A Unified Bézier Curve-Based Flux-Linkage Model for Position Sensorless Model Predictive Current Control of SRMs

Precise flux-linkage characterization plays a crucial role in current and torque control strategies of switched reluctance motors, which are becoming increasingly common in industrial and automotive applications. Moreover, due to the challenges posed by physical rotor position sensors, there is a constant demand for a cost-effective position estimation method that can be easily implemented on a digital computation platform. This work presents a method for modelling and control of a position sensorless SRM drive, using a technique considering a unified Bézier curve approach that reduces memory requirements typically associated with large lookup tables used in conventional control techniques. The inherent simplicity of this technique facilitates its application in precise position estimation and model predictive current control (MPCC), and ensures that it can be implemented on low-cost digital processors with low computational burden. The presented method is experimentally tested on a 2.2 kW SRM, demonstrating its efficacy in position sensorless current control. The performance of the presented method in tracking current reference is found to be superior to that of conventionally used current controllers without imposing high memory requirements or computational burden.