A two-stage decomposition method for integrated optimization of islanded AC grid operation scheduling and network reconfiguration. (March 2022)

• Record Type: Journal Article
• Authors: Damgacioglu, Haluk; Celik, Nurcin
• Is Part Of: International journal of electrical power & energy systems. Volume 136(2022)
• Publication Date: 2022-03
• Subjects: Grid operation scheduling -- Network reconfiguration -- Radial networks -- Mixed integer programming; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2021.107647
• Languages: English
• ISSNs: 0142-0615

Abstract

Abstract: A restructuring of electricity systems promises energy resilience, prosperity, and sustainability for communities. In particular, Reconfigurable networks, such as distribution systems or microgrids, as smart electricity systems are more controllable and flexible compared to traditional systems. In this study, we integrate network reconfiguration decisions into a multi-period grid operation scheduling problem to minimize the operational costs while taking into account both technical and inter-temporal constraints for distributed energy resources, physical constraints on alternating current (AC) network, and system reliability constraints. We develop a comprehensive optimization model for the grid operation planning problem by taking into account the operation scheduling and economic dispatch decisions from earlier cycles and grid operational constraints while determining the optimal network topology. Due to the NP-hard nature of the problem, we introduce a two-stage decomposition algorithm where the solution of the operation scheduling problem (without switching variables) feeds the network reconfiguration problem embedding the dispatch problem for each time period. The performance of the proposed two-stage decomposition algorithm is demonstrated on three different available IEEE testbeds against CPLEX and COUENNE solvers for benchmarking purposes. The selected IEEE testbeds have been widely adopted by previous literature for the aforementioned grid operation planning problem due to their capability in reproducibility. The proposed decomposition algorithm outperforms CPLEX in terms of both solution quality and computational times, particularly our algorithm provides 0.5% and 3.17% better solutions in mid-size and large-size networks, respectively. Highlights: We propose MILP for grid scheduling problem in active distribution networks. The proposed model integrates grid operation planning and network topology decisions. We develop a two-stage decomposition algorithm to solve the problem. We demonstrate the performance the proposed algorithm on three separate test-beds.