In order to improve the operation performance and decrease the torque ripple of switched reluctance motor (SRM), a new current control method for SRM based on voltage space vector was proposed. In which the stator windings are excited by sinusoidal current with a DC offset to realise the unipolar current excitation. In this control system, zero vectors were used to control zero-sequence current alone and non-zero voltage space vector was selected to control d-q-axis current considering coupling. The reluctance torque component was offset by injecting third harmonic current into the DC bias current, and copper loss was minimised by selecting appropriate current ratio. Experiences verified that the new method can effectively control SRM current. The torque ripple can be reduced with third times harmonic current injection, but the efficiency is also decreased. Compared with the traditional control method, the proposed method is simple, easy to implement, and can effectively suppress torque ripple.

To reduce the effect of estimation position angle error caused by mutual inductance on sensorless control of switched reluctance motor (SRM) in clear command channel mode, a SRM rotor position estimation method is presented. Using this method, the effect of mutual inductance can be eliminated without measuring mutual inductance. First, different working states are analysed. When working at single-phase excitation, mutual inductance can be ignored, inductance model method is used to estimate the position angle. When working at two-phase excitation situation, mutual inductance is non-zero, based on flux linkage-difference a new method is used to eliminate mutual inductance influence. Finally, the proposed method is verified by experiments. Experiment results show that compared with the mutual inductance ignored methods, the proposed method has higher estimation accuracy. This method can realise stable and reliable running of SRM sensorless control in wider speed range.

No abstract

The phase inductance of a switch reluctance motor (SRM) is significantly nonlinear. With different saturation conditions, the phase inductance shape is clearly changed. This study focuses on the relationship between coefficient and current in an inductance model with ignored harmonics above the order of 3. A position estimation method based on the variable coefficient inductance model is proposed in this paper. A four-quadrant sensorless control system of the SRM drive is constructed based on the relationship between variable coefficient inductance and rotor position. The proposed algorithms are implemented in an experimental SRM test setup. Experimental results show that the proposed method estimates position accurately in operating two/four-quadrants. The entire system also has good static and dynamic performance.

To reduce torque ripple of switched reluctance motor (SRM), a low-torque ripple control method based on current vector is proposed in this study. In which, i d = 0 control strategy is adopted. Then, i q is decomposed into the vector sum of twophase currents in different positions. A Fourier series model about T e and i q is established and this model satisfies the magnetic saturation condition. According to the established Fourier series model, the change rule of i q based on constant reference torque is obtained. In this study, torque models of SRM under traditional unipolar sinusoidal excitation control method, constant i q unipolar excitation control method and variable i q unipolar excitation control method are analysed. Finally, experimental platform of SRM control system based on TMS320F2812 is built. Experimental results show that the variable i q unipolar excitation method based on current vector decomposition can effectively reduce the torque ripple of SRM.