Joint Research Project “Coordinated Power Network Protection Based on Machine Learning Techniques”granted by DFG at FAU

Symbolic picture for the article. The link opens the image in a large view.

Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) is thrilled to announce the approval of a significant research project by the German Research Foundation (DFG). The project, entitled “Coordinated Power Network Protection Based on Machine Learning Techniques” (Koordinierter Netzschutz auf Basis maschineller Lernverfahren), is a pioneering venture in the field of electrical energy systems and machine learning.

Project Overview

The joint initiative, led by Professor Johann Jäger, Chair of Electrical Energy Systems (LEES), and Professor Andreas Maier, Chair of Computer Science 5 (Pattern Recognition Lab) at FAU, aims to revolutionize network protection in electrical supply networks. With a focus on high-voltage levels, the project will leverage machine learning and neural networks to automate and enhance fault detection, localization, and coordinated fault clearing.

Innovative Approach

The traditional protective devices in electrical networks will be replaced with specially trained agents, capitalizing on the non-linear classification, competency learning, and generalization capabilities of neural networks. This approach promises to significantly improve reliability and traceability in protective reactions, with options for both centralized and decentralized solutions.

Goals and Methodology

The project’s ambitious goals include:

  1. Generating training data to bridge the gap between simulation and reality.
  2. Developing a universal coordinated network protection solution.
  3. Creating a concept for traceability and reliability based on Known Operator Learning.
  4. Demonstrating and testing the developed solution in a real-time laboratory.

These objectives are supported by methods and insights gained from extensive preliminary work.

Funding and Support

The DFG has generously funded this project for a duration of three years. This grant will also support the doctoral studies of two PhD students, fostering the next generation of experts in this critical field.

Impact and Future Prospects

This project promises to set new standards in network protection technology, integrating seamlessly with digital switching stations and potentially communicating across decentralized agents. Its success will position FAU at the forefront of research in electrical energy systems and machine learning.

About FAU

FAU is a leading research institution in Germany, renowned for its interdisciplinary approach and cutting-edge research in engineering, medical technology, digital humanities, law, and economics.