Grid-Connected DFIG Driven Wind System for Low Voltage Ride Through Enhancement using Neural Predictive Controller

Abstract

Doubly Fed Induction Generators (DFIGs) are exposed to severe grid faults. In such cases, Low Voltage Ride Through (LVRT) enhances the DFIG’s performance under fault conditions. This paper investigates the LVRT enhancement of the DFIG system under grid disturbance. The paper proposes the Neural Predictive (NP) controller for the DFIG based Wind Turbine (WT) generator during grid faults. This controller operates with the Levenberg-Marquardt (LM) algorithm for its fast convergence. The algorithm based Neural Predictive (NP) controller is operated for large signal stability. The proposed controller has the benefit of reducing the peak values and uncertainties, which are raised for the system parameters during grid faults. Further, the proposed controller outcome is compared with existing controllers in the literature such as PI, PID, Feed-Forward Neural Network (FNN), and 2nd order Sliding Mode Controller (SOSMC) with the help of MATLAB/SIMULINK. The Hardware-In-Loop (HIL) is used to validate the simulation results, which have been performed on the OPAL-RT setup. According to the results that are obtained in this study, the proposed controller improved the LVRT performance of the DFIG-Wind Turbine (WT) system while operating under dynamic conditions. © 2022, The Institution of Engineers (India).