b'Archiving Experimental BreederLinking multi-physics models with a centralized database containing Reactor-II Metallic Fuel Testhistorical experimental information accelerates nuclear fuel development Data Using the Nuclear Energyandqualification.Advanced Modeling andQ ualification of a fuel type for use in a new nuclear design can be very complicated and time-consuming, sometimes taking several decades. Simulation Program to AccelerateCombining data from experiments performed on similar fuel designs for different reactors into a single database allows those experimental results to be used Fast Reactor Fuel Qualification to help qualification of the new fuel design. The data can also be used to validate the accuracy of fuel performance models. The database and model would be linked together so that design parameters for the new reactor fuel can be used to search the database for appropriate data for validation.The goal of this research was to select an existing database, improve it as needed, TOTAL APPROVED AMOUNT:and demonstrate the database and model linkage for a new fuel design. The $1,137,000 over 3 years sodium-cooled fast reactor was chosen as the example and the associated metal fuel database further selected because much of it had already been collected PROJECT NUMBER:under the earlier Integral Fast Reactor Materials Information System program. 19A39-103 New fuel performance modeling used by the Nuclear Energy Advanced Modeling PRINCIPAL INVESTIGATOR:and Simulation program was also selected for the demonstration. Argonne Douglas Porter National Laboratory had already compiled the information into a new database, the Fuels Irradiation & Physics Database. A collaborative effort between the two CO-INVESTIGATORS: laboratories resulted in successful linkage between the model and the database. Andrei Gribok, INL It was shown that these data could be analyzed statistically to provide measures Nancy Lybeck, INL of accuracy of the model, which is required for use in fuel qualification and safety Pavel Medvedev, INL analysis support. Machine learning and image analysis techniques were applied COLLABORATORS: to some of the rough data to enhance its usefulness. For example, radiography Argonne National Laboratory of irradiated test fuel pins was examined to extract specific information from the Pacific Northwest National Laboratory radiographs, such as fuel swelling.A representation of some of the performance data that can be tracked and calculated for a single fuel pin from the database.32'