Abstract
Principle of sequential excitation for stator windings of switched reluctance motor (SRM) causes many problems including torque ripples and substantial requirement for rotor position sensing. This paper presents an online torque sharing function (TSF) to minimize the undesirable torque ripples in SRM which operates under sensorless drive system. In classical TSF, variance between incoming and outgoing motor torque responses makes tracking of motor torque to reference torque cannot be achieved. Moreover, the TSFs are mainly designed for operating at rated speed. As motor speed changes, the torque ripples increase because the commutation period (CP) becomes inappropriate for commutation process. The proposed TSF is designed to guarantee the good matching between the motor torque response and TSF reference. So, a linear function is chosen as a torque reference for the incoming phase while an exponential function is selected for the outgoing phase. Additionally, the online tuning of commutation period is employed in the proposed hybrid TSF. The sliding mode observer (SMO) based model is formulated for rotor position and speed estimations purpose. The performance of the modified TSF is validated throughout computer simulations and experimental results. Furthermore, the comparisons between the proposed TSF and conventional TSF are carried out.