Bidding strategy for virtual power plants with the day-ahead and balancing markets using distributionally robust optimization approach

Abstract

Virtual power plant (VPP) coordinates the energy consumption or production of its components and trades power in both day-ahead market (DAM) and balancing market (BM) to maximize operating margins, where consists of intermittent distributed generation, energy storage devices, and flexible demand. Due to the uncertainty of electricity prices and wind power output and imbalance penalties, VPP bidding is risky. Meanwhile, both traditional stochastic optimization (SO) and robust optimization (RO) algorithms have certain limitations and shortcomings in dealing with wind power output uncertainties. Therefore, a two-stage distribution robust optimization (DRO) model is proposed in this paper for determining the optimal bidding strategy for VPP participation in the energy market and combining L1norm with L∞norm to simultaneously constrain the confidence set of uncertain probability distributions. The column-and-constraint generation (CCG) algorithm is used to solve it. The robustness and feasibility of the proposed model are verified by a case study.

Keywords

• Distributionally robust optimization
• Virtual power plant
• Uncertainty