EuroPED-NN: Uncertainty aware surrogate model
Name
24ja071_full.pdf
Size
4.31 MB
Format
Adobe PDF
Checksum (MD5)
f7009fc95774b1f966a7a8d1cb24ad12
Author(s)
Panera Alarez, A.
Ho, Aaron
Järvinen, A.
Saarelma, S.
Wiesen, S.
JET contributors
ASDEX Upgrade Team
Date Issued
June 2024
Journal
Plasma Physics and Controlled Fusion
Publisher
IOP
Abstract
This work successfully generates an uncertainty-aware surrogate model of the EuroPED plasma pedestal model using the Bayesian neural network with noise contrastive prior (BNN-NCP) technique. This model is trained using data from the JET-ILW pedestal database and subsequent model evaluations, conforming to EuroPED-NN. The BNN-NCP technique has been proven to be a suitable method for generating uncertainty-aware surrogate models. It matches the output results of a regular neural network while providing confidence estimates for predictions as uncertainties. Additionally, it highlights out-of-distribution (OOD) regions using surrogate model uncertainties. This provides critical insights into model robustness and reliability. EuroPED-NN has been physically validated, first, analyzing electron density ne(ψ_pol = 0.94) with respect to increasing plasma current, Ip, and second, validating the Δ−β_p,ped relation associated with the EuroPED model. This affirms the robustness of the underlying physics learned by the surrogate model. On top of that, the method was used to develop a EuroPED-like model fed with experimental data, i.e. an uncertainty aware experimental model, which is functional in JET database. Both models have been also tested in ~50 AUG shots.
Description
Submitted for publication in Plasma Physics and Controlled Fusion
MIT Department
Massachusetts Institute of Technology. Plasma Science and Fusion Center
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1088/1361-6587/ad6707
