Adaptive speed synchronization of dual-motor drives using a recurrent Type-2 fuzzy NARX-CMAC network
This paper proposes a recurrent Type-2 fuzzy Nonlinear autoregressive networks with exogenous input cerebellar model articulation controller (NARX-CMAC) control strategy for speed synchronization in dual-motor drive systems operating under nonlinear dynamics, external disturbances, and load variations. Precise synchronization between two motors is a critical requirement in many industrial applications, where even small speed deviations may degrade product quality, increase mechanical stress, and reduce system reliability. However, conventional controllers such as PID and linear model-based approaches often show limited performance when the plant is subject to parameter uncertainty, nonlinear effects, and measurement noise. To address these challenges, the proposed method integrates three complementary mechanisms into a unified nonlinear control framework. Type-2 fuzzy inference is employed to enhance uncertainty handling, the NARX captures the temporal behavior of the system and improves dynamic prediction, while the CMAC provides fast local learning and efficient online adaptation. In addition, the recurrent structure enables the controller to exploit past system information, thereby improving transient synchronization and disturbance rejection capability. The experimental results demonstrate that the proposed controller achieves high synchronization accuracy, rapid dynamic response, and stable operation even in the presence of noise and time-varying loads. Compared with conventional control methods, the developed Type-2 fuzzy NARX-CMAC scheme offers superior robustness and shows strong potential for intelligent synchronization control in high-performance multi-drive applications.