MASTER’S THESIS – GRAPH NEURAL NETWORKS FOR TOPOLOGY AWARE PREDICTIVE MAINTENANCE AND ENVIRONMENTAL MONOTORING IN EMERGING RENEWABLE ENERGY INFRASTRUCTURES – MICRO-HYDROPOWER DEVICES SWARMS (M/F/D)

fortiss is the research institute of the Free State of Bavaria for software-intensive systems and services with headquarters in Munich. The institute currently employs around 150 employees, who collaborate on research, development and transfer projects with universities and technology companies in Bavaria, Germany and Europe. Research is focused on state-of-the-art methods, techniques and tools of software development, systems & service engineering and their application to reliable, secure cyber-physical systems, such as the Internet of Things (IoT). fortiss has the legal structure of a non-profit limited liability company (GmbH). Its shareholders are the Free State of Bavaria (as majority shareholder) and the Fraunhofer Society for the Promotion of Applied Research. 

To extend our research lines in our Architectures and Services for Critical Infrastructures team, we are offering a: 

MSc Thesis on Graph neural networks for topology aware predictive maintenance, and environmental monitoring in emerging renewable energy infrastructures – micro-hydropower device swarms (m/f/d)

In recent years, the landscape of energy distribution systems has undergone a remarkable transformation. The modernization and diversification of the energy networks has come with severe challenges and opportunities, and the advent of smart grids have initiated an era of massive data generation and exchange in this domain. Data from several measuring devices distributed across the network are now available, and this together with the power of modern machine learning techniques presents opportunities for a more resilient, efficient, and sustainable energy infrastructure.

The aim of the project is to develop an automated process for robust and accurate predictive maintenance, diagnosis and environmental monitoring in an emerging form of renewable energy composed of a swarm of river hydropower plants. To this end, learning methods involving graph neural networks considering topological information, external knowledge sources must be researched. Our research relies on incorporating data from heterogeneous measuring devices at different locations and incorporate elements from previous work on fault localization in low and medium voltage energy grids.