Speakers
Description
Here we report on concepts and activities of the AC department in data management and automation.
We apply the idea of the FAIR principle$^1$ and seek to establish the basis for the application of data analysis and data mining methods.
This requires reliable, reproducible data sets with high diversity. To generate such data, handbooks are used in the research in which the characterisation of catalysts and the determination of kinetic data are precisely prescribed.$^2$ These handbooks should specify the minimum data set that should be generated for each catalyst and how the measurements should be performed. Handbooks are important for quality control and allow reproducibility and comparability of kinetic experiments by reference to external benchmarks.
To implement such a concept, we are working on a system that consists of the following components:
- AC database (archive)
- EPICS or LabVIEW for communicating with devices and collecting measurement data
- Archiver appliance for storage of time series
- Phoebus for the creation of graphical user interfaces
- Python/Bluesky/Jupyter notebooks for creating automations and evaluations
- S3-Storage as long-term storage for large amounts of data
The workflow of a scientific experiment will consist of the following steps:
1. Create/select a method/recipe
2. Carrying out the experiment by executing the method/recipe
3. Data analysis
All data obtained in an experiment are linked to the sample and all metadata and can be accessed via the web interface of the AC archive.
References
1. Mark D. Wilkinson, et al., Scientific Data 3, 160018 (2016).
2. Annette Trunschke, et al., Topics in Catalysis 63, 1683 (2020).
