High-fidelity inference of local impurity profiles in C-2W using Bayesian tomography.

Area of Science:

• Plasma Physics
• Fusion Energy Research
• Spectroscopy

Background:

• The C-2W device sustains steady-state, beam-driven field-reversed configuration plasmas in a magnetic mirror.
• Accurate impurity profile measurements are crucial for understanding plasma behavior and confinement.
• Plasma non-uniformity complicates direct spectral analysis from Doppler spectroscopy.

Purpose of the Study:

• To develop and apply a robust method for inferring local impurity profiles from line-integrated spectroscopic measurements.
• To overcome challenges posed by plasma non-uniformity in the C-2W device.
• To model oxygen impurity ion populations (thermal and beam) using advanced analysis techniques.

Main Methods:

• Utilized Bayesian tomography with a comprehensive forward model of the Doppler spectroscopy diagnostic.
• Modeled O4+ triplet line emission, considering distinct thermal and beam ion populations.
• Employed Gaussian processes for local profile generation and inference, incorporating diagnostic geometry, instrument function, calibration, and noise models.
• Applied Markov chain Monte Carlo (MCMC) sampling for high-fidelity uncertainty quantification.

Main Results:

• Successfully reconstructed local O4+ impurity profiles in the C-2W plasma.
• Inferred profiles demonstrated consistency with data from other diagnostics.
• Results showed good agreement with established physics models for plasma transport and biasing circuits.

Conclusions:

• Bayesian tomography is a superior technique for analyzing non-uniform plasma spectroscopic data.
• The reconstructed impurity profiles provide valuable insights into plasma conditions and transport phenomena in C-2W.
• This methodology enhances the capability to diagnose and understand complex fusion plasmas.