Energy Trading in Local Energy Markets: A Comprehensive Review of Models, Solution Strategies, and Machine Learning Approaches

The increasing adoption of renewable energy sources and the emergence of distributed generation have significantly transformed the traditional energy landscape, leading to the rise of local energy markets. These markets facilitate decentralized energy trading among different market participants at the community level, fostering greater energy autonomy and sustainability. As local energy markets gain momentum, the application of artificial intelligence techniques, particularly reinforcement learning, has gained substantial interest in optimizing energy trading strategies by interacting with the environment and maximizing the rewards by addressing the decision complexities by learning. This paper comprehensively reviews the different energy trading projects initiated at the global level and machine learning approaches and solution strategies for local energy markets. State-of-the-art reinforcement learning algorithms are classified into model-free and model-based methods. This classification examines various algorithms for energy transactions considering the agent type, learning methods, policy, state space, action space, and action selection for state, action, and reward function outputs. The findings of this work will serve as a valuable resource for researchers, stakeholders, and policymakers to accelerate the adoption of the local energy market for a more efficient, sustainable, and resilient energy future.