FORECASTING AND MANAGEMENT IN SMART GRID WITH ARTIFICIAL INTELLIGENCE

Abstract

As conventional power industries transition toward the integration of smart grid features, decarbonization, and distributed energy systems, more sources of uncertainties are being introduced into existing power systems. The uncertainties significantly complicate grid analyses and increase the risk of control in smart grids. This thesis proposes an artificial intelligence (AI)-based framework for uncertainty forecasting and management in smart grids with three data-driven components integrated, which are data preprocessing, uncertainty analysis, and uncertainty management. The three data-driven components are validated using three corresponding tasks. It is shown that the proposed AI-based framework significantly outperforms state-of-the-art methods in terms of accuracy, efficiency, and reliability, with quantified improvements given. The proposed AI-based framework has considerable potential in enhancing uncertain smart grid control in the sense that it can reduce the risk of control by providing accurate uncertainty quantification and adequate control strategies.