Comparative Performance of PI and ANN Controllers for DTC of a PMSG-Based Direct-Drive Wind Energy Conversion System

This paper presents an enhanced control strategy for direct torque control (DTC) in a wind energy conversion system (WECS) using a permanent magnet synchronous generator (PMSG). The conventional DTC scheme is improved by replacing proportional-integral (PI) controllers with artificial neural network (ANN)-based controllers. The PI controller parameters are first optimized using the artificial rabbits optimization (ARO) algorithm. In the proposed ANN-based DTC, the ARO-optimized PI controller is replaced with neural network-based regulators for more precise torque and flux control. A comparative analysis in MATLAB/Simulink shows that the ANN-based DTC achieves a torque ripple of 0.4735 Nm, compared to 0.5784 Nm for the ARO-optimized PI-based DTC. The flux ripple is improved from 0.0178 Wb to 0.0056 Wb. The ANN-based DTC also provides better dynamic behavior and robustness to system parameter variations. This work demonstrates the potential of ANN-based controllers in improving DTC strategies for PMSGs, offering more robust and high-performance control systems for renewable energy applications.