A novel unsupervised machine learning algorithm for automatic Alfvénic activity detection in the TJ-II stellarator

A novel sparse encoding algorithm is developed to detect and study plasma instabilities automatically. This algorithm, called Elastic Random Mode Decomposition, is applied to the Mirnov coil signals of a dataset of 1291 discharges of the TJ-II stellarator, enabling the identification of the Alfvénic...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Nuclear fusion Ročník 64; číslo 12; s. 126057 - 126077
Hlavní autoři: Zapata-Cornejo, E.d.D., Zarzoso, D., Pinches, S.D., Bustos, A., Cappa, A., Ascasibar, E.
Médium: Journal Article
Jazyk:angličtina
Vydáno: IOP Publishing 01.12.2024
Témata:
Alfvénic activity
Artificial Intelligence
Computer Science
elastic net
Gabor atoms
Physics
random dictionaries
sparse coding
TJ-II
unsupervised learning
Alfvénic activity
MHD instabilities
TJ-II
Gabor atoms
clustering
elastic net
random dictionaries
sparse coding
unsupervised learning
ISSN:0029-5515, 1741-4326
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:A novel sparse encoding algorithm is developed to detect and study plasma instabilities automatically. This algorithm, called Elastic Random Mode Decomposition, is applied to the Mirnov coil signals of a dataset of 1291 discharges of the TJ-II stellarator, enabling the identification of the Alfvénic activity. In the presented approach, each signal is encoded as a collection of basic waveforms called atoms, drawn from a signal dictionary. Then the modes are identified using clustering and correlations with other plasma signals. The performance of the proposed algorithm is dramatically increased by using elastic net regularization and taking advantage of GPU architectures. Therefore the signal size and the number of dictionary elements are no longer limiting factors for encoding complex signals. Once the modes are retrieved from the shots, standard clustering and dimensionality reduction techniques are applied to obtain a 2D map featuring of the physical mode characteristics of this subset of TJ-II shots. The clustering features consider the relationship with the plasma current I p , the diamagnetic energy W , and inverse square root of electron density 1 / n , profiling different subtypes of Alfvénic activity. The proposed algorithm can potentially create large databases of labeled modes with unprecedented detail.
Bibliografie:NF-107115.R2
ISSN:0029-5515
1741-4326
DOI:10.1088/1741-4326/ad85f4