Day-ahead and intra-day optimization for energy and reserve scheduling under wind uncertainty and generation outages

• Published in: Electric power systems research 2021-06, Vol.195, p.107133, Article 107133
• Main Authors: Ji, Yongli, Xu, Qingshan, Zhao, Jun, Yang, Yongbiao, Sun, Lu
• Format: Article
• Language: English
• Subjects: Algorithms; Alternative energy sources; Energy and reserve scheduling; Energy sources; N-1 generation outage contingency; Optimization; Reserve scheduling; Robust optimization; Robustness; Stochastic models; Stochastic optimization; Uncertainty; Wind power; Wind uncertainty
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2021.107133

•A day-ahead to intra-day optimization framework for energy and reserve scheduling is proposed.•Robust optimization and stochastic optimization approaches are combined to cope with the system's uncertainty nature.•Supply and demand side resources are coordinated in the normal/uncertainty cases to enhance the economy and reliability.•Continuous load/wind power and discrete unit state uncertainties are considered. The procurement of reserves becomes indispensable to hedge against the increasing penetration of renewable energy resources with variable nature and limited predictability. This paper presents a multi-time-scale optimal scheduling framework within the context of co-optimized electricity markets for energy and reserve, wherein the supply-side and demand-side resources are coordinated in the normal case and uncertainty cases to accommodate the variations of prediction errors over time and enhance the system cost-effectiveness and reliability. The optimization process consists of two phases, day-ahead planning and intra-day adjustment. In the day-ahead planning, a robust energy and reserve dispatch model is put forward considering continuous load/wind power and discrete generator state uncertainties, which can be solved efficiently by the combination of column-and-constraint generation (C&CG) algorithm and Karush-Kuhn-Tucker (KKT) optimality conditions. The optimal results provide robust operating plans for the intra-day dispatch. In the intra-day adjustment, the source-load deviations caused by various uncertainties can be compensated by the proposed rolling stochastic optimization (SO) model with quick-start units, which modulates the energy and reserve contributions of units and demand-side resource providers (DRPs). The superiority and validity of the proposed models are verified in case studies.