AI-Based Damping of Electromechanical Oscillations by Using Grid-Connected Converter

The proliferation of grid-connected converter interfaced energy sources in Smart Grids, enhance sustainability and efficiency as well minimizing power losses costs. However, concerns arise regarding the stability reliability future smart grids due to this wide integration electronic devices, which are recognized affect dynamic response system, especially during disturbances. For instance, apart from lower damping existing electromechanical modes, new low-frequency oscillations begin appear. Yet, ability converters provide grid support functionalities can alleviate aforementioned challenges. Relevant studies show that these be enhanced even further, if information oscillation characteristics available. Traditional methods for extracting modal very suited monitoring purposes, however, they pose certain limitations when considered control applications. Therefore, paper proposes a multi-band intelligent damper (MiPOD) exploits 1) inherent damp multiple 2) modeling capabilities Artificial Intelligence (AI) predicting frequency operating conditions change. Essentially, MiPOD integrates AI model loop attenuate modes oscillation. proposed controller is validated different disturbances randomly generated points two area system. Specifically, case Random Forest ensemble regressor developed tracking modes. As it shown, improve overall performance system under various contingency scenarios with only 6% corresponding total nominal capacity synchronous generators. In addition, abilities demonstrated vast range just by tuning parameters; predicted frequencies local inter-area mode.