Data Driven Energy Management Solutions for Smart Grid Applications

Abstract

The growing integration of renewable energy sources, such as wind and solar power, into modern power systems introduces significant challenges due to their inherent variability and intermittency. These challenges affect energy management, grid stability, and operational efficiency in active distribution networks. This thesis proposes a series of novel data-driven approaches to address these challenges, focusing on enhancing energy forecasting accuracy, optimizing grid control strategies, and improving the resilience and stability of power systems under high renewable energy penetration.This thesis demonstrates how the proposed hybrid models, advanced control frameworks, and data-driven algorithms can be used to optimize energy management and control in modern active distribution networks. Several Case studies and benchmark comparations are conducted to evaluate the effectiveness of the proposed methods, demonstrating improvements in forecasting accuracy, voltage control performance, and resilience against fluctuations. This research contributes to advancing the scalability, adaptability, and effectiveness of data-driven solutions for smart grid applications, supporting the transition to more resilient and efficient energy systems.