ESTIMATION OF DOUBLY FED INDUCTION GENERATOR BASED WIND ENERGY CONVERSION SYSTEM WITH HAMMERSTEIN-WIENER MODEL

Abstract

The aim of this paper is to estimate the parameters such as real power, reactive power, DC link voltage and generator torque of the doubly fed induction generator based wind energy conversion system with Hammerstein-Wiener models. The Hammerstein-Wiener models are developed for the complete wind energy conversion system and the parameters of Hammerstein-Wiener models are estimated with different procedures such as piecewise linear, sigmoid network, saturation, dead zone, wavelet network and one dimensional polynomial. Wavelet network and sigmoid network are found to be the techniques which produce the best fit percentage for all the parameters except for the generator torque in the low wind speed region and real power of medium wind speed region. Dead zone technique produces the best fit percentage for generator torque in low wind speed region. Similarly, saturation technique produces the best fit percentage for real power of doubly fed induction generator in medium wind speed region. The stability is analyzed with transient response, frequency response and pole-zero maps of the appropriate models.