Open AccessA 24-sectors direct power control-feedforward neural network method of DFIG integrated to dual-rotor wind turbine
Author(s) -
Habib Benbouhenni
Publication year - 2021
Publication title -
international journal of applied power engineering
Language(s) - English
Resource type - Journals
eISSN - 2722-2624
pISSN - 2252-8792
DOI - 10.11591/ijape.v10.i4.pp291-306
Subject(s) - control theory (sociology) , ac power , wind power , total harmonic distortion , feed forward , stator , rotor (electric) , induction generator , turbine , engineering , feedforward neural network , power control , power (physics) , computer science , artificial neural network , control engineering , control (management) , voltage , electrical engineering , physics , mechanical engineering , artificial intelligence , machine learning , quantum mechanics
In this work, a 24-sector direct power control (24-sector DPC) of a doubly-fed induction generator (DFIG) based dual-rotor wind turbine (DRWT) is studied. The major disadvantage of the 24-DPC control is the steady-state ripples in reactive and active powers. The use of 24 sectors of rotor flux, a feedforward neural network (FNN) algorithm is proposed to improve traditional 24-sector DPC performance and minimize significantly harmonic distortion (THD) of stator current and reactive/active power ripple. The proposed method is modeled and simulated by using MATLAB/Simulink software under different tests and compared with conventional 24-sector DPC.