ENTRUST: Energy trading under uncertainty in smart grid systems

In this paper, real-time energy trading in smart grid is modeled as an optimization process under uncertainties of demand and price information - a problem perspective that is divergent from the ones in the existing literature. Energy trading in smart grid is affected by demand uncertainties - intermittent behavior of renewable energy sources, packet loss in the communication network, and fluctuation in customers' demands. Energy trading is also affected by price uncertainty due to the demand uncertainties. In such uncertainty-prone scenario, we propose the algorithm named ENTRUST using the principles of robust game theory to maximize the payoff values for both sides - customers, and grid. We show the existence of robust-optimization equilibrium for establishing the convergence of the game. Simulation results show that the proposed scheme performs better than the existing ones considered as benchmarks in this study. Utilities for the customers are also maximized in order to promote cost-effective and reliable energy management in the smart grid.