Robust stochastic optimal dispatching method of multi-energy virtual power plant considering multiple uncertainties.

• Propose an optimal scheduling method to deal with multi-energy and multi-uncertainty. • Combine deep learning and clustering algorithm to deal with load-side uncertainty. • Adjusting robust coefficients can obtain scheduling schemes under different risk. • The proposed method can reduce the overall scheduling cost. In recent years, with the rapid development of the energy Internet and the deepening of the complementary coupling of various energy sources, the concept of multi-energy virtual power plant comes into being. At the same time, insufficient research on optimal scheduling of multi-energy virtual power plants under multiple uncertainties. Here we propose a robust stochastic optimal dispatching method to solve the scheduling problem under multiple uncertainties. For the source side uncertainties, the uncertain set of cardinalities with a robust adjustable coefficient is adopted to describe the output of wind turbines and photovoltaics. For the load side uncertainties, the Wasserstein generative adversarial network with gradient penalty is used to generate electric, thermal, cooling, and natural gas load scenarios, and the K-medoids clustering is used to get typical scenes. A two-stage robust stochastic optimal model of the min-max-min structure was established. Based on the dual transformation theory and the column constraint generation algorithm, the original model was solved alternately. Finally, the effectiveness of the proposed model and algorithm is verified by simulation analysis. The proposed method can get the scheduling scheme with the lowest operating cost in the worst scenario and is conducive to reducing the overall scheduling cost of the system. [ABSTRACT FROM AUTHOR]