DOE OSTI · 1779436
Data-driven profile prediction for DIII-D
Abstract
A new, fully data-driven algorithm has been developed that uses neural networks to predict plasma pro les on a scale of τ E into the future given actuators and the present plasma state. The model was trained and tested on DIII-D data from the 2013-2018 experimental campaigns. The model is accurate on average, with q predictions the worst and pressure predictions the best. The model can run in milliseconds and is very simple to use. This makes it a potentially useful tool for operators and physicists when planning plasma scenarios. It also is a candidate for doing phase-space exploration without going through the DIIID database or complicated and computationally expensive simulation codes. Here, a reduced model using only realtime diagnostics has also been developed and formed the basis for a model-predictive control algorithm implemented and successfully tested on DIII-D.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Abbate, Joseph, Conlin, R., Kolemen, E.. 2021-03-19. Data-driven profile prediction for DIII-D. https://doi.org/10.1088/1741-4326%2Fabe08d. https://searchnasa.com/record/osti%3A1779436/data-driven-profile-prediction-for-diii-d
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Search NASA by Indiciae: BibTeX · RIS.