Enhancing variable frequency drive efficiency using fractional hybrid Particle Swarm Optimization and comprehensive thermal management

This study proposes a Fractional Calculus Hybrid Approach to enhance the dynamic and thermal performance of vector-controlled Permanent Magnet Synchronous Motor (PMSM) drives. The method employs a Fractional Hybrid Particle Swarm Optimization (FHPSO) algorithm that embeds fractional-order calculus within the conventional PSO structure and hybridizes it with Simulated Annealing (SA) to reinforce convergence stability and global exploration. By precisely tuning the Proportional-Integral (PI) controller parameters, the proposed algorithm achieves significant reductions in overshoot and settling time, thereby improving both transient and steady-state drive responses. Simultaneously, a detailed thermal model of the Variable Frequency Drive (VFD) is developed to predict and regulate temperature variations under diverse load and speed conditions, ensuring efficient heat dissipation and enhanced cooling. The integrated control-thermal co-optimization yields a 13.3% to 0.98% and a ~ 64% improvement in temperature regulation (reducing MOSFET temperatures from over 80 °C to ~ 29 °C), validated through comprehensive simulations and Hardware-in-the-Loop (HiL) experiments. Overall, the proposed approach demonstrates that fractional-order hybrid optimization provides a mathematically rigorous and computationally efficient pathway toward improving energy efficiency, thermal reliability, and long-term durability of PMSM-based VFD systems for advanced electric vehicle applications.