Effect of Input and Fuzzy Membership Function on ANFIS Controller in DFIG-Based Wind Energy Conversion System

This work investigates how input and fuzzy membership functions affect the effectiveness of an Adaptive Neuro-Fuzzy Inference System (ANFIS) controller designed for managing active and reactive power in a grid-coupled Doubly Fed Induction Generator Wind Energy Conversion System (DFIG-WEC). The test scheme involves of two interconnected DFIG wind turbines that interface through the grid over a bidirectional converter. To control the Direct link capacitor potential difference in the DFIG generation plant, the ANFIS regulatory framework is applied directly to the Grid Side Converter. The proposed method considers single, double input and various membership functions, such as triangular, trapezoidal, and Gaussian, to examine their effectiveness in the ANFIS controller in power output control in changing situations. In this work, the effect of these different membership functions has been tested under conditions of varying wind speeds and grid failures using extensive simulations in MATLAB/Simulink. These results demonstrate how the membership function selection impacts the controller’s performance. This study emphasises the significance of selecting membership functions suitable for practical settings to guarantee the overall stability and effectiveness of DFIG-WECS. These results are, therefore, very useful in designing and optimizing ANFIS controllers for renewable energy systems, targeting the critical role of membership functions in attaining improved dynamic performance and power grid compatibility. The findings thus set a foundation for further progress in intelligent control strategies for sustainable energy solutions.