Optimal Virtual Power Plant Operational Regime Under Reserve Uncertainty
Han Wang, Youwei Jia, Chun Sing Lai, Kang Li
Abstract
Virtual power plant (VPP) has become an important resource for reserve provision owing to its fast-responding capability. In this paper, an optimal VPP operational regime considering reserve uncertainty is proposed, which includes a novel day-ahead offering strategy and a real-time dispatching model. At the day-ahead stage, the offering strategy gives the VPP’s price-dependent offers in the energy market under multiple uncertainties on market price, renewable generation, and calls of reserve deployment. A hybrid stochastic minimax regret (MMR) model is proposed to facilitate making offering decisions in the electricity market. At the real-time dispatching stage, generation scheduling can be realized based on the MMR criterion in an online fashion. To alleviate the intrinsic conservativeness of the dispatching model, a self-adaptive algorithm is also proposed to instantly modify the confidence bounds. The proposed regime is comprehensively tested through extensive case studies, which demonstrate the effectiveness of our method in obtaining operational decisions that are less conservative.