Stochastic look-ahead dispatch for coupled electricity and natural-gas networks

The increasing employment of natural-gas fired generators for balancing intermittent renewable energy sources has greatly strengthened the interdependency between electricity and gas networks. This work therefore proposes two-stage stochastic look-ahead dispatch for coupled electricity and gas networks, which includes first-stage dispatch for a base case and second-stage regulation dispatch for each representative scenario. An enhanced second-order cone (SOC) relaxation of the dynamic gas flow equations is applied to model the optimization problem as SOC programming (SOCP), and difference-of-convex programming (DCP) is deployed to ensure the feasibility and optimality of the solution provided by SOCP. Moreover, the alternating direction method of multipliers (ADMM) is employed to reduce the computational complexity by decomposing the scenario-based stochastic optimization problem. Test results on a modified IEEE 39-node system and a Belgian 20-node gas system verify the effectiveness of the enhanced SOCP, DCP, and ADMM approaches. In addition, comparison results between the deterministic dispatch and the proposed stochastic dispatch demonstrate the value of stochastic solution.