Enhancing Performance Toward Torque and Flux Control Through a Hybrid Approach of Intelligent and DTC for SRM Drives

Switched Reluctance Motors (SRMs) are considered the preferred choice for next-generation electric vehicle (EV) traction motor applications due to their simple rotor structure and favorable power and torque characteristics. However, SRM drives suffer from high torque ripples, which can cause severe vibrations and acoustic noise. Moreover, the highly nonlinear characteristics of SRMs complicate their control and modeling, making it a significant challenge to achieve optimal performance. Therefore, this paper proposes a Direct Torque Control (DTC) strategy for SRMs, which incorporates a modified dynamic precise model and an Adaptive Neuro-Fuzzy Inference System (ANFIS). First, a DTC technique based on a two-level hysteresis controller with an appropriate switching scheme is introduced for instantaneous torque control, ensuring proper torque tracking with a smooth torque profile. Second, a novel hybrid intelligent ANFIS model is developed for accurate flux and torque estimation and optimal selection of switching angles. The proposed method employs a multi-objective function to fully exploit the torque production capability of the SRM, minimize torque ripple, and enhance operational efficiency. The proposed solutions are compared with conventional DTC schemes, and experimental evaluations demonstrate significant reductions in torque ripple, flux ripple, and current distortion.