Dongyang Cai ^1* Jiewen Zuo ² Xiaolong Hao ²

• ¹ Guangzhou Electric Power Company, Guangzhou, China
• ² China University of Mining and Technology, Xuzhou, Jiangsu Province, China

The rapid integration of renewable energy sources such as wind and solar into power grids presents significant challenges due to their inherent variability and unpredictability. Traditional power systems, designed for stable, fossil-fuel-based generation, struggle to maintain reliability and cost-effectiveness in this new landscape. This paper introduces a novel framework that integrates robust optimization with online learning to address these challenges, offering a dynamic and adaptive approach to managing power transmission systems. By leveraging robust optimization, the model ensures that the system remains resilient under worst-case scenarios, while the online learning component continuously refines decision-making based on the latest data. Simulation results demonstrate a substantial improvement in system performance, with operational costs reduced by up to 12% and system reliability enhanced by 1.4% as renewable integration increases from 10% to 50%. The framework also significantly reduces the need for reserve power, particularly under high variability conditions, showcasing its effectiveness in enhancing both economic and environmental outcomes in modern power grids. This integrated approach represents a significant advancement in energy management, providing a robust solution for the sustainable integration of renewable energy into power systems.