A Robust Reserve Scheduling Technique for Power Systems with High Wind Penetration

In this paper a novel algorithm to determine the minimum cost reserves for power systems with high wind penetration, is presented. We first determine the status and schedule of each generating unit by solving a deterministic unit commitment problem, where only the wind power forecast is taken into account. Having then computed the generation schedule, we formulate a linear stochastic program with chance constraints, whose solution provides the level of reserves to achieve production-demand balance with a certain probability. To solve the resulting optimization program, a Markov chainbased model for the wind power error is constructed, and the “scenario approach” is employed, leading to a tractable optimization problem with probabilistic performance guarantees. The efficiency of the proposed approach is compared with benchmark techniques via Monte Carlo simulations. Two different market implementations are considered; a day-ahead and an intra-day market. The improvement afforded by the additional information that the intra-day market provides, is quantified in terms of reserve cost, load shedding and wind generation spillage.