Experimental Validation of Second-Order Adaptive Fuzzy Logic Controller for Grid-Connected DFIG Wind Power Plant

Abstract

This paper introduces a second-order adaptive fuzzy logic controller (SO-AFLC) to enhance the characteristics of a doubly fed induction generator (DFIG) inside a grid-tied wind power plant (WPP). SO-AFLCs were utilized to maximize the output of the DFIG wind power plant (WPP) and improve dynamic responsiveness with extremely low mean square error. When comparing the mean square error of SO-AFLC with proportional-integral controllers (PI) and adaptive fuzzy logic controllers (AFLC), the reductions are 87.38% and 40.39%, respectively. This controller prevents overshoots and oscillations. DFIG wind power plant is modeled and simulated using Matlab/Simulink package. Under the unit step wind speed profile, SO-AFLC improved the steady-state error in the Cp waveform by 63.25% compared with the PI controller and 13.12% compared with AFLC. DSpace1104 is used to conduct an experimental investigation to validate the simulation results. In addition, realistic data from the wind farm at RAS Ghareb in the Gulf of Suez, Egypt, are used in this study to achieve more realistic results. Compared to those obtained with PI and AFLC, the results obtained using SO-AFLCs showed fast time response, high convergence rate, reduced peak overshoot, less undershoot, and low steady-state error in terms of power coefficient of the turbine, DC link voltage control, and rotor speed tracking. In addition, a wind turbine performance index based on gross system integral absolute error (IAE) is provided. This index is used to illustrate the SO-AFLC methodology's viability compared to AFLC and PI under the same wind turbine conditions.