuracy and feasibility of the proposed strategy, the flux linkage characteristics obtained by the proposed method are compared with those measured by rotor clamping method. The result shows that a good agreement can be obtained by using the proposed strategy with the experimental samples.