Optimization of BLDC Motor Geometry using Particle Swarm Optimization Algorithm to Achieve Efficiency Balance Across Various Electric Vehicle Traction Requirements

Gasoline vehicles (GVs) contribute significantly to global energy crises and environmental pollution, while electric vehicles (EVs) offer a more sustainable alternative. However, the current development and deployment of EVs are largely limited to ideal operating conditions, such as urban roads. To compete effectively with GVs, EVs must have drivetrain systems that maintain high efficiency even in non-ideal environments, including rural areas and rough terrains. This study proposes a geometry optimization method for a 1 kW Brushless DC (BLDC) motor to improve energy efficiency under three primary EV traction scenarios: climbing, acceleration, and cruising. The optimization targets nine geometric parameters—outer and inner stator radius, magnet thickness, rotor yoke thickness, shoe stator thickness, magnet width, shoe stator width, stator pole width, and back-iron thickness. The optimization is performed using a Particle Swarm Optimization (PSO) algorithm integrated with Finite Element Method Magnetics (FEMM) and analytical performance evaluation. The optimization constraints are derived from traction dynamics, weight, and volume limitations based on the regulations of the Indonesian Electric Vehicle Competition (Kompetisi Mobil Listrik Indonesia, KMLI). The results show that the optimized BLDC motor geometry can increase efficiency by up to 24.3% and torque by 11.3% compared to the baseline design. This research contributes a high-efficiency BLDC motor design tailored for dynamic EV traction demands under regulatory and extreme operational constraints, making it highly suitable for further development, including additional performance scenarios such as deceleration and cornering.