Abstract
Electric utilities in California have historically been linked to up to 10% of wildfires. To mitigate this risk, Southern California Edison has invested significantly in wildfire prevention strategies, including undergrounding cables and enhancing equipment inspections. This article explores a novel approach to fire prevention by detecting anomalies in the distribution system that may indicate potential fire hazards. The focus is on identifying arcing conditions through high-resolution point-on-wave (POW) measurements. Arcing, a precursor to fires, is challenging to detect due to its subtle transients and complex system topology. The article discusses the use of advanced signal processing and machine learning techniques, such as spectral correlation function and discrete wavelet transform, to extract features from POW data and accurately identify arcing events. The study demonstrates a high accuracy rate in detecting arcing, paving the way for improved fire prevention measures in electric distribution systems.
| Original language | English |
|---|---|
| Pages (from-to) | 83-90 |
| Number of pages | 8 |
| Journal | IEEE Power and Energy Magazine |
| Volume | 22 |
| Issue number | 6 |
| DOIs | |
| State | Published - 2024 |
Funding
This work was partly performed under the auspices of the DOE by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. This article has been authored by UT-Battelle, LLC, under Contract DE-AC05- 00OR22725 with the DOE. The publisher acknowledges the U.S. government license to provide public access under the DOE Public Access Plan (https://energy.gov/doe-public -access-plan).