Privacy-Preserving Economic Dispatch for An Active Distribution Network With Multiple Networked Microgrids

Microgrids are increasingly deployed and networked at the power distribution level in the transition toward an active distribution network (ADN) that is managed by a distribution system operator (DSO). This paper presents a decentralized economic dispatch approach for an ADN, when the DSO collaborates with microgrid central controllers (MGCCs) residing in a set of networked microgrids for optimizing the ADN’s energy management. This paper first develops a comprehensive economic dispatch model in the form of mixed-integer second-order cone programming (MISOCP) for ensuring the efficiency and security of the ADN’s operation. The proposed model takes full advantage of advanced control entities, such as on-load tap-changing transformers, static Var compensators, and remotely-controlled switches, while facilitating the plug-and-play participation of microgrids in providing energy and auxiliary services to the ADN. Next, generalized Benders decomposition is applied to solve the proposed MISOCP problem in a decentralized and iterative manner. After receiving power exchange requests from the DSO, MGCCs iteratively provide feedback in terms of Benders cuts without revealing their private operational information in order to help the DSO find the optimal dispatch decision. Moreover, a specific set of cutting planes are derived to ensure the tightness of the second-order cone relaxation of branch flows in the solution process. Last, case studies based on the modified IEEE 33-bus distribution system are conducted for evaluating the computational performance of the proposed approach. Numerical results have validated that the proposed approach has favorable efficiency, accuracy, and robustness, with great potential in the energy management of an ADN.