Machine-learning researcher at FAU Erlangen-Nürnberg building open benchmarks for ML-based power system protection.
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
2
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-15) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
Not yet scored: the underlying output data is insufficient for a stable measure. This axis would aggregate total works and their distribution across years and venues; the current record (h-index 2, i10-index 1) is too early-stage to score.
Impact—
Not yet scored: citation data is insufficient for this axis, which would measure citations and h-/i10-indices relative to field and career stage. The current record — h-index 2, i10-index 1 — is too thin for a reliable score.
Focus0/100
The focus axis scores 0/100, the profile's lowest. The record splits evenly between two adjacent field labels — 'electrical power systems / applied machine learning' and 'power systems / machine learning' at 0.5 each — and spans five topic clusters: power system optimization and stability (21), fault detection (20), protection (13), smart grid security (4) and power quality (2). Thematically the work reads as coherent; the score reflects this even split across a short, early record.
Collaboration53/100
Collaboration scores 53/100. The OpenAlex record lists 13 distinct coauthors across 20 works in the last three years, and both reviewed papers place Kordowich in a middle author position without corresponding status — consistent with team-based rather than solo research. Recurring partners include Julian Oelhaf, Siming Bayer, Andreas Maier and Johann Jäger, with additional coauthorship links to Paula Andrea Pérez-Toro, Tomás Arias-Vergara and Christian Bergler.
Activity89/100
Activity scores 89/100, the strongest axis in this profile. The OpenAlex counts-by-year series runs 1 (2022), 1 (2023), 10 (2025) and 10 (2026), i.e. 20 works within the recent three-year window. After a thin 2022-2023 start, output accelerated sharply from 2025 and held that pace into 2026, when both reviewed arXiv papers — EvEMTBench and the systematic ML fault-detection evaluation — appeared.
Reproducibility60/100
Reproducibility scores 60/100 as the mean over two Evolve-reviewed papers. EvEMTBench rates high for reproducibility readiness — an explicitly open, versioned benchmark spanning 4 grids and 24 tasks — while the systematic fault-detection evaluation rates low, with a single-topology PowerFactory setup, no deep-learning baselines and flagged possible cross-validation leakage. The average conceals a real spread between the two artifacts.
Research summary
Georg Kordowich is an early-career researcher at Friedrich-Alexander-Universität Erlangen-Nürnberg working at the interface of electrical power systems and applied machine learning, focused on machine-learning methods for fault detection, classification, localization and line identification in power grids. The profile is name-resolved from OpenAlex bibliographic data (ORCID 0000-0003-2225-7926). The bibliometric record is short but steep: one indexed work each in 2022 and 2023, then ten per year in 2025 and 2026 — twenty works within the recent three-year window — with an h-index of 2 and an i10-index of 1, meaning the citation footprint has not yet caught up with the recent acceleration. Topic frequencies concentrate on power system optimization and stability (21), power systems fault detection (20) and electrical fault detection and protection (13), with smaller activity in smart grid security and resilience (4) and power quality and harmonics (2). Evolve has reviewed two 2026 arXiv papers, both in middle-author positions: EvEMTBench, an open, versioned ML-protection benchmark covering 4 grids and 24 tasks, rated high for reproducibility readiness; and a systematic evaluation of ML fault detection reporting F1 0.991 within 10 ms windows, rated low because of a single-topology PowerFactory setup, absent deep-learning baselines and possible cross-validation leakage. Collaboration is team-based: 13 distinct coauthors, with Julian Oelhaf, Siming Bayer and Andreas Maier among the recurring partners.
Collaboration pattern
Team-based rather than solo: 13 distinct coauthors appear across 20 recent works (collaboration score 53/100), and both reviewed papers place Kordowich in a middle author position without corresponding status. Recurring partners include Julian Oelhaf and Siming Bayer, alongside Andreas Maier, Johann Jäger, Paula Andrea Pérez-Toro, Tomás Arias-Vergara and Christian Bergler — a stable, group-internal coauthor pool.
Research keywords
Highlights
EvEMTBench: an open, versioned benchmark for ML in power system protection
Systematic ML evaluation reports F1 0.991 in 10 ms fault windows
Timeline
- 2022First recorded OpenAlex output: 1 indexed work
- 2023Output stayed thin at 1 indexed work
- 2025Output accelerated to 10 indexed works
- 202610 works, including both Evolve-reviewed arXiv papers: EvEMTBench and the systematic ML fault-detection evaluation
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Bibliometric metrics come from an OpenAlex snapshot only; no field-normalized citation or Scopus/Web of Science data is included.
- Qualitative assessments draw only on the two 2026 arXiv papers reviewed by Evolve; other works in the OpenAlex record were not reviewed.
- The citation record is thin (h-index 2, i10-index 1), so the output and impact axes are left unscored rather than estimated.
- The supplied frequent-coauthor list contained duplicates and the researcher's own name; the collaboration note uses the deduplicated set of 7 external coauthor names.