Learning Bidding Strategies in Local Electricity Markets using Ant Colony optimization

Local energy markets (LM) are attracting significant interest due to their potential of balancing generation and consumption and supporting the adoption of distributed renewable sources at the distribution level. Besides, LMs aim at increasing the participation of small end-users in energy transactions, setting the stage for transactive energy systems. In this work, we explore the use of ant colony optimization (ACO) for learning bidding strategies under a bi-level optimization framework that arises when trading energy in an LM. We performed an empirical analysis of the impact of ACO parameters have in the learning process and the obtained profits of agents. After that, we analyze and compare ACO performance against an evolutionary algorithm under a realistic case study with nine agents trading energy in the day-ahead LM. Results suggest that ACO can be efficient for strategic learning of agents, providing solutions in which all agents can improve their profits. Overall, it is shown the advantages that an LM can bring to market participants, thereby increasing the tolerable penetration of renewable resources and facilitating the energy transition.