Enhancing doubly-fed induction generator performance in wind energy systems using intelligent control

This study investigates the optimization of doubly-fed induction generator (DFIG) performance in wind energy systems using intelligent control techniques. The research focuses on two primary control methods: fuzzy logic and artificial neural networks, complemented by neural space vector modulation (NSVM) for enhanced power quality. The paper presents a comprehensive model of the wind energy conversion chain and vector control of the DFIG, with particular emphasis on improving active and reactive power control. Simulation results demonstrate the effectiveness of both fuzzy logic and neural network systems in tracking power changes. However, the neural network-based control exhibits superior performance in reducing current and electromagnetic torque ripples. a key aspect of the study is the total harmonic distortion (THD) analysis of the source current. the neural network system achieves a lower THD value (0.11%) compared to the fuzzy logic system (0.32%), indicating better power quality. additionally, a robustness test involving alterations to generator parameters reveals the fuzzy logic system's greater adaptability to these changes. the research also explores the implementation of space vector modulation (SVM) based on neural networks (NSVM) to replace conventional switching pulse-width modulation (PWM) techniques. This approach significantly improves power quality and enhances performance in the presence of machine parameter variations. The study concludes by highlighting the potential of these intelligent control techniques in improving wind energy system performance. it provides valuable insights for developing more efficient and reliable control systems for DFIG in wind energy applications. The findings contribute to ongoing efforts to enhance renewable energy efficiency and reliability, potentially accelerating the adoption of wind power in the global energy mix.