Smart Neural-Fuzzy Control of Brushless DC Motors for Reduced Torque Ripple and Harmonic Distortion

BLDC motors require accurate control under varying conditions, while conventional controllers suffer from nonlinearities and disturbances that increase overshoot, torque ripple, and harmonic distortion. Conventional methods such as Field-Oriented Control (FOC), Direct Torque Control (DTC), and PI/PID controllers are widely used but remain sensitive to nonlinearities and disturbances, resulting in overshoot, torque ripple, and Total Harmonic Distortion (THD). Recent advances in Artificial Intelligence (AI) have enabled significant improvements in BLDC motor control. To address these challenges, this paper proposes a hybrid artificial neural network (ANN)-PI coupled with a Fuzzy-PID controller that combines the learning capability of artificial neural networks with the robustness of fuzzy logic. The main challenges addressed in this study are the need for adaptive parameter adjustment, improved disturbance rejection, and simultaneous optimization of speed tracking accuracy, torque ripple, and harmonic distortion in BLDC drives. The proposed hybrid strategy aims to overcome the limitations of conventional controllers by combining ANN learning capability with fuzzy robustness. Although not based on a conventional adaptive control framework, the proposed ANN-PI/Fuzzy-PID controller exhibits adaptive behavior by continuously updating control parameters in response to system nonlinearities and disturbances. All results presented in this work are obtained from detailed MATLAB/Simulink simulations. The results demonstrate that the hybrid controller achieves faster transient response (0.02 s), minimal overshoot (~2.5%), low steady-state error (~0.00012), and significantly reduced THD (1.85%) and torque ripple (0.05 N.m/A), outperforming conventional PI and MPC strategies. Additional comprehensive comparison with recent publications further confirms the competitiveness and robustness of the proposed approach.