Evolutionary Game Theory as a Catalyst in Smart Grids: From Theoretical Insights to Practical Strategies

Abstract

The increase in the installation of distributed energy resources (DERs) globally has led to a remarkable transformation in the structure of smart grids due to the growing number of energy participants. Recently, electricity markets (EM) have received substantial attention as a viable solution for the complex issue of managing DERs. Modeling the power grid as a complex system of interacting components facilitates investigating the interaction among electricity producers and consumers to maintain the total generation and demand at a balance. In this work, we present a review of the recent advances in adopting evolutionary game theory (EGT), to mitigate challenges in the emerging smart grid, as a decision-making framework for trading dynamics and considering large populations. It includes a taxonomy of various EGT applications in energy trading dynamics, DER management, and policy and infrastructure development. Finally, the linkage between multi-agent reinforcement learning (MARL) and EGT is provided, highlighting their mathematical parallels in the context of smart grid applications.

Keywords

• Game theory
• evolutionary games
• electricity markets
• reinforcement learning
• multi-agent system
• dynamic populations