ADVANCING SMART GRID RELIABILITY FOR ENHANCED PERFORMANCE BY INTEGRATING ARTIFICIAL INTELLIGENCE FOR MODELING RENEWABLE ENERGY AND OVERVOLTAGES

Abstract

Smart grids have revolutionized energy systems by enhancing efficiency, reliability, and sustainability through the integration of advanced communication, control technologies, and renewable energy sources. However, their increased complexity introduces challenges in evaluating reliability, especially with the integration of distributed energy resources, communication disruptions, and cybersecurity risks. This paper reviews advancements in smart grid reliability evaluation, focusing on probabilistic modeling, artificial intelligence (AI), machine learning (ML), and big data analytics. Methods like Monte Carlo simulations and Markov chains are adapted to address uncertainties in grid operations. Key performance indicators (KPIs) such as SAIDI and SAIFI help quantify reliability, while AI and ML improve fault detection and predictive maintenance. Case studies from utilities like Pacific Gas and Electric and the UK’s National Grid demonstrate the practical benefits of these approaches. Additionally, the paper explores overvoltage modeling and insulation reliability. Despite progress, challenges remain in standardizing models, mitigating cybersecurity threats, and optimizing renewable energy integration. Future research will focus on refining predictive models and exploring the impact of emerging technologies like IoT and blockchain.