We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Simulation of hard X-ray time evolution in plasma tokamak by using the NARX-GA hybrid neural network.
- Authors
Alavi, Amir; Saadat, Shervin; Ghanbari, Mohamad Reza; Alavi, Seyed Enayatallah; Kadkhodaie, Ali
- Abstract
The NARX-GA hybrid neural network was applied to simulate the time evolution of runaway electrons (REs) in the plasma tokamak. This particular type of artificial neural network was created specifically for time series prediction. The NARX-GA network was built using inputs from some plasma diagnostic signals (loop voltage, hard X-ray) collected during all phases of plasma tokamak discharges. The network output predicts the time evolution of hard X-ray (HXR) signals up to 500 μs, which can be achieved with high accuracy (MSE = 3.76 × 10 - 5 ). The real-time application of this methodology can pave the way for prompt REs control action. The confinement time increases as the REs generation decreases, and their destructive effects on the tokamak wall decrease as well. Early prediction of RE behavior is critical in attempting to mitigate their potentially dangerous effects.
- Subjects
HARD X-rays; TOKAMAKS; RECURRENT neural networks; PLASMA flow; TIME series analysis
- Publication
European Physical Journal D (EPJ D), 2022, Vol 76, Issue 10, p1
- ISSN
1434-6060
- Publication type
Article
- DOI
10.1140/epjd/s10053-022-00511-6