U.S. Department of Energy Office of Scientific and Technical Information
Towards fast and accurate predictions of radio frequency power deposition and current profile via data-driven modelling: applications to lower hybrid current drive
Three machine learning techniques (multilayer perceptron, random forest and Gaussian process) provide fast surrogate models for lower hybrid current drive (LHCD) simulations. A single GENRAY/CQL3D simulation without radial diffusion of fast electrons requires several minutes of wall-clock time to complete, which is acceptable for many purposes, but too slow for integrated modelling and real-time control applications. The machine learning models use a database of more than 16 000 GENRAY/CQL3D simulations for training, validation and testing. Latin hypercube sampling methods ensure that the database covers the range of nine input parameters ( $$n_{e0}$$ , $$T_{e0}$$ , $$I_p$$ , $$B_t$$ , $$R_0$$ , $$n_{\|}$$ , $$Z_{{\rm eff}}$$ , $$V_{{\rm loop}}$$ and $$P_{{\rm LHCD}}$$ ) with sufficient density in all regions of parameter space. The surrogate models reduce the inference time from minutes to $$\sim$$ ms with high accuracy across the input parameter space.
Wallace, G. M., et al. "Towards fast and accurate predictions of radio frequency power deposition and current profile via data-driven modelling: applications to lower hybrid current drive." Journal of Plasma Physics, vol. 88, no. 4, Aug. 2022. https://doi.org/10.1017/S0022377822000708
Wallace, G. M., Bai, Z., Sadre, R., Perciano, T., Bertelli, N., Shiraiwa, S., Bethel, E. W., & Wright, J. C. (2022). Towards fast and accurate predictions of radio frequency power deposition and current profile via data-driven modelling: applications to lower hybrid current drive. Journal of Plasma Physics, 88(4). https://doi.org/10.1017/S0022377822000708
Wallace, G. M., Bai, Z., Sadre, R., et al., "Towards fast and accurate predictions of radio frequency power deposition and current profile via data-driven modelling: applications to lower hybrid current drive," Journal of Plasma Physics 88, no. 4 (2022), https://doi.org/10.1017/S0022377822000708
@article{osti_1882254,
author = {Wallace, G. M. and Bai, Z. and Sadre, R. and Perciano, T. and Bertelli, N. and Shiraiwa, S. and Bethel, E. W. and Wright, J. C.},
title = {Towards fast and accurate predictions of radio frequency power deposition and current profile via data-driven modelling: applications to lower hybrid current drive},
annote = { Three machine learning techniques (multilayer perceptron, random forest and Gaussian process) provide fast surrogate models for lower hybrid current drive (LHCD) simulations. A single GENRAY/CQL3D simulation without radial diffusion of fast electrons requires several minutes of wall-clock time to complete, which is acceptable for many purposes, but too slow for integrated modelling and real-time control applications. The machine learning models use a database of more than 16 000 GENRAY/CQL3D simulations for training, validation and testing. Latin hypercube sampling methods ensure that the database covers the range of nine input parameters ( $n_{e0}$ , $T_{e0}$ , $I_p$ , $B_t$ , $R_0$ , $n_{\|}$ , $Z_{{\rm eff}}$ , $V_{{\rm loop}}$ and $P_{{\rm LHCD}}$ ) with sufficient density in all regions of parameter space. The surrogate models reduce the inference time from minutes to $\sim$ ms with high accuracy across the input parameter space. },
doi = {10.1017/S0022377822000708},
url = {https://www.osti.gov/biblio/1882254},
journal = {Journal of Plasma Physics},
issn = {ISSN 0022-3778},
number = {4},
volume = {88},
place = {United Kingdom},
publisher = {Cambridge University Press (CUP)},
year = {2022},
month = {08}}
Elwasif, Wael R.; Bernholdt, David E.; Shet, Aniruddha G.
2010 18th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), 2010 18th Euromicro Conference on Parallel, Distributed and Network-based Processinghttps://doi.org/10.1109/PDP.2010.63