Integration of AI-driven digital twins for real-time optimization of renewable energy grids

The increasing integration of renewable energy sources, such as solar photovoltaics, wind turbines, hydropower, and energy storage systems, introduces substantial variability and complexity into modern power grids. This variability challenges grid stability, supply-demand balancing, and operational resilience. Digital twin (DT) technology, which provides a dynamic, real-time virtual representation of physical assets and systems, has emerged as a transformative tool for monitoring, analyzing, and optimizing energy grids. The incorporation of artificial intelligence (AI) into digital twins further enhances their capabilities, enabling predictive analytics, adaptive control, fault detection, and real-time decision-making for grid-specific objectives such as voltage/frequency regulation, congestion management, DER coordination, curtailment reduction, and resilience under fast renewable ramps. Machine learning, deep learning, and reinforcement learning techniques facilitate accurate forecasting of energy generation and demand, intelligent dispatch of distributed energy resources, and predictive maintenance, while hybrid models combining physics-based simulations with AI improve prediction accuracy in data-sparse or high-uncertainty environments. Despite these advancements, challenges persist, including data quality and availability, computational scalability, cybersecurity risks, and interoperability issues. This review synthesizes current research on AI-driven digital twins in renewable energy grids, highlights methodological and technological gaps, and identifies future research directions for developing resilient, scalable, and adaptive energy systems. The findings underscore the potential of AI-integrated digital twins to accelerate the transition toward intelligent, sustainable, and climate-resilient energy infrastructures. In addition, this review incorporates sustainability-oriented intelligent-system methodologies such as energy-aware edge–cloud cyber-physical architectures and digital-twin-enabled lifecycle sustainability frameworks to align DT optimization with contemporary sustainability practice better.