Bayesian inference of ion velocity distribution function from laser-induced fluorescence spectra

Area of Science:

• Plasma Physics
• Statistical Mechanics
• Spectroscopy

Background:

• The velocity distribution function (VDF) is crucial for understanding particle kinetics and macroscopic properties in many-body systems.
• Laser-induced fluorescence (LIF) spectroscopy is a key technique for measuring local VDFs in spatially inhomogeneous plasmas.
• Determining the precise analytic form of the VDF can be challenging in non-equilibrium conditions due to complex factors.

Purpose of the Study:

• To propose and validate a novel method for selecting the appropriate analytic form of the VDF using Bayesian statistics.
• To apply Bayesian inference to ion VDFs using LIF spectral data from a linear magnetized plasma.
• To evaluate spatial inhomogeneity by verifying the analytic forms of local VDFs.

Main Methods:

• Formulation of Bayesian inference for ion velocity distribution functions.
• Application of the Bayesian approach to LIF spectra obtained at multiple spatial locations.
• Analysis of spectral data to infer and validate VDF models.

Main Results:

• Successfully applied Bayesian inference to LIF data for ion VDFs in a magnetized plasma.
• Demonstrated the capability to assess spatial inhomogeneity by analyzing local VDF forms.
• Validated the proposed method for selecting appropriate VDF analytic forms.

Conclusions:

• The Bayesian statistical approach provides a robust framework for experimentally determining VDFs.
• This method is effective in characterizing spatial variations in plasma VDFs.
• The approach has broad applicability to plasmas, gases, and liquids for establishing VDFs.