Distributionally robust day-ahead operation of power systems with two-stage gas contracting. (15th September 2021)

• Record Type: Journal Article
• Authors: Sayed, Ahmed Rabee; Wang, Cheng; Chen, Sheng; Shang, Ce; Bi, Tianshu
• Is Part Of: Energy. Volume 231(2021)
• Publication Date: 2021-09-15
• Subjects: Day-ahead operation -- Distributionally robust optimization -- Gas systems -- Gas contracting -- Two-stage decision-making; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2021.120840
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Due to stronger interactions between power and gas systems promoted by the deployment of fast-response gas-fired power plants and the advanced technology of power-to-gas facilities, secure and optimal energy management becomes highly challenging. In industrial practice, the two systems are operated by different regulatory authorities, which is commonly neglected in the existing studies, therefore, information exchange is limited, and bilateral energy transactions are vital. This paper revisits the day-ahead operation of the coupled power and gas system with large-scale wind power generation integration, where a two-stage gas contracting mechanism is proposed to incorporate cost-effective day-ahead and rarely-event real-time gas contracts. To balance the robustness and the conservativeness of the operation strategy, a distributionally robust optimization (DRO) based decision-making framework is derived, which effectively handles the proposed mechanism. The intractable DRO model is reformulated to be solved by the nested column and constraint generation algorithm. The nonconvex gas flow equations are approximated as convex ones, which are solved by a difference-of-convex programming-based procedure to find feasible solutions. Simulation results demonstrate the benefits of the DRO and two-stage contract modeling. Highlights: A two-stage bidirectional gas contracting mechanism is proposed. A distributionally robust dispatch model is formulated for power system operation. A tractable convex reformulation is obtained and solved by the NC&CG algorithm. DC programming-based procedure is designed for feasibility guarantee.