Distributed Peer-to-Peer Optimization Based on Robust Reinforcement Learning with Demand Response: A Review
Andrés Martínez, Paúl Arévalo
Abstract
The increasing adoption of renewable energy resources and the growing need for efficient and adaptable energy management have emphasized the importance of innovative solutions for energy sharing and storage. This study aims to analyze the application of advanced optimization techniques in decentralized energy systems, focusing on strategies that improve energy distribution, adaptability, and reliability. This research employs a comprehensive review methodology, examining reinforcement learning approaches, demand response mechanisms, and the integration of battery energy storage systems to enhance the flexibility and scalability of P2P energy markets. The main findings highlight significant advancements in robust decision-making frameworks, the management of energy storage systems, and real-time optimization for decentralized trading. Additionally, this study identifies key technical and regulatory challenges, such as computational complexity, market uncertainty, and the lack of standardized legal frameworks, while proposing pathways to address them through intelligent energy management and collaborative solutions. The originality of this work lies in its structured analysis of emerging energy trading models, providing valuable insights into the future design of decentralized energy systems that are efficient, sustainable, and resilient.