Open AccessLow‐loss image‐based compression for synchrophasor measurements
Author(s) -
Acharya Sowmya,
DeMarco Christopher L.
Publication year - 2020
Publication title -
iet generation, transmission and distribution
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.92
H-Index - 110
eISSN - 1751-8695
pISSN - 1751-8687
DOI - 10.1049/iet-gtd.2020.0551
Subject(s) - lossy compression , compression (physics) , data compression , computer science , phasor , volume (thermodynamics) , compression ratio , image compression , data compression ratio , lossless compression , transmission (telecommunications) , range (aeronautics) , transient (computer programming) , compressed sensing , electronic engineering , electric power system , real time computing , power (physics) , algorithm , engineering , artificial intelligence , image processing , telecommunications , materials science , image (mathematics) , automotive engineering , aerospace engineering , composite material , quantum mechanics , internal combustion engine , physics , operating system
Deployments of high‐sampling rate synchronised phasor measurement units (PMUs) are growing rapidly throughout the world, and with the advent of microPMUs, spreading from bulk transmission through distribution systems. The growing volume of PMU data presents challenges in its communication and storage, motivating consideration of compression algorithms. This study presents a novel lossy compression algorithm that exploits particular characteristics of power system measurements to improve the compression. Concepts successfully applied in image compression are tailored to the spatio‐temporal correlations induced between electrical quantities via their network interconnections. The quality of the resulting compression is judged on the balance of storage space savings versus the accuracy of data reconstruction. In representative real‐world and transient simulation datasets, the technique developed can provide storage compression in the range of 40:1 when different physical quantities are compressed together. The compression ratios can be in the range of 90:1 for voltage magnitudes and 190:1 for frequency when the measurements are compressed separately. The high‐compression ratios are achieved while maintaining low‐loss (high‐accuracy) reconstruction.