Speaker: Geert Verdoolaege, Department of Applied Physics, Ghent University, Ghent, Belgium
Title: Bayesian Probability for Integrated Data Analysis and Pattern Recognition in Fusion Devices
Abstract: Fusion energy research can benefit greatly from modern data science methods, both for increasing the understanding of the underlying plasma physics and for optimizing the design and operation of fusion devices. From basic statistical techniques for model fitting, to Bayesian methods for probabilistic analysis of data from single or multiple diagnostics, to the latest machine learning techniques for anomaly detection and uncertainty quantification: the applications are numerous and the possible approaches originate from a broad range of subfields of the information sciences. In this talk, I will highlight a number of recent applications of Bayesian inference in fusion. After introducing the principles and treating a few general applications, integrated analysis of fusion diagnostic data will be discussed. Using Bayesian techniques, this approach enables a systematic, joint treatment of data from multiple, heterogeneous diagnostics. I present some applications where Bayesian probability gracefully handles tomographic inversion and error propagation through complex measurement models. Opportunities for design optimization and machine learning techniques for speeding up the inference process are also touched upon, with a view to real-time integrated data analysis in future devices. I then proceed to applications of Bayesian inference and information geometry for pattern recognition in fusion data, concentrating on robust estimation of scaling laws in complex, multi-machine data sets, as well as anomaly detection for predictive maintenance in fusion devices.
Bio: Geert Verdoolaege is an Associate Professor at Ghent University (UGent, Belgium), where he leads the research unit Nuclear Fusion (infusion). He obtained the M.Sc. degree in theoretical physics in 1999 and the Ph.D. in engineering physics in 2006, both at UGent. His research activities comprise development of data analysis techniques using methods from probability theory, machine learning and information geometry, and their application to nuclear fusion experiments. In addition, he teaches master courses on plasma physics and on continuum mechanics. He is also involved in international fusion education through the European fusion master FUSION-EP and joint PhD programs. He serves on the editorial board of the multidisciplinary journal Entropy and is a member of the scientific committees of several conferences on nuclear fusion and information science. Furthermore, he is a consulting expert in the International Tokamak Physics Activity (ITPA) Topical Groups on Diagnostics, as well as Transport and Confinement.
This talk will be offered virtually. If you wish to participate remotely, please send an email to [email protected].