Synchronous reluctance motors (SynRMs) have gained renewed interest as economical, mag-net-free alternatives to permanent magnet and induction machines, particularly for small- and medium-power electric vehicle propulsion systems where cost, efficiency, and sustainability are critical. Their main advantages include simple construction, high reliability, and suitability for sustainable drive systems. However, SynRMs inherently suffer from torque ripple, acoustic noise, and limited power factor, mainly due to the anisotropic nature of their magnetic design. This study presents a fully integrated Multi-Objective Genetic Algorithm (MOGA)–Finite Ele-ment Method (FEM) framework, which simultaneously minimizes torque ripple and maximizes output power by systematically optimizing rotor flux barrier geometry, in contrast to traditional trial-and-error or single-objective optimization methods. The optimization process evaluates key rotor design parameters, including flux barrier angles, widths, and insulation ratios, while explor-ing a wide design space and corresponding objective function values. FEM analysis is employed at each optimization iteration to accurately assess electromagnetic performance and ensure con-vergence toward the optimal solution. The optimized rotor demonstrates a substantial perfor-mance improvement, with the average torque increasing from 13.15 Nm to 19 Nm and the output power rising from 4.13 kW to 6 kW. The power factor is improved from 0.85 to 0.9, indicating enhanced efficiency and reduced reactive current demand. Moreover, torque ripple is quantita-tively reduced from 8.95% to 8.5%, confirming the effectiveness of the proposed optimization strategy. The introduction of barrier fillets further decreases flux density in the tangential ribs, thereby reducing mechanical stress and core losses. These results confirm that systematic opti-mization of flux barrier geometry can significantly enhance SynRM performance without the use of permanent magnets or complex rotor structures, preserving simplicity and cost-effectiveness.
Boudjelida et al. (Sun,) studied this question.