Optimized ANN–RF hybrid model with optuna for fault detection and classification in power transmission systems
Hasan Uzel, Yıldırım Özüpak, Feyyaz Alpsalaz, Emrah Aslan · 2025
Read Optimized ANN–RF hybrid model with optuna for fault detection and classification in power transmission systems by Hasan Uzel, Yıldırım Özüpak, Feyyaz Alpsalaz, Emrah Aslan (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-12795842.txt.
Description
Hasan Uzel and co-authors pair two learning methods to catch faults on power lines before they cascade. Their 2025 paper trains a hybrid of a neural network and a random forest on simulated current and voltage traces from an 11-kilovolt network. Fault types include single and double lines touching ground and a full three-phase short.
A tuning tool called Optuna searches for the best settings, and an oversampling step balances the rarest fault classes. The hybrid scores 99.8 percent accuracy, edging out several classic classifiers and two heavier deep models. The authors credit the network's pattern extraction, the forest's steady voting, and the tuning together. They caution that simulated data must still prove itself against real grid recordings before control rooms can rely on it.
Catalog record
| CREATOR | Hasan Uzel, Yıldırım Özüpak, Feyyaz Alpsalaz, Emrah Aslan |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 86 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |