Data-driven robust dispatch for integrated electric-gas system considering the correlativity of wind-solar output

Abstract The increasing popularity of distributed renewable power generation represented by wind turbines and solar plants and the deployment of gas turbines (GTs) and power-to-gas (P2G) facilities have promoted stronger interdependence between the power grid and the natural gas system, which makes the economic dispatch of the integrated electrical and gas system (IEGS) more challenging. This paper proposes a two-stage dispatch model for distribution-level IEGS based on a data-driven robust optimization (DDRO) method. First, for distribution-level IEGS, a network model of IEGS is established, and the nonconvex constraints are relaxed by the big M method and second-order cone (SOC) relaxation. Considering the correlation between wind and solar output in multiperiod, a wind-solar output ellipsoid uncertain set is constructed through the minimum volume enclosed ellipsoid (MVEE) algorithm to obtain extreme scenarios. Finally, a column-and-constraint generation (C&CG) method based on extreme scenarios is proposed to solve the two-stage robust optimization model. Simulation results show that the proposed DDRO method can reduce the day-ahead and real-time dispatch costs of energy conversion equipment while ensuring the robustness of IEGS system dispatch. In addition, the economics of IEGS system dispatch is improved.