Intelligent Control of Doubly-Fed Induction Generators.

Abstract

This thesis investigates the intelligent flux-oriented control of active and reactive powers in a Doubly Fed Induction Generator (DFIG) to optimize its performance. The study begins with a comprehensive review of the state-of-the-art in Doubly-fed induction machines, focusing specifically on generators. It then gets into Artificial Neural Networks, first as a concept and then as an advanced control strategy for our system. After that, it also explores the Adaptive Neuro-Fuzzy Inference System (ANFIS), theoretically, and then its application to our control system in a similar manner. Various tests were conducted to evaluate the efficiency and robustness of both control strategies. At the end, a comparative analysis was performed to highlight the strengths and weaknesses of each approach. Simulation results in the MATLAB/SIMULINK environment demonstrate that both controllers deliver excellent performance in term of robustness. However, ANFIS exhibits a slight edge regarding response time and robustness. The findings underscore the potential of intelligent control techniques in optimizing DFIG-based wind turbines; This improvement is attributed to the adaptive and flexible nature of intelligent controllers, which better handle the complexities and uncertainties inherent in wind energy systems.