Decentralized Implementation of Unit Commitment with Analytical Target Cascading: A Parallel Approach
This paper presents a decentralized solution algorithm for network-constrained unit commitment (NCUC) in multiregional power systems. The proposed algorithm is based on our previous work in which a local NCUC was formulated for each control entity (i.e., region) and an analytical target cascading (ATC) based distributed but partially parallelized algorithm requiring a central coordinator was presented. The primary objective of this paper is to present a decentralized approach that relaxes the need for any form of central coordinator in ATC and allows fully parallelized solutions of the local NCUCs. To achieve this objective, we formulate a bilevel optimization problem for each control entity. While the upper level solves the NCUC problem of the control entity, the lower level seeks to coordinate the control entity with its neighboring regions. The lower level is a convex optimization, which can be further replaced in the upper level problem by the Karush-Kuhn-Tucker conditions. The control entities communicate directly with each other and synchronously solve their local NCUCs. Having no need for any form of central coordinator, the proposed algorithm is potentially less vulnerable to cyber-attacks and communication failures than the distributed methods utilizing a coordinator.
Publication Source (Journal or Book title)
IEEE Transactions on Power Systems
Kargarian, A., Mehrtash, M., & Falahati, B. (2018). Decentralized Implementation of Unit Commitment with Analytical Target Cascading: A Parallel Approach. IEEE Transactions on Power Systems, 33 (4), 3981-3993. https://doi.org/10.1109/TPWRS.2017.2787645