Computational design of a novel soft-Xray-based turbulence diagnostic in NSTX-U
Turbulence transport poses a significant challenge in fusion research. The measurement of turbulent fluctuations is critical for comprehending turbulence transport, predicting its behavior, and ultimately controlling it to maximize fusion gain. However, there is a notable scarcity electron temperatu...
Main Author: | |
---|---|
Other Authors: | |
Format: | Thesis |
Published: |
Massachusetts Institute of Technology
2024
|
Online Access: | https://hdl.handle.net/1721.1/157154 https://orcid.org/0000-0003-0557-1489 |
Summary: | Turbulence transport poses a significant challenge in fusion research. The measurement of turbulent fluctuations is critical for comprehending turbulence transport, predicting its behavior, and ultimately controlling it to maximize fusion gain. However, there is a notable scarcity electron temperature fluctuation diagnostics, including for high density tokamak plasmas and in spherical tokamaks. The ultimate aim of our research is to develop a novel diagnostic tool for temperature fluctuations. Before experimental exploration, conducting a numerical feasibility study is essential for the proposed diagnostic. The high spatial and temporal resolutions that are attainable using Soft X-ray (SXR) imaging makes it a promising candidate. The primary objective of the thesis is to assess the feasibility of an electron temperature fluctuation diagnostic based on SXR imaging.
The feasibility study involves gathering fluctuation data and constructing a numerical diagnostic model. This model computes synthetic SXR measurements, which are then reconstructed using tomographic algorithms to derive electron temperature fluctuations. These reconstructions are then compared against the ground truth to assess diagnostic performance. Optimization of performance is achieved by adjusting diagnostic parameters to identify the optimal set for feasibility analysis.
This study consists of two primary parts. First, we utilize a simplified toy model with circular plasma geometry and synthetic fluctuation data abstracted from gyrokinetic simulation fluctuation spectra and we employ a pseudolocal tomography algorithm for reconstruction and demonstrate reliable measurement of electron temperature fluctuations for X-ray detectors with sufficiently high signal-to-noise ratio. Second, we conduct a more comprehensive feasibility study using fluctuation data directly generated from gyrokinetic simulations, in a real (spherical tokamak) NSTX-U configuration with complex plasma geometry. Assumptions from the previous study, such as infinitely thin beam size, are relaxed to assess their impact on reconstruction. Additionally, we enhance the reconstruction algorithm using neural networks, enabling reconstruction of both electron density and temperature fluctuations, as well as cross-phase analysis. Overall, the study confirms the feasibility of the SXR diagnostic given that SXR detectors meet minimum requirements. Furthermore, we explore fluctuations generated from different gyrokinetic simulations, demonstrating the diagnostic's ability to differentiate fluctuations originating from different instabilities under the same configuration.
This research provides a theoretical foundation and guidance for developing a practical SXR-based electron temperature fluctuation diagnostic for experimental use. It outlines the measurable quantities, their limitations, and the minimum requirements for SXR hardware to ensure reliable measurements. This contribution significantly advances our understanding of plasma turbulence transport, addressing a key challenge in fusion research. However, the current study's limitations employ a simplified emissivity model. Utilizing a more comprehensive model incorporating atomic data could yield more robust conclusions. Additionally, incorporating real hardware parameters would enhance the reliability of the conclusion. |
---|