Akademik Veri Yönetim Sistemi
IEEE Access, cilt.14, ss.45500-45516, 2026 (SCI-Expanded, Scopus)
This paper proposes a fault-tolerant control strategy for PMSM speed sensors that integrates a novel single-neuron PIDD controller, a backstepping controller, and an extended Kalman filter (EKF). The proposed system parameters are computed using the Zebra Optimization Algorithm (ZOA) to eliminate the need for trial-and-error. The main novelty lies in introducing the first single-neuron implementation of a PIDD structure and embedding it in a cascade with integral backstepping, while utilizing a filtered feedforward branch to enhance transient tracking without compromising closed-loop stability. Simulation results demonstrate that the proposed controller outperforms comparable methods in the literature without introducing chattering or oscillations under fault conditions. In addition, the suggested controller exhibits better performance under parameter fluctuations than other reported methods, demonstrating its simplicity and robustness. Finally, two independent real-time evaluations are conducted: (i) hardware-in-the-loop using Arduino and a PMSM model to confirm real-time feasibility, and (ii) a physical test on the Quanser Aero 2 platform to compare the proposed SNPIDD with its closely related counterpart, SNPID. The Aero 2 results show that SNPIDD achieves reductions of more than 10% in overshoot and settling time, providing robust disturbance rejection with smaller control effort, reduced oscillations, and at least 15% improvement in recovery time and maximum deviation compared to SNPID. Thus, SNPIDD enhances closed-loop stability and supports reliable operation in the presence of abrupt sensor outages and disturbances.