Modeling and Robust Control with Wind Speed Estimation by Artificial Neural Networks of a DFIG Wind Turbine Under Both Normal Operation and Grid Fault

Abstract

For high performance and reliability of the wind turbine generator based on doubly fed induction generator connected to the electrical network through back-to-back converter, we propose in this paper the wind speed estimation algorithm based on artificial neural network in order to eliminate the need of using the anemometer to measure wind speed. On the basis of the ANN algorithm to estimate the wind speed, a developed control strategy consists on using the field oriented technique without any simplification in the DFIG model. The objective to be achieved with this control strategy is to maximize the captured power with the MPPT algorithm at low to medium wind speeds. The major advantage of the proposed control law is that when the stator voltage drops to an acceptable value, this control system is not too much influenced. Simulation results for 1,5 MW, with a DFIG based Wind Energy Conversion System (WECS) confirm the effectiveness and the performance of the global proposed approaches.