Back-feed power restoration using distributed constraint optimization in smart distribution grids clustered into microgrids

M. Zaki El-Sharafy, Hany E.Z. Farag

    Research output: Contribution to journalArticlepeer-review

    24 Scopus citations


    In this paper, an optimization problem is formulated for the automatic back-feed service restoration in smart distribution grids. The formulated problem relies on the structure of smart distribution grids, clustered into multi-microgrids, capable of operating in both grid-connected and islanded modes of operation. To that end, three types of power transfer between the neighboring microgrids, during the restoration processes are introduced: load transfer, distributed generation (DG) transfer, and combined load–DG transfer. The formulated optimization problem takes into account the ability of forming new, not predefined islanded microgrids, in the post-restoration configuration, to maximize service restoration. To obviate the need for a central unit, the optimization problem is reformulated, in this work, as a distributed constraint optimization problem, in which the variables and constraints are distributed among automated agents. To reduce the problem complexity, the restoration problem is decomposed into two sequential and interdependent distributed sub-problems: supply adequacy, and optimal reconfiguration. The proposed algorithm adopts the Optimal Asynchronous Partial Overlay (OPTAPO) technique, which is based on the distributed constraint agent search to solve distributed sub-problems in a multi-agent environment. Several case studies have been carried out to evaluate the effectiveness and robustness of the proposed algorithm.

    Original languageBritish English
    Pages (from-to)1102-1117
    Number of pages16
    JournalApplied Energy
    StatePublished - 2017


    • Distributed constraint optimization
    • Distributed generation
    • Microgrids
    • Restoration
    • Smart distribution grids


    Dive into the research topics of 'Back-feed power restoration using distributed constraint optimization in smart distribution grids clustered into microgrids'. Together they form a unique fingerprint.

    Cite this