AI-BASED VIRTUAL INERTIA CONTROL STRATEGY FOR DFIG WIND TURBINES TO ENHANCE GRID FREQUENCY STABILITY

Abstract

The increasing penetration of renewable energy sources has significantly reduced the effective inertia of modern power systems. Wind turbines based on Doubly-Fed Induction Generators (DFIGs) are widely deployed due to their efficiency and controllability; however, their power electronic interface decouples the mechanical rotor from the grid, limiting the natural inertia contribution. As a result, large frequency deviations and high rates of change of frequency (ROCOF) may occur during disturbances. Virtual inertia control has emerged as an effective technique for enabling wind turbines to provide synthetic inertial support to the grid. This paper presents an artificial intelligence (AI)-based approach for enhancing the virtual inertia response of a DFIG wind turbine system. A detailed mathematical model of the wind turbine and DFIG is developed, followed by the design of a virtual inertia controller. A neural-network-based adaptive controller is proposed to dynamically adjust the inertia gain according to system frequency deviation and rotor speed. The proposed controller is implemented and evaluated using MATLAB/Simulink under load disturbance scenarios. Simulation results demonstrate that the AI-based virtual inertia controller improves frequency stability by reducing frequency deviation, lowering ROCOF, and improving frequency nadir compared with conventional fixed-gain controllers.