2019/12/09 by Nico Hübner, Hübner, Nico, Yannick Rink +5
Computer Science · Engineering · #Distributed #FOS: Computer and information sciences #FOS: Electrical engineering #HVDC Systems and Fault Protection #Microgrid Control and Optimization #Optimal Power Flow Distribution #Parallel #Power System Optimization and Stability #Railway Systems and Energy Efficiency #Systems and Control (eess.SY) #and Cluster Computing (cs.DC) #cs.DC #cs.SY #eess.SY #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1912.03942
arxiv created 2019/12/09 · openalex publication_date 2019/12/09 · arxiv updated 2019/12/10 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
Distributed or multi-area optimal power flow appears to be promising in order to cope with computational burdens in large-scale grids and without the regional system operators losing control over their respective areas. However, algorithms are usually tested either in small test cases or single countries. We present a realistic case study in the interconnected European transmission grid with over 5000 buses and 315 GW load. The grid is partitioned into 24 areas which correspond to the respective single countries. We use a full alternating current model and integrate multi-terminal direct current systems. The decomposed problem is solved via a modified Alternating Direction of Multipliers Method (ADMM), in which the single countries only exchange border node information with each neighbor. In terms of generation costs, the solution of the distributed optimal power flow problem deviates only 0.02% from a centrally computed one. Consensus between all regions is reached before 200 iterations, which is remarkable in such a large system.