An Autonomous Demand Response Program in Smart Grid with Foresighted Users

In smart grid, demand response is a viable approach to motivate users towards shifting the demand during the peak load periods. Each user can also benefit by reducing its total cost. In most of the existing studies, the demand response program is modeled as a one-shot game among myopic users, who aim to minimize their cost in one period of time. In this paper, we show that the Nash equilibrium (NE) in the one-shot game can be inefficient in reducing the peak load demand and the users' cost. We address the inefficiency of the NE by modeling the demand response program as a repeated game. A grim-trigger strategy is proposed to determine the subgame perfect equilibrium. To address the issue of fairness, we partition the set of users into groups. In each time period, only one group of users are required to participate in the demand response program. Simulation results show that the proposed demand response repeated game can benefit both the users, by reducing their long-term cost, and the utility company, by reducing the peak-to-average ratio in the aggregate load demand.