performance improvement of combined wind farms using ann-based statcom and grey wolf optimization-based tuning

Changes in the electric supply can significantly affect electronic devices since they are very sensitive. due to a nonlinear system with multiple interconnected and unpredictable demands in the smart grid, the electricity system is facing several issues, including power quality, reactive power management, and voltage drop. to address these problems, a static synchronous compensator (statcom) is frequently used to compensate and correct the voltage level at the power bus voltage. in this study, an artificial neural network (ann) and gwo based controlled statcom has been developed to replace the traditional pi based controller and enhance the overall statcom performance. the ann controller is preferred due to its simplicity, adaptability, resilience, and ability to consider the non-linearities of the power grid. to train the classifier offline, data from the pi controller was utilized. the matlab/simulink software was employed to assess the effectiveness of statcom on a 25 km transmission line during increased load and three faults. the combined results of the pi and ann controllers indicate that the ann controller significantly improves statcom efficiency under different operating conditions. moreover, the ann controller outperforms the traditional pi controller in terms of results.