AI-driven power system flexibility: a review of demand response, renewable integration, and grid reliability

The global energy sector is undergoing a significant transformation, driven by the need for deep decarbonization and the rapid advancement of artificial intelligence (AI). This review explores the integration of AI-driven flexibility, demand response (DR), and renewable energy. It aims to achieve a secure and sustainable energy transition. We present a systematic analysis of high-quality research publications, emphasizing recent progress in AI applications for power systems. We establish a structured taxonomy of AI techniques tailored for power system flexibility. We also investigate critical challenges like renewable intermittency and grid stability. Moreover, we discuss influential AI methodologies deployed across power systems. A critical analysis evaluates their techno-economic trade-offs and implementation readiness. Our review offers an integrated analysis of AI-driven grid applications that addresses both technical performance and operational challenges. It culminates in a proposed roadmap for the most promising interdisciplinary approaches. The reviewed evidence indicates that AI-driven flexibility reduces integration costs for variable renewables. It also significantly improves DR participation rates. Furthermore, it enhances grid reliability beyond conventional methods. This review synthesizes diverse research into a coherent roadmap. It serves researchers and practitioners working on intelligent energy systems.