A machine learning approach for active/reactive power control of grid-connected doubly-fed induction generators, By Ardashir Mohammadzadeh

Research

Title	A machine learning approach for active/reactive power control of grid-connected doubly-fed induction generators
Type	Article
Keywords	Doubly-Fed Induction Generator (DFIG) Radial Basis Function Network (RBFN) Recurrent Type-II Fuzzy Neural Networks (RT2FNN) Wind turbine Active/reactive power Artificial intelligence Renewable energies
Researchers	Jafar Tavoosi، Ardashir Mohammadzadeh، Bahareh Pahlevanzadeh، Morad BagherzadehKasmani، Shahab S، Rabia Safdar، Amirhosein Mosavi

Abstract

This paper suggests a new fuzzy method for active power (AP) and reactive power (RP) control of a power grid that includes wind turbines and Doubly Fed Induction Generators (DFIGs). A Recurrent Type-II Fuzzy Neural Networks (RT2FNN) controller based on Radial Basis Function Networks (RBFN) is applied to the rotor side converter for the power control and voltage regulation of the wind turbine equipped with the DFIG. In order to train a model, the voltage profile at each bus, and the reactive power of the power grid are given to the RT2FNN as the input and output, respectively. A wind turbine and its control units are studied in detail. Simulation results, obtained in MATLAB software, show the well performance, robustness, good accuracy and power quality improvement of the suggested controller in the wind-driven DFIGs.