Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz

AKPIK 6: AI Methods for Materials Science

AKPIK 6.4: Vortrag

Donnerstag, 20. März 2025, 17:15–17:30, H5

Reflectivity Analysis with AI: The LISA Data Pipeline at P08/DESY — •Julia Kobus^1,2, Lukas Petersdorf^1,2, Svenja Hövelmann^1,2, Alexander Hinderhofer³, Vladimir Starostin³, Chen Shen⁴, Florian Bertram⁴, Linus Pithan⁴, Frank Schreiber³, and Bridget Murphy^1,2,4 — ¹Institute of Experimental and Applied Physics, Kiel University, Leibnizstr. 19, 24118 Kiel, Germany — ²Ruprecht-Haensel Laboratory, Olshausenstr. 40, 24098 Kiel, Germany — ³University of Tübingen, 72076 Tübingen, Germany — ⁴Deutsches Elektronen-Synchrotron DESY, Notkestraße 85, 22607 Hamburg, Germany

We present a data analysis pipeline that is under development within TA3 of DAPHNE4NFDI for the LISA instrument at the P08 beamline at DESY. This pipeline is adapted from the development on solid surface XRR AI analysis to be used for liquid surfaces and interfaces. The pipeline aims at performing data reduction and subsequent analysis for reflectivity measurements using AI-based models. This approach increases measurement efficiency by processing data in real time, allowing flexible adjustments during the experiment and providing immediate insights. This will help users make better-informed decisions for subsequent measurements. The pipeline also will make advanced data analysis accessible to less experienced users. Our work will demonstrate the potential of AI to transform experimental workflows and data interpretation, and provides a blueprint for similar developments in the scientific community.

Keywords: data analysis; maschine learning; X-ray reflectivity