Julian Oelhaf, M. Sc.

Researcher

Department of Computer Science
Chair of Computer Science 5 (Pattern Recognition)

Room: Room 09.157
Martensstr. 3
91058 Erlangen

About:
I am a PhD researcher at FAU Erlangen–Nürnberg working on machine-learning-based methods for
power system protection and fault management.
My research focuses on improving how faults are detected, classified, localized, and cleared in
modern electricity grids with high shares of renewables and power electronics.

I enjoy bridging theory and practice: my work combines high-fidelity simulation,
large-scale data analysis, and real-time laboratory validation, and I collaborate closely with
industry partners to translate AI research into deployable grid solutions.

Focus Areas

  • Fault Management: smarter detection, classification, and localization of grid faults
  • Protection & Restoration: AI-driven strategies to speed up response and recovery after disturbances
  • Digital Twins & Data: large-scale simulations and data pipelines for testing new protection approaches
  • Anomaly Detection: modern AI methods to identify unusual events and improve system monitoring

Approach & Tools

  • Machine Learning & AI (deep learning, reinforcement learning, transformers)
  • Handling complex time-series data at scale with reproducible pipelines
  • Simulation and real data: PowerFactory EMT, PMU/SCADA, and digital twins

Collaboration Opportunities

  • Joint pilot projects with utilities, TSOs/DSOs, and technology providers
  • Benchmarking and validation of new AI-based protection methods
  • Advisory and knowledge transfer on ML adoption for grid applications

  • Since 2024:
    PhD Student at the Pattern Recognition Lab, FAU Erlangen-Nürnberg
  • 2021 – 2024
    Master’s degree in Computer Science, FAU Erlangen-Nürnberg
  • 2017 – 2020
    Bachelor’s degree in Aerospace Computer Science, JMU Würzburg

2024

  • Coordinated grid protection based on machine learning methods

    (Third Party Funds Single)

    Project leader: , ,
    Term: July 1, 2024 - June 30, 2027
    Acronym: Netzschutz-KI
    Funding source: DFG-Einzelförderung / Sachbeihilfe (EIN-SBH)

2026

Unpublished Publications

2025

Journal Articles

Conference Contributions

Type Title Status
MA thesis AI for Smart Grids: Sensor-Efficient Fault Localization in High-Voltage Power Systems open
MA thesis Known Operator Learning for Fault Localization in Electric Power Grids running
MA thesis Federated Learning for Local Fault Analysis in Power Systems open
MA thesis Deep Learning for Fault Localization in High-Voltage Power Grids open
Project Universal and Relay-Generalizable Machine Learning for Protection in Power Grids open
Project Machine Learning for Cyber-Physical Event Detection in Smart Grids running
MA thesis Parameter Efficient Finetuning of Universal Time Series Transformers for Energy Forecasting running
Project Conventional vs. Reinforcement Learning–Based Relays for Power System Protection running
BA thesis Classification of the State of Electrical Contacts of Circuit Breakers with Explainable Artificial Intelligence running
Project Evaluating Time-Frequency Representations for Intelligent Fault Analysis in Power System Protection running
Project Multi-Task Learning for Integrated Fault Analysis in Power System Protection running
MA thesis Reinforcement Learning for Adaptive Protection in Power Grids finished
Project Reproducible Reinforcement Learning on a Real-World Power Grid Control Problem running
Project Reinforcement Learning for Centralized Fault Coordination in Power Systems finished
Project Transformer-Based Forecasting Model for Fault Detection in Power System Protection finished
MA thesis Advanced Machine Learning-Based High Demand Forecasting of Household Energy Consumption for Enhancing Grid Operations finished
MA thesis Deep Learning-Based Fault Detection and Classification in Power System Protection: A Comparative Study finished
MA thesis Wind Power Forecasting through Probabilistic Machine Learning Models finished

 

AI Grid Protection – Coordinated Power System Protection using Machine Learning