A Spot-Risk-Based Approach for Addressing Problems of Decision-Making under Uncertainty

Considerable research has been devoted on the development of decision-making models suitable for tackling decision problems integrating some amount of uncertainty. However, such approaches are either problem-specific, either too general to suit certain problems. In this paper we propose an approach for performing decision-making under uncertainty suitable for problems in which decisions must be made sequentially (i.e.: in time). We apply the proposed model to the case of a virtual power plant (comprising a wind farm and an energy storage device) operating under market conditions with the objective of reducing the imbalances generated by the plant while maximizing the profit of the plant operator. Using real-world data, we have simulated and evaluated several operation scenarios for an entire year of operation. Wind power forecasts produced by a state-of-art wind power forecasting model (based on KDE estimators) were used.