I work as an Associate Professor at the Electrical Engineering, Mathematics and Computer Science Faculty at the Delft University of Technology. I also conduct research at the Delft AI Energy Lab, work as a principal scientist at the Austrian Institute of Technology in the Center for Energy, and co-direct (non-executive) the Canadian organization Student Energy. Before that, I worked on Machine Learning applications for power systems at Imperial College London, Operations Research and Applied Control at Carnegie Mellon University and the Massachusetts Institute of Technology (MIT). During my studies, I worked in the chemical and energy industry, in China and Germany. I hold an M.Sc. in Chemical Engineering, a B.Sc. in Electrical Engineering, and a B.Sc. in Mechanical Engineering from RWTH Aachen University, Germany. I am an active member of the IEEE PES Taskforces and CIGRE C2 working groups. I also serve as Associate Editor for the IEEE PES Transactions on Power Systems and on several technical programme committes like for the Power System Computing Conference.

Aside from these commitments, I enjoy playing squash, running, hiking, going to the sauna, sailing, skiing in the Alps, travelling to visit friends and family, and playing video games.

Delft AI Energy Lab

Our research develops fundamental methods based on AI and machine learning that lead to new use cases in energy systems ranging from demand response to distributed real-time control, as well as centralized coordinated operations in real time. My team develops novel algorithms that can process substantial amounts of data and advancing energy system operations from societal, sustainable, and economic perspectives. Explore more about Delft AI Energy Lab

Research

My vision is towards a fully data-driven operating paradigm of the control centres of the future, in distribution and transmission. This does not mean neglecting physical models but to include those as part of the “data” (others might call it hybrid or physics-informed approaches). Artificial Intelligence represents an alternative to successfully handling complex systems, particularly in images, molecular graphs, and human communication. The future electricity system requires fast (control and monitoring) actions and new foundations addressing the complexity of the operation of complex power systems. Our team develops either general AI-based models that can operate multiple (operating) tasks at once, and specialised AI-based models that can operate just one task but outperform conventional tooling. Eventually, my vision is that these will merge towards a system of AI in the control room. To achieve this vision my research focuses on the reliability, security and resiliency of power system operation that requires monitoring and controlling distribution and transmission systems. My team approaches these applications by developing new computational methods from machine learning (our focus) but also operations research and control that are suited to this application area. Our approach and transferable expertise involve learning and rapidly developing new computational methods to unlock new use cases in this field. To realise this, our team combines “domain” expertise on these applications with a deepened understanding of machine learning techniques.

Explore more about Publications

Team

Our interdisciplinary team combines specialists in data science, operations research, and electrical engineering. We are part of the Delft AI Energy Lab which I co-direct with Peyman Mohajerin Esfahani (based at TU Delft and University of Toronto). Structurally, we are within the Intelligent Electrical Power Grid section in the Department of Electrical Sustainable Energy that Peter Palensky heads. Currently, my team members are

• Olayiwola Arowolo, topic “Graph-based Learning for Electromagnetic Transients”, TU Delft
• Paul Bannmüller, topic “Reinforcement Learning in Topology Control”, TU Delft
• Périne Cunat, topic “Modelling Generating Alternatives for Heating and Electricity Networks”, TU Delft, Austrian Institute of Technology
• Demetris Chrysostomou, topic “Estimating Flexibility of Distribution Grids at TSO/DSO interface”, TU Delft
• Luca Hofstadler, topic “Quantum Computing for Transmission Systems”, TU Delft, Austrian Institute of Technology
• Mert Karaçelebi, topic “Neural Networks for Dynamic Security Assessment” TU Delft
• Betül Mamudi, topic “Machine Learning for State Estimation”, TU Delft, Alliander
• Basel Morsy, topic “Reinforcement Learning for Topological Reconfigurations”, TU Delft, Austrian Institute of Technology
• Ali Rajaei, topic “System-theoretic Machine Learning for Congestion Management” TU Delft
• Jochen Stiasny, topic “Physics-Informed Machine Learning for Power System Dynamics”, TU Delft
• Haiwei Xie, topic “Game Theory and Control of Distributed Energy”, TU Delft
• Runyao Yu, topic “Machine Learning for Outliers in Grids Operations and Intraday Energy Markets’, TU Delft, Austrian Institute of Technology
• Viktor Zobernig, topic “Trading Strategies in Redispatch Markets”, TU Delft, Austrian Institute of Technology

Possibilities to join my team as PhD researcher

We offer an excellent environment from tailored education to key background and competencies to conduct research in the area described above on a high-scientific quality. Additionally, we offer a competitive salary package to all our PhD researchers. We expect our new team members to have an excellent background in related domains (computer science, power systems, control or electrical engineering or related education background). We strongly encourage you to get in touch if you bring in this background on an excellent level, for example being among the top 10% among your peer students, and/or have already publications that demonstrate your research ambitions and previous performance, having won a merit-based award(s). If you meet these criteria, please note that diversity is a key priority for our team! Diverse teams are the most fun and best teams. I encourage everyone to apply.

Possibilities to join my team as MSc thesis researcher

We are supporting MSc students to undertake research in this area within the scope of their thesis which frequently leads to scientific publications. Get in touch if you are interested.

Internship as visiting researcher (MSc or PhD)

If you plan to spend around 1 year at TU Delft, have a good PhD performance, and have an initial idea of how your research could fit into our work please get in touch. However, due to our strong internal student base, we have limited capacity to host visiting researchers from other universities. This rarely justifies the administrative and supervisory overhead of stays shorter than one year, though exceptions may be made.

If you are already an exchange student at TU Delft for your MSc degree, we may be able to arrange for you to write your MSc thesis in our team, for example, we have good experience hosting Erasmus students who consider extending their stay at TU Delft for their thesis work.

List of graduated PhD researchers

• Al-Amin Bugaje, “Data Augmentation for Power System Security Assessment”, Imperial College London, now a researcher at Hitachi Energy in Canada
• Federica Bellizio, “Topological Changes in Data-Driven Dynamic Security Assessment for Power System Control”, Imperial College London, now at the Swiss Federal Laboratories for Materials Science and Technology and Co-Founder of Kuafu