Application of Spatial Iterative Learning Control for Direct Torque Control of Switched Reluctance Motor Drive

In this paper, a novel direct torque controller for switched reluctance motor (SRM) is proposed using spatial iterative learning control (ILC). SRM magnetization characteristics are highly non-linear, and torque is a complex and coupled function of phase current and rotor position. Direct torque control (DTC) scheme avoids the complexity of torque-to-current conversion as required in indirect torque control scheme. Traditional DTC scheme uses a hysteresis controller and leads to large amount of torque ripples when implemented using a digital controller. Advanced non-linear control methods can be used to improve the performance of DTC in SRM. However, such methods are often too complex for real-time implementation or require an accurate model of SRM magnetization characteristics. As shown here, ILC only uses a linearized magnetization characteristics and a simple learning law to obtain the desired control signal. An ILC based DTC scheme for SRM torque control for constant motor torque, has been developed and experimentally verified on a 1-hp, 4-phase SRM. Experimental results show the effectiveness of the proposed scheme in terms of average torque control and ripple minimization.