Hybrid ARIMA–TCN–Self‐Attention Control With Superb Fairy‐Wren Optimization for Adaptive BLDC Motor Speed Regulation in Electric Vehicles

Electric vehicles (EVs) have gained significant attention in recent years and are rapidly emerging as a sustainable alternative to fossil fuel‐based transportation due to their efficiency and environmental benefits. The primary objectives in EV development are to maximize energy utilization and extend driving range both of which depend heavily on enhancing motor performance. Among various motor types, the Brushless Direct Current (BLDC) motor stands out as the most suitable choice for EVs owing to its superior speed and position control capabilities, high efficiency, and robust performance. This paper proposes a hybrid intelligent control strategy that integrates an autoregressive Integral moving average (ARIMA), Temporal Convolutional Network (TCN) and a self‐attention mechanism (SAM) for adaptive speed regulation of BLDC motors by providing enhanced learning adaptability and memory characteristics for dynamic system control. To further optimize the ARIMA‐TCN‐SAM performance, Superb Fairy‐wren Optimization Algorithm (SFOA) is employed for hyperparameters tuning. The SFOA algorithm improves convergence speed and prevents premature stagnation during training, ensuring optimal controller performance. Performance metrics such as error, rise time, settling time and overshoot are evaluated and compared with conventional controllers. The ARIMA‐TCN‐SAM–SFOA hybrid controller demonstrates superior results, achieving low overshoot, rise time of 0.0072, 0.0090, and 0.0108 s, and settling time of 0.0151, 0.0188, and 0.0228 s for low, medium, and high inertia cases. With a high regression correlation coefficient of 0.9987 and a lower mean square error (MSE), the controller ensures accurate speed control. Consequently, the integration of ARIMA‐TCN‐SAM with SFOA presents a robust and efficient solution for BLDC motor speed control, marking a substantial advancement in intelligent EV propulsion systems and contributing to the evolution of sustainable and responsive transportation technologies.