Speaker
Description
The EXFOR database is the major source to retrieve experimental data as input for nuclear data evaluations. It aggregates decades worth of international nuclear data measurements. EXFOR is specifically designed to store experimental data as it was reported by the experimentalists, with limited corrections made as needed. For this reason, the database does not store the subjective alterations to the data sets by users, such as evaluators. For example, evaluators may re-normalize data sets or augment uncertainties to increase consistency between data sets in their evaluation. These altered data sets, which are needed to reproduce the evaluations, are not readily available to others in the field. To improve the reproducibility of evaluations, the choices made by the evaluators need to be stored in a database that builds on the EXFOR database. In addition, some data in EXFOR are not easy to interpret automatically, including meta-data about experiments. If the new database improves the interpretability of this data with a different format, applying automatic processing and using machine learning techniques will be more tenable. To design this new database, an international collaboration has been established through the Organisation for Economic Co-operation and Development (OECD)/Nuclear Energy Agency (NEA) Working Party on International Nuclear Data Evaluation Co-operation (WPEC) Subgroup 50 (SG50): Developing an Automatically Readable, Comprehensive and Curated Experimental Reaction Database. The goals of SG50 are to create a requirements document, a specifications document, and example files. The requirements document will contain the metadata to be stored in the new database, while the specifications document will outline the data structures used to store the information. This presentation will outline the current progress of SG50.
LA-UR-21-28324