Adaptive Fuzzy Logic-Based SMO Utilizing Higher Order Soft-Sign Function With ROA Tuned PI Gains for Sensorless PMSM Drive

This article presents an advanced sensorless control strategy for permanent magnet synchronous motors (PMSMs) that integrates optimal proportional integral (PI) controller tuning via the remora optimization algorithm (ROA) with robust speed estimation using a fuzzy-adaptive higher-order soft-sign sliding mode observer (SMO). The ROA, a population-based optimizer characterized by low computational complexity and fast convergence, is employed in an offline optimization mode to tune the PI gains for the speed, direct-axis ($\boldsymbol{i}_{\boldsymbol{d}}$), and quadrature-axis ($\boldsymbol{i}_{\boldsymbol{q}}$) current loops within a field-oriented control (FOC) framework. Compared to particle swarm optimization (PSO) and genetic algorithm (GA)-based tuning, the ROA achieves approximately 50% faster convergence with fewer control parameters and stable global optimization behavior. The proposed SMO incorporates a higher order soft-sign switching function whose parameters are adaptively tuned by a fuzzy logic controller (FLC) using Gaussian bell shaped membership functions, enabling real time adjustment of the function’s slope and smoothness. Experimental validation on a TMS320F28379D DSP-based PMSM drive demonstrates a maximum steady state speed ripple of$\boldsymbol{\pm}$0.5 rpm, a peak overshoot of 2.5 rpm, a settling time of 3.85 s within the 300–1500 rpm range, and approximately 14.2% improvement in settling time compared to the sigmoid-based observers approach. The integrated ROA–FLC–SMO framework thus delivers improved robustness, adaptability, and computational efficiency for real-time PMSM drives.