# Robust PMSM Speed Control for EV Traction Drives: A FOPSO-Optimized Hybrid Fuzzy Fractional-Order PI Strategy

## TL;DR

This paper introduces a new control strategy for electric vehicle motors that improves speed control despite real-world challenges like sensor noise and changing road conditions.

## Contribution

A novel FOPSO-optimized hybrid fuzzy fractional-order PI controller validated using a high-fidelity co-simulation framework.

## Key findings

- The proposed controller eliminates overshoot in step responses and maintains stability with 2.0× inertia mismatch.
- It reduces speed tracking RMSE by 75.0% under the EPA urban driving cycle compared to standard PI controllers.
- The method is computationally feasible for real-time implementation in EV traction drives.

## Abstract

High-performance speed control of Permanent Magnet Synchronous Motor (PMSM) drives in Electric Vehicle (EV) applications faces significant challenges due to inherent nonlinearities, parameter variations, and signal non-idealities such as sensor noise and measurement latency. To address these issues, this paper proposes a robust PI-based Fractional-Order PSO-Fuzzy Weight Controller (PI-FOPSOFWC). The proposed strategy integrates a fractional-order PI (FOPI) core to ensure iso-damping robustness, a fuzzy inference mechanism for online gain scheduling against nonlinear load dynamics, and a novel Fractional-Order Particle Swarm Optimization (FOPSO) algorithm for optimal parameter tuning. A key contribution of this study is the validation of the control strategy within a high-fidelity co-simulation framework coupling MATLAB/Simulink with CarSim 2023, which incorporates realistic vehicle dynamics and time-varying road loads unavailable in conventional simplified simulations. Co-simulation results demonstrate that the proposed controller effectively eliminates overshoot in step responses and maintains stability under significant parameter mismatches (2.0× inertia). Furthermore, under the EPA urban driving cycle, the proposed method reduces the speed tracking Root Mean Square Error (RMSE) by 75.0% compared to the standard PI controller. Computational complexity analysis further confirms the feasibility of the proposed algorithm for real-time implementation in commercial EV traction drives.