Using synthetic data for pretraining partial discharge detection in overhead transmission lines
Lukáš Klein, Jan Fulneček, Ondřej Kabot, Jiří Dvorský, Lukáš Prokop · 2025
Read Using synthetic data for pretraining partial discharge detection in overhead transmission lines by Lukáš Klein, Jan Fulneček, Ondřej Kabot, Jiří Dvorský, Lukáš Prokop (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12749304.txt.
Description
Lukas Klein and co-authors use synthetic data to train detectors of partial discharge in power lines in this 2025 paper. Partial discharge is a small electrical fault that warns of damage ahead. Finding it early prevents costly outages. Real labeled data is scarce, and radio interference is common on overhead lines.
The team built a simulator that varies fault rate, signal size, vegetation contact, and noise. Networks were pretrained on the synthetic signals, then fine tuned on a small set of real measures. Several network types were compared on signal images. Convolutional networks learned the task well and beat the other types. The study maps which synthetic settings help most, so future training sets can be built with care.
Catalog record
| CREATOR | Lukáš Klein, Jan Fulneček, Ondřej Kabot, Jiří Dvorský, Lukáš Prokop |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 161 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |