Electric Vehicle (EV) technology has been gaining attention these last few years, with huge possibilities of reducing the environmental burden in comparison with fossil fuel-powered vehicles. This has made the comparison of different types of motors for traction applications an interesting topic. Among these, brushless DC (BLDC) motors have superior torque density, near-unity power factor, and minimal maintenance. When there are a growing interest and demand for these specialized BLDC motors for numerous applications, to develop design methodologies and optimize motor performance. This paper proposes an algorithm called Particle Swarm Optimization with Self-Adaptive Velocity Update Strategy (PSOSAV) and its application to the BLDC motor design problems. The self-adaptive mechanism adjusts the trade-off between explorative and exploitative behaviors in the right amount throughout the entire process of optimization, ensuring better motor performance. The primary performance criteria targeted by the proposed algorithm are Total Harmonic Distortion (THD), torque, and back electromotive force (back-EMF), which have been optimized to improve motor efficiency and reliability. The PSOSAV algorithm achieved optimized parameters of Br = 1.3481 T, g = 7.5×10⁻⁴ m, and α = 0.7663, resulting in a torque constant of 0.11408 and a THD value of 8.85%, outperforming conventional PSO and other recent metaheuristic algorithms. A prototype traction BLDC motor based on the proposed algorithm has achieved a power output of 2.45 kW, with results superior to theoretical predictions than those predicted through theoretical models. A comparative study shows the superiority of the proposed PSOSAV over canonical PSO and some other recently proposed metaheuristic algorithms for parameter optimization of BLDC motors.
TÖREN et al. (Sun,) studied this question.