Artificial neural network for standalone induction generator :control wind dc microgrid

Abstract

This study aims to develop an intelligent system to estimate the generator rotor speed in a wind power generation system without the need to use mechanical sensors directly, using advanced techniques such as Kalman Filter and Artificial Neural Network. The rotor speed is a vital variable that directly affects the generator control process, so obtaining an accurate estimate of it contributes to improving system efficiency and increasing the amount of energy extracted from the wind. This approach provides a low-cost and more reliable alternative to traditional sensors, which may be damaged or require regular maintenance, especially in harsh environments. Through this model, the overall performance of the electro-mechanical system can be improved, operational costs can be reduced, and better energy productivity can be achieved in renewable energy systems.