A virtual reservoir electricity market design applied to the Brazilian system using an Agent Based Model

This paper discusses the current Brazilian electricity market, brings out some dilemmas that should be examined in order to implement a more market-oriented approach, and describes a new market design to overcome these issues. The proposed market design is based on virtual reservoirs and aims at enhancing the flexibility to enable market participants to comply with their contracts, while still ensuring the efficient use of the water and maintaining the current level of the security of supply. In addition, to simulate the behavior of the market participants in this new framework, an Agent-Based Model-ABM where agents use reinforcement Q-Learning - is developed and applied to a case study that includes a centralized dispatch as currently exists in Brazil. The results obtained so far show that this new design is suitable to allow hydros managing their commercial contract commitments with extra flexibility.