Feature Selection for Fault Prediction in Distribution Systems

Kordowich G, Oelhaf J, Bayer S, Maier A, Kereit M, Jäger J (2026)

Publication Language: English

Publication Status: Accepted

Publication Type: Journal article, Original article

Future Publication Type: Journal article

Publication year: 2026

Journal

Book Volume: 261

Article Number: 113498

DOI: 10.1016/j.epsr.2026.113498

Abstract

While conventional power system protection isolates faulty components only after a fault has occurred, fault prediction approaches try to detect faults before they can cause significant damage. Although initial studies have demonstrated successful proofs of concept, development is hindered by scarce field data and ineffective feature selection. To address these limitations, this paper proposes a surrogate task that uses simulation data for feature selection. This task exhibits a strong correlation (r = 0.92) with real-world fault prediction performance. We generate a large dataset containing 20,000 simulations with 34 event classes and diverse grid configurations. From 1556 candidate features, we identify 374 optimal features. A case study on three substations demonstrates the effectiveness of the selected features, achieving an F1-score of 0.80 and outperforming baseline approaches that use frequency-domain and wavelet-based features.

Authors with CRIS profile

Related research project(s)

Involved external institutions

Siemens AG Germany (DE)

How to cite

APA:

Kordowich, G., Oelhaf, J., Bayer, S., Maier, A., Kereit, M., & Jäger, J. (2026). Feature Selection for Fault Prediction in Distribution Systems. Electric Power Systems Research, 261. https://doi.org/10.1016/j.epsr.2026.113498

MLA:

Kordowich, Georg, et al. "Feature Selection for Fault Prediction in Distribution Systems." Electric Power Systems Research 261 (2026).

BibTeX: Download