Optimized Discrete-Time LQI Current Control for Indirect Torque Regulation of Switched Reluctance Motors

Torque ripple remains a key challenge in switched reluctance motors (SRMs) due to their highly nonlinear magnetic characteristics. In order to achieve improved torque smoothness, this paper proposes an enhanced discrete-time current controller based on Linear Quadratic Integral (LQI) regulation, combined with an offline Torque Sharing Function (TSF) to enable indirect torque control. A performance-oriented optimization methodology is introduced, where the weighting matrices of the LQI formulation are tuned using a Genetic Algorithm (GA) to balance current tracking accuracy and control effort. The proposed controller consistently maintains low root- mean-square errors in both current and torque across a wide range of speeds and load conditions, while ensuring torque smoothness. Such performance is sustained throughout all operating points, despite the highly nonlinear and strongly speed/load/position-dependent behavior of the SRM. The strategy also presents high robustness to parametric uncertainty, maintaining closed-loop stability under inductance variations of up to$\pm$50% relative to the design model. Furthermore, its reduced computational burden makes it suitable for lower-cost processors or multi-drive applications. Experimental results validate the effectiveness of the proposed strategy, demonstrating reduced torque ripple and current-tracking error under both steady-state and dynamic conditions.