Search NASA⌕ Search

Engineering topics

Rognlien, Thomas

Publications and source records attributed to Rognlien, Thomas.

Data-driven model for divertor plasma detachment prediction

We present a fast and accurate data-driven surrogate model for divertor plasma detachment prediction leveraging the latent feature space concept in machine learning research. Our approach involves constructing and training two neural networks: an autoencoder that finds a proper latent space representation (LSR) of plasma state by compressing the multi-modal diagnostic measurements and a forward model using multi-layer perception (MLP) that projects a set of plasma control parameters to its corresponding LSR. By combining the forward model and the decoder network from autoencoder, this new data-driven surrogate model is able to predict a consistent set of diagnostic measurements based on a few plasma control parameters. In order to ensure that the crucial detachment physics is correctly captured, highly efficient 1D UEDGE model is used to generate training and validation data in this study. The benchmark between the data-driven surrogate model and UEDGE simulations shows that our surrogate model is capable of providing accurate detachment prediction (usually within a few per cent relative error margin) but with at least four orders of magnitude speed-up, indicating that performance-wise, it has the potential to facilitate integrated tokamak design and plasma control. Comparing with the widely used two-point model and/or two-point model formatting, the new data-driven model features additional detachment front prediction and can be easily extended to incorporate richer physics. This study demonstrates that the complicated divertor and scrape-off-layer plasma state has a low-dimensional representation in latent space. Understanding plasma dynamics in latent space and utilising this knowledge could open a new path for plasma control in magnetic fusion energy research.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Integrated Liquid Metal Flowing First Wall and Open-Surface Divertor for Fusion Nuclear Science Facility: Concept, Design, and Analysis

The Fusion Energy System Studies (FESS) Fusion Nuclear Science Facility (FNSF) project team in the United States is examining the use of liquid metals (LMs) for plasma-facing components (PFCs). Our approach has been to utilize an already established fusion design, FESS-FNSF, which is a tokamak-based machine with 518 MW fusion power, a 4.8-m major radius, a 1.2-m minor radius, and a machine average neutron wall loading of ~1 MW/m 2 . For this design, we propose a PFC concept that integrates a flowing LM first wall (FW) and an open-surface divertor. The flowing LM first removes the surface heat flux from the FW and then proceeds to the lower section of the vacuum chamber to form a large area LM surface for absorbing high peak surface heat flux in the divertor region. In pursuing the application of large open LM surfaces in the FNSF, two new computer codes have been developed and then applied to the analysis of free-surface magnetohydrodynamic flows and heat transfer, including fast thin flowing liquid layers over the solid FW (liquid wall), a tublike divertor, and a fast flow divertor. The analysis is aimed at optimization of the liquid wall design by matching certain proposed design criteria and also at evaluation of the maximum heat fluxes, using liquid lithium (Li) as a working fluid. It was demonstrated that the flowing Li FW (at ~2 cm and ~10 m/s) can tolerate a surface heat flux of ~1 MW/m 2 , while the open-surface Li divertor can remove a maximum high peak heat flux of 10 MW/m 2 . Furthermore, the paper also focuses on the underlying science. One such example is the evaluation and characterization of heat transfer mechanisms and heat transfer intensification in the tublike Li divertor.

36 MATERIALS SCIENCE↗