Abstract:

In order to cope with the uncertainties brought by the growing scale of renewable energy, this paper proposed a distributed optimization for virtual power plant (VPP) considering uncertainties. On the VPP side, a scenario-based stochastic optimization model was used to depict the stochastic characteristics of photovoltaic and wind power output. On the distribution network side, considering the uncertainty of electricity prices when purchasing power from the grid, a two-stage robust optimization model and an opportunity model based on information gap decision theory were established to enhance the model's robustness against electricity price fluctuations or to capture potential benefits amidst volatile electricity prices. The problem was solved distributedly between the VPP and the distribution network through the alternating direction multiplier method algorithm to protect their privacy. The results show that the proposed model is more flexible than the robust model and can better balance the risks and benefits under uncertain conditions. The distributed algorithm has a good convergence performance and can better utilize computing resources compared to the centralized algorithm.

Key words: virtual power plant, distribution network, tradable green certificate market, uncertainty, distributed optimization