An Inductance Characteristics-Based Position-Sensorless Startup Method for Doubly Salient Electromagnetic Machine

In this article, a high-performance low-speed sensorless startup method is proposed based on the inductance characteristics of doubly salient electromagnetic machine (DSEM). First, by establishing the nonlinear inductance model related to the rotor position, a position estimation method is developed by capturing the mutual inductance information between the armature and field windings from terminal voltages. Then, for the inductance nonlinearity caused by the doubly salient structure of the motor, a compensation method is proposed to eliminate the position estimation error caused by high-order harmonics of inductance during standstill and startup. Moreover, for the initial rotor position estimation, an accurate convergence method is proposed by analyzing the convergence of the initial estimation. For the position estimation during the startup process, the estimation error caused by polarity changes of field current slope is analyzed, and a corresponding compensation strategy is presented. Additionally, the compensation method is investigated for the commutation error. Compared with existing methods, the proposed method avoids the injection of test pulses and considers the effect of inductance nonlinearity, which can improve the startup performance on the basis of enhancing the estimation accuracy of position angle. Finally, the experiments on a 12/8-pole DSEM validate the correctness of the proposed method.