%0 Journal Article
%@ 1932-8184
%A Yang, B
%A Katsaros, KV
%A Chai, WK
%A Pavlou, G
%D 2018
%F discovery:1523757
%J IEEE Systems Journal
%K Phasor measurement units, Power grids, Delays, Network topology, Topology, Monitoring, State estimation
%N 1
%P 948-958
%T Cost-Efficient Low Latency Communication Infrastructure for Synchrophasor Applications in Smart Grids
%U https://discovery.ucl.ac.uk/id/eprint/1523757/
%V 12
%X With the introduction of distributed renewable energy resources and new loads, such as electric vehicles, the power grid is evolving to become a highly dynamic system that necessitates continuous and fine-grained observability of its operating conditions. In the context of the medium voltage (MV) grid, this has motivated the deployment of phasor measurement units (PMUs), that offer high-precision synchronized grid monitoring, enabling mission-critical applications such as fault detection/location. However, PMU-based applications present stringent delay requirements, raising a significant challenge to the communication infrastructure. In contrast to the high voltage domain, there is no clear vision for the communication and network topologies for the MV grid; a full-fledged optical fiber-based communication infrastructure is a costly approach due to the density of PMUs required. In this study, we focus on the support of low-latency PMU-based applications in the MV domain, identifying and addressing the tradeoff between communication infrastructure deployment costs and the corresponding performance. We study a large set of real MV grid topologies to get an in-depth understanding of the various key latency factors. Building on the gained insights, we propose three algorithms for the careful placement of high capacity links, targeting a balance between deployment costs and achieved latencies. Extensive simulations demonstrate that the proposed algorithms result in low-latency network topologies while reducing deployment costs by up to 80% in comparison to a ubiquitous deployment of costly high capacity links.
%Z This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.