Advanced Parametric Estimation Methods for Permanent Magnet BLDC Motor Under Uncertainties: Quantification of EV Dynamics

Permanent magnet brushless DC motors (PMBLDCMs) are widely used in vehicles due to their high efficiency, better power density, and reliability. Precise motor parameter estimation is significant in improving the performance of the vehicle drive. This paper presents a comparative investigation of advanced parameter estimation techniques-namely Monte Carlo (MC), Markov Chain Monte Carlo (MCMC), Kalman Filter (KF), Extended Kalman Filter (EKF), Sequential Monte Carlo/Particle Filter (SMC/PF), and Unscented Kalman Filter (UKF) applied to a 250W PMBLDCM. With these advanced parametric estimation methods, PM-BLDC motors can be fine-tuned, resulting in more precise control over the motor’s torque, speed, and position of the drive. This leads to the development of fault-tolerant systems that can detect and respond to motor abnormalities in real-time, even under dynamic operating states. The motor model includes key parameters such as stator resistance ( $2.875\Omega $ ),stator inductance (8.5mH), inertia ( $0.8\ast 10^{-3}$ kg.m ${}^{2}$ ), viscous damping ( $1\ast 10^{-3}$ N.m.s), flux linkage (0.175wb) and torque constant (0.8N.m/A) by considering electric vehicle (EV) uncertainty quantifications. By incorporating uncertainty quantification techniques, the proposed approach enhances the reliability and accuracy of motor control, which is crucial for optimizing dynamic performance. Key outcomes shows that the UKF achieves superior accuracy and stability across all estimated parameters, making it highly effective for real-time PMBLDCM control in EV applications.