Deep learning for numerical solutions of Hesegawa-Wakatani system.
In this paper， we use high-order splitting method to discretize the scheme of 2D-Hesegawa-Wakatani system ，generate numerical solutions of this system by BOUT++ and use DNN neural network to train a model from row-zonally average electrical density and vorticity to its local flux.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/175608 |
| Summary: | In this paper， we use high-order splitting method to discretize the scheme of 2D-Hesegawa-Wakatani system ，generate numerical solutions of this system by BOUT++ and use DNN neural network to train a model from row-zonally average electrical density and vorticity to its local flux. |
|---|