Towards fast surrogate models for interpolation of tokamak edge plasmas
One of the major design limitations for tokamak fusion reactors is the heat load that can be sustained by the materials at the divertor target. Developing a full understanding of how machine or operation parameters affect the conditions at the divertor requires an enormous number of simulations. A p...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-03-01
|
| Series: | Nuclear Materials and Energy |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352179123000352 |
| _version_ | 1811154356349698048 |
|---|---|
| author | Stefan Dasbach Sven Wiesen |
| author_facet | Stefan Dasbach Sven Wiesen |
| author_sort | Stefan Dasbach |
| collection | DOAJ |
| description | One of the major design limitations for tokamak fusion reactors is the heat load that can be sustained by the materials at the divertor target. Developing a full understanding of how machine or operation parameters affect the conditions at the divertor requires an enormous number of simulations. A promising approach to circumvent this is to use machine learning models trained on simulation data as surrogate models. Once trained such surrogate models can make fast predictions for any scenario in the design parameter space. In future such simulation based surrogate models could be used in system codes for rapid design studies of future fusion power plants. This work presents the first steps towards the development of such surrogate models for plasma exhaust and the datasets required for their training. Machine learning models like neural networks usually require several thousand data points for training, but the exact amount of data required varies from case to case. Due to the long runtimes of simulations we aim at finding the minimal amount of training data required. A preliminary dataset based on SOLPS-ITER simulations with varying tokamak design parameters, including the major radius, magnetic field strength and neutral density is constructed. To be able to generate more training data within reasonable computation time the simulations in the dataset use fluid neutral simulations and no fluid drift effects. The dataset is used to train a simple neural network and Gradient Boosted Regression Trees and test how the performance depends on the number of training simulations. |
| first_indexed | 2024-04-10T04:15:27Z |
| format | Article |
| id | doaj.art-d5a5c272a0eb42e7bb8a0d767ac4e55a |
| institution | Directory Open Access Journal |
| issn | 2352-1791 |
| language | English |
| last_indexed | 2024-04-10T04:15:27Z |
| publishDate | 2023-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Nuclear Materials and Energy |
| spelling | doaj.art-d5a5c272a0eb42e7bb8a0d767ac4e55a2023-03-12T04:21:36ZengElsevierNuclear Materials and Energy2352-17912023-03-0134101396Towards fast surrogate models for interpolation of tokamak edge plasmasStefan Dasbach0Sven Wiesen1Corresponding author.; Forschungszentrum Jülich GmbH, Institut für Energie- und Klimaforschung - Plasmaphysik, Jülich, 52425, GermanyForschungszentrum Jülich GmbH, Institut für Energie- und Klimaforschung - Plasmaphysik, Jülich, 52425, GermanyOne of the major design limitations for tokamak fusion reactors is the heat load that can be sustained by the materials at the divertor target. Developing a full understanding of how machine or operation parameters affect the conditions at the divertor requires an enormous number of simulations. A promising approach to circumvent this is to use machine learning models trained on simulation data as surrogate models. Once trained such surrogate models can make fast predictions for any scenario in the design parameter space. In future such simulation based surrogate models could be used in system codes for rapid design studies of future fusion power plants. This work presents the first steps towards the development of such surrogate models for plasma exhaust and the datasets required for their training. Machine learning models like neural networks usually require several thousand data points for training, but the exact amount of data required varies from case to case. Due to the long runtimes of simulations we aim at finding the minimal amount of training data required. A preliminary dataset based on SOLPS-ITER simulations with varying tokamak design parameters, including the major radius, magnetic field strength and neutral density is constructed. To be able to generate more training data within reasonable computation time the simulations in the dataset use fluid neutral simulations and no fluid drift effects. The dataset is used to train a simple neural network and Gradient Boosted Regression Trees and test how the performance depends on the number of training simulations.http://www.sciencedirect.com/science/article/pii/S2352179123000352SolpsPlasma exhaustDivertorSurrogateMachine learningNeural network |
| spellingShingle | Stefan Dasbach Sven Wiesen Towards fast surrogate models for interpolation of tokamak edge plasmas Nuclear Materials and Energy Solps Plasma exhaust Divertor Surrogate Machine learning Neural network |
| title | Towards fast surrogate models for interpolation of tokamak edge plasmas |
| title_full | Towards fast surrogate models for interpolation of tokamak edge plasmas |
| title_fullStr | Towards fast surrogate models for interpolation of tokamak edge plasmas |
| title_full_unstemmed | Towards fast surrogate models for interpolation of tokamak edge plasmas |
| title_short | Towards fast surrogate models for interpolation of tokamak edge plasmas |
| title_sort | towards fast surrogate models for interpolation of tokamak edge plasmas |
| topic | Solps Plasma exhaust Divertor Surrogate Machine learning Neural network |
| url | http://www.sciencedirect.com/science/article/pii/S2352179123000352 |
| work_keys_str_mv | AT stefandasbach towardsfastsurrogatemodelsforinterpolationoftokamakedgeplasmas AT svenwiesen towardsfastsurrogatemodelsforinterpolationoftokamakedgeplasmas |