Index

[MS project] Reinforcement Learning for 110 kV Distribution Grid Restoration in Blackout situation

Background

Restoring power in a 110 kV distribution grid after a blackout involves complex, sequential decisions under strict operational constraints (e.g. voltage, frequency). Traditional rule-based approaches lack flexibility for unexpected scenarios and operational experience.

Challenge

To support operators in real time, reinforcement learning must evaluate safe, interpretable actions within milliseconds. Key requirements include integration with a pre-existing grid restoration simulator (including test grid) built into PowerFactory, as well as strict adherence to stability limits.

Tasks

– Model a reinforcement learning environment for the restoration process

– Incorporate operational constraints

– Implement and train an RL agent (e.g., DQN, PPO)

– Evaluate agent performance (success rate, stability violations)

– Visualize and interpret agent decisions for transparency

– (Optional) Integrate safety mechanisms (shielded learning)

– (Optional) Benchmark inference speed for real-time GridAssist

Requirements

– Basic knowledge in electrical engineering

– Understanding of electrical power systems (PowerFactory)

– Advanced programming skills (Python)

– (basic/advanced) knowledge of reinforcement learning

– Fluent in English

Start: July 15th
End: Jan 15th
Type: Master research project
Language: English
Contact: Changhun Kim(changhun.kim@fau.de), Simon Linnert(simon.linnert@fau.de)

Application:
Please apply by email with the subject line “[RL-Restoration Project Application 2025]”.
Include your CV and transcript of records (grade overview). Applications without these documents will not be considered.
In the body of the email, briefly describe (approx. 100 words) why you are interested in this specific project and how your background prepares you for it.

References:Safe_RL_Power_Grid_restoration

[1] X. Chen, Y. Xu, and P. Zhang, “Deep reinforcement learning for distribution system restoration with DER coordination,” IEEE T. Smart Grid, vol. 13, no. 2, pp. 987-999, 2022.
[2] J. Ding, H. Wang, and S. Low, “Safe policy gradient for microgrid black-start restoration,” in Proc. IEEE PES GM, 2024.
[3] H. Liu et al., “Explainable reinforcement learning: A survey,” ACM Comput. Surveys, vol. 55, no. 7, pp. 1-38, 2023.
[4] R. Li, T. Liu, J. Yu et al., “Graph neural network based voltage-control reinforcement learning for distribution systems,” IEEE T. Smart Grid, vol. 12, no. 6, pp. 5269-5280, 2021.
[5] A. Molina García et al., “Switching impact on MV equipment-a sixyear field study,” CIRED Workshop, 2020.
[6] S. Ross, G. Gordon, and D. Bagnell, “A reduction of imitation learning and structured prediction to no-regret online learning,” in AISTATS, 2011.
[7] J. Achiam, D. Held, A. Tamar, and P. Abbeel, “Constrained policy optimization,” in Proc. ICML, 2017.
[8] M. Alshiekh, R. Bloem, R. Ehlers et al., “Safe reinforcement learning via shielding,” in AAAI, 2018.
[9] M. Du, N. Liu, Q. Hu et al., “Techniques for interpretable deep learning,” Commun. ACM, vol. 63, no. 1, pp. 68-77, 2020.

Interpretable Vision Transformers with Attention Maps for Phonological Precision Assessment from MRI

[Master Project] Machine Learning for Fault Localization and Classification in Power Grids

Context:
As electrical grids evolve to accommodate renewable energy and decentralized generation, accurate and fast fault analysis becomes increasingly critical. This project investigates machine learning (ML) methods to perform two key protection tasks using high-frequency voltage and current waveforms:

Fault classification (e.g., three-phase, line-to-ground)
Fault localization (estimating the exact distance to the fault along a line)

ML offers a promising alternative to traditional protection schemes, especially under real-world constraints like noise, missing data, and rapidly changing grid topologies.

Project Goals:

Implement and evaluate ML models for fault type classification and localization using multivariate time series.
Benchmark model performance (accuracy, F1, runtime) across fault types and window lengths (10–50 ms).
Optional: Analyze robustness to incomplete or noisy measurements.

What’s Provided:

A pre-generated, realistic simulation dataset with transient fault signals and ground truth labels.
Modular Python pipeline for data processing and model evaluation.
Close supervision and guidance from researchers in ML and power systems.

Requirements (strict):

Solid foundation in machine learning (e.g., coursework or projects involving classification/regression).
Strong Python programming skills, especially with ML libraries such as scikit-learnand PyTorch,
Experience with or strong interest in time series, signal processing, or power systems.
High level of independence and reliability – this is a hands-on project with a fixed timeline and deliverables.

Start: July 1st or July 15th
End: September 1st or September 15th
Type: Master research project
Language: English or German
Contact: Julian Oelhaf

Application:
Please apply by email with the subject line “[Fault-ML Project Application 2025]”.
Include your CV and transcript of records (grade overview). Applications without these documents will not be considered.
In the body of the email, briefly describe (approx. 100 words) why you are interested in this specific project and how your background prepares you for it.

Reinforcement Learning for Centralized Fault Coordination in Power Systems

Latent Space Modeling for Event Detection in Power Grid Data

This project explores how latent representations learned from raw grid waveforms can reveal underlying structure and enable early detection of abnormal events. By modeling high-frequency voltage and current signals, we aim to distinguish critical disturbances from normal behavior with minimal delay.

Report Generation in pathology using WSIs

This project focuses on developing methods for processing large-scale digital pathology datasets and extracting meaningful features from whole slide images to support automated report generation. Emphasis is placed on efficient handling of gigapixel image data and preparing it for use in vision-language models for clinical applications.