Machine Learning for Cyber-Physical Event Detection in Smart Grids

Type: Project

Status: running

Date: March 1, 2026 - July 31, 2026

Supervisors: Julian Oelhaf

Machine Learning for Cyber-Physical Event Detection in Smart Grids

Modern power grids combine electrical infrastructure with communication and control systems. This creates new risks: cyber attacks can mimic real grid disturbances. In this project, you will use machine learning to distinguish normal operation, faults, and cyber events using publicly available industrial control system datasets.

What you’ll do

  • Reproduce a published ML pipeline (literature-based baseline) [1][2]
  • Train and evaluate classifiers for event detection
  • Analyze explainability (e.g., feature importance / SHAP-style reasoning)
  • Extend the baseline (robustness, new models, better evaluation)

What you’ll learn

  • Applied machine learning on real-world energy system data
  • Smart grid cybersecurity basics (ICS / OT perspective)
  • Reproducible research workflows and evaluation

Requirements

  • Solid Python skills and basic knowledge of machine learning (e.g., supervised learning, evaluation metrics)
  • Power systems or cybersecurity knowledge is helpful but not required.
  • Able to attend the weekly in-person meeting in Erlangen (Tuesdays, 14:00)

Project type

Bachelor / Master project (research-oriented)
Start date: flexible (by arrangement)

Apply

Send one PDF to julian.oelhaf@fau.de with the subject:

Application | Project (10 ECTS) | ML for Cyber-Physical Event Detection | <Your Full Name>

Email body (max. 200 words):
Short motivation and your earliest start date.

Attach as one PDF:

  • CV
  • Transcript of records (dated)
  • Optional: links to code projects (GitHub, portfolio, etc.)

📌 Incomplete applications will not be considered.

[1] Farsi, M., Alwateer, M., Alsaedi, S.A. et al. Detection of disturbances and cyber-attacks in smart grids using explainable machine learning. Sci Rep (2026). https://doi.org/10.1038/s41598-026-35449-x

[2] Beaver, Justin M., Borges-Hink, Raymond C., Buckner, Mark A., “An Evaluation of Machine Learning Methods to Detect Malicious SCADA Communications,” in the Proceedings of 2013 12th International Conference on Machine Learning and Applications (ICMLA), vol.2, pp.54-59, 2013. doi: 10.1109/ICMLA.2013.105