Artificial Intelligence Techniques for Optimal Scheduling of Distributed Energy Resources with an IoT-Enabled Smart Energy Management Device: Trends and Challenges

Abstract

The increasing penetration of distributed energy resources (DERs), including solar photovoltaics, wind turbines, and energy storage systems, has transformed modern power systems into decentralized and dynamic smart grids requiring intelligent scheduling strategies. Artificial intelligence (AI), integrated with Internet of Things (IoT)-enabled smart energy management devices, has emerged as an effective solution for optimizing energy generation, distribution, and consumption through real-time monitoring and adaptive decision-making. This review examines AI-driven optimization techniques, including machine learning, deep learning, reinforcement learning, and hybrid metaheuristic algorithms, for optimal DER scheduling. Advanced models such as deep neural networks, convolutional neural networks, long short-term memory networks, and deep reinforcement learning are analyzed for load forecasting, demand prediction, and adaptive energy scheduling. Optimization approaches including particle swarm optimization, genetic algorithms, mixed-integer linear programming, and multi-objective evolutionary algorithms are evaluated for improving operational cost, energy efficiency, and grid stability. The review further investigates the role of IoT devices, edge computing, and cloud platforms in enabling scalable, data-driven energy management across IEEE benchmark systems, residential datasets, and industrial microgrids. Finally, key challenges involving cybersecurity, scalability, data heterogeneity, and renewable uncertainty are discussed, while future research directions include federated learning, blockchain-enabled energy trading, explainable AI, and advanced edge intelligence for resilient smart energy systems.