Scenario Reduction for Futures Market Trading in Electricity Markets

Abstract

To make informed decisions in futures markets of electric energy, stochastic programming models are commonly used. Such models treat stochastic processes via a set of scenarios, which are plausible realizations throughout the decision-making horizon of the stochastic processes. The number of scenarios needed to accurately represent the uncertainty involved is generally large, which may render the associated stochastic programming problem intractable. Hence, scenario reduction techniques are needed to trim down the number of scenarios while keeping most of the stochastic information embedded in such scenarios. This paper proposes a novel scenario reduction procedure that advantageously compares with existing ones for electricity-market problems tackled via two-stage stochastic programming.