Optimizing performance of artificial neural network-based tomography model at golem tokamak: impact of training data quantity and quality
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61389021%3A_____%2F24%3A00616842" target="_blank" >RIV/61389021:_____/24:00616842 - isvavai.cz</a>
Result on the web
<a href="https://ojs.cvut.cz/ojs/index.php/PPT/article/view/9980/7184" target="_blank" >https://ojs.cvut.cz/ojs/index.php/PPT/article/view/9980/7184</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Optimizing performance of artificial neural network-based tomography model at golem tokamak: impact of training data quantity and quality
Original language description
The paper presents an Artificial Neural Network (ANN)-based model for the tomography reconstruction of visible plasma radiation distribution at the GOLEM tokamak. To train the model, the training dataset is constructed using emissivity phantoms with associated synthetic measurements from one poloidal cross-section of the GOLEM tokamak. The trained model is validated by the test dataset. The performance optimization of the ANN-based model is investigated by considering the effect of the quantity and quality of the training data.
Czech name
—
Czech description
—
Classification
Type
O - Miscellaneous
CEP classification
—
OECD FORD branch
10305 - Fluids and plasma physics (including surface physics)
Result continuities
Project
—
Continuities
I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace
Others
Publication year
2024
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů