Holistic parameter optimization of fractional-order sliding mode control for switched reluctance motors using genetic algorithm

This paper presents an improved Fractional-Order Sliding Mode Control (FSMC) strategy for speed regulation of a 4-phase Switched Reluctance Machine (SRM). While conventional FSMC designs require manual or partial tuning of parameters—often focusing only on the fractional order α —this work identifies and addresses a critical research gap: the need for holistic optimization of all three key parameters ( α , λ , K ) to achieve optimal performance trade-offs. A multi-objective Genetic Algorithm (GA) is employed to simultaneously optimize this parameter triplet, minimizing both tracking error and control energy. The proposed Intelligent FSMC (IFSMC) is rigorously benchmarked against a conventional fixed-gain FSMC to demonstrate its superior performance. Simulation results show that the IFSMC achieves: (1) significant chattering reduction (control effort variance reduced by 75%), (2) improved dynamic response (settling time reduced by 51%), and (3) enhanced robustness to parameter variations, including ± 20% stator resistance changes and load torque disturbances with less than 0.8% speed deviation. This holistic optimization approach demonstrates that co-adaptation of all FSMC parameters is essential for unlocking the full potential of fractional-order sliding mode control in nonlinear electromechanical systems.