Hybrid fuzzy-PSO based self-tuning fractional order PI controller for BLDC motors in electric vehicles: Comparative analysis and experimental validation

Abstract

The performance of brushless DC (BLDC) motor drives in electric vehicles is highly dependent on effective speed control, which becomes challenging under dynamic load and environmental variations. However, conventional control strategies typically rely on offline parameter tuning and lack real-time adaptability, making them less effective in rapidly changing operating conditions. To overcome these limitations, this paper presents a novel hybrid fuzzy logic–Particle Swarm Optimization (PSO) self-tuning fractional-order proportional–integral (FOPI) controller for precise speed control of BLDC motors in electric vehicles. In the proposed method, the proportional and integral gains are optimized online using the PSO algorithm, while a fuzzy logic controller continuously adjusts the fractional-order integral coefficient. This combination leverages the local adaptability of fuzzy logic and the global optimization capabilities of PSO, enabling real-time adaptation to system variations. To assess its effectiveness, the proposed approach is compared against three conventional FOPI controllers tuned offline using established metaheuristic methods: PSO, Bee Colony Optimization (BCO), and Grey Wolf Optimization (GWO). For validation, a hardware-in-the-loop (HIL) platform was implemented using two dSPACE CLP1104 systems, along with a dedicated test bench that replicates realistic driving scenarios. The controller’s performance was evaluated under various conditions, including dynamic speed transitions, external load disturbances, and torque ripple behavior, following the New European Driving Cycle (NEDC) profile. The experimental results demonstrate that the proposed controller achieves faster convergence, stronger disturbance rejection, reduced torque ripple, and shorter rise time compared to offline-tuned controllers. These results validate the effectiveness and practical relevance of this adaptive control strategy for high-precision, real-time BLDC motor applications in electric vehicles.