A hierarchical approach to automated identification of anomalous electrical waveforms

Power utilities employ "smart'' field devices capable of digitally recording electrical waveforms. The relationship between events and their recorded waveforms can be exploited for characterization of the power grid’s state over any period of time and facilitating the impact electrical disturbances have on equipment, subsystems, and systems. Over a period of one month, these devices record approximately 2,000 electrical disturbance waveforms. Currently, analysis of these waveforms is conducted using by-hand approaches; thus, severely limiting the analysis to roughly 2%. The analysis is done hours to days after the events occurred, which negates informed, timely corrective actions. This document presents an automated hierarchical approach capable of identifying specific events using the electrical disturbance waveforms stored using COMmon format for TRAnsient Data Exchange (COMTRADE) files. The developed approach processes a single file in 1.8 seconds and has demonstrated successful identification of 140 events with a success rate of 91%.

Grantor dc:publisher

University of Tennessee at Chattanooga

Year dc:date.available

2020

Author dc:creator

• Wilson, Aaron

Contributors dc:contributor

• Reising, Donald R.
• Loveless, Thomas D.; Karrar, Abdelrahman A.; Hay, Robert W.
• College of Engineering and Computer Science

• Electric circuit analysis
• Electric power systems
• Electric power distribution

Licence

• https://rightsstatements.org/page/InC/1.0/?language=en

Language dc:language

English, eng

Repository record dc:identifier

https://scholar.utc.edu/theses/596

OAI identifier oai:identifier

oai:scholar.utc.edu:theses-1748