Decentralized Energy Management for Networked Microgrids in Future Distribution Systems

An increasing amount of distributed energy resources is being integrated into both distribution systems (DSs) and networked microgrids (MGs). The coordination among DSs and MGs becomes essential for DS operators and MG operators considering the high uncertainties of renewable energies and load demands. In this paper, we propose a decentralized energy management framework to coordinate the power exchange between DS and MGs based on the alternating direction method of multipliers algorithm in a fully decentralized fashion. The energy management model in each entity (DSs or MGs) is formulated using two-stage robust optimization to address the uncertainties of renewables and load demands. The problem is treated as a second order cone programming one based on a relaxed distflow model. Moreover, the robust model is solved by column and constraint generation algorithm, where cutting planes are introduced to ensure the exactness of second-order cone relaxation. Numerical results on a modified IEEE 33-bus system with three MGs validate the effectiveness of the proposed method.