BUA Institution

Technische Universität Berlin

Research Description

As a professor of computer science, I work on methods and systems for data analytics, including acquisition, organization and modeling, storage and compression, data integration, query processing, and machine learning/AI. The focus is on large sensor time series and analytics on those, e.g., forecasting. The main application area is digital energy, especially modeling, optimizing, and trading energy flexibility from devices such as batteries, wind turbines, PV panels, and EVs. 

Where in the world has your career been largely based until now?

Mostly Denmark, with longer stays in Scotland and California. 

Why Berlin?

Berlin, and especially TU Berlin, has a world-class research environment within systems and techniques for large-scale data processing and analytics. I have known my host, Prof. Matthias Böhm, for over 15 years and already work with him. 

What fascinates you about your research area?

That data is the foundation of learning, both for humans and machines. Some say that AI = 75% data and only 25% machine learning. 

How will your research change the world?

One line of my research aims to provide methods and systems for data analytics that are an order of magnitude more sustainable, in terms of resource consumption, than current methods and systems. My other line of research aims to enable a cost-efficient energy system running mostly on renewable energy. 

What major short-term goal are you currently working towards with your research?

1. Making our ModelarDB time series management system the fastest and cheapest for massive high-frequency time series. 2. Making FlexOffers, our general and scalable model for energy flexibility in time, amount, and type, a de-facto standard in digital energy. 

How did you become interested in your specific topic?

By working both in the academic data management community as well as with companies from the energy sector.