[Paper Review] Modeling of Particle Transport, Neutrals and Radiation in Magnetically-Confined Plasmas with Aurora

F. Sciortino, T. Odstrčil+9 more|arXiv (Cornell University)|Jun 8, 2021

Magnetic confinement fusion researchPhysics and Astronomy31 references23 citations

TL;DR

Aurora is an open-source, multi-language simulation package for modeling particle transport, neutral particles, and radiation in magnetically confined fusion plasmas. It enables high-performance 1.5D impurity transport simulations using atomic data from ADAS and other sources, with key innovations including the superstaging approximation for heavy ions and validation that charge exchange has negligible impact on core radiated power in ITER.

ABSTRACT

We present Aurora, an open-source package for particle transport, neutrals and radiation modeling in magnetic confinement fusion plasmas. Aurora's modern multi-language interface enables simulations of 1.5D impurity transport within high-performance computing frameworks, particularly for the inference of particle transport coefficients. A user-friendly Python library allows simple interaction with atomic rates from the Atomic Data and Atomic Structure database as well as other sources. This enables a range of radiation predictions, both for power balance and spectroscopic analysis. We discuss here the superstaging approximation for complex ions, as a way to group charge states and reduce computational cost, demonstrating its wide applicability within the Aurora forward model and beyond. Aurora also facilitates neutral particle analysis, both from experimental spectroscopic data and other simulation codes. Leveraging Aurora's capabilities to interface SOLPS-ITER results, we demonstrate that charge exchange is unlikely to affect the total radiated power from the ITER core during high performance operation. Finally, we describe the ImpRad module in the OMFIT framework, developed to enable experimental analysis and transport inferences on multiple devices using Aurora.

Motivation & Objective

• To develop a modern, open-source framework for integrated modeling of impurity transport, radiation, and neutral particles in magnetic fusion devices.
• To reduce computational cost in modeling complex ions through the superstaging approximation.
• To enable high-performance, in-memory simulations for rapid inference of transport coefficients from experimental data.
• To facilitate interoperability with major fusion codes like SOLPS-ITER and FIDASIM, and with the OMFIT framework for experimental analysis.
• To support spectroscopic analysis and radiative power predictions using standardized atomic data from ADAS and other sources.

Proposed method

• Aurora implements a 1.5D forward model for ion transport, solving coupled partial differential equations for charge state densities using user-defined or inferred transport coefficients.
• It integrates atomic rate coefficients from the ADAS database and other sources via automated online retrieval or local file input.
• The superstaging approximation groups multiple charge states into superlevels to reduce computational cost while preserving accuracy in radiation and transport predictions.
• The package provides a Python library for parsing ADF15 (PECs), ADF04 (GCR), and other atomic data formats, enabling spectral emissivity and radiated power calculations.
• It supports post-processing of neutral particle data from codes like KN1D, FIDASIM, and SOLPS-ITER, enabling charge exchange and neutral beam analysis.
• The ImpRad module in OMFIT enables experimental inference of transport coefficients across multiple devices using Aurora’s physics models and Bayesian inference tools.

Experimental results

Research questions

• RQ1 To what extent does charge exchange with edge neutrals affect the total radiated power within the last closed flux surface in ITER?
• RQ2 How accurate is the superstaging approximation for heavy ions like tungsten in predicting radiation and transport?
• RQ3 Can Aurora’s 1.5D transport model enable efficient inference of particle transport coefficients from experimental data?
• RQ4 How well can Aurora reproduce experimental spectroscopic signals using atomic data from ADAS and other sources?
• RQ5 To what extent can Aurora’s modular design and open architecture support integration with diverse fusion simulation and analysis workflows?

Key findings

• The superstaging approximation significantly reduces computational cost for modeling complex ions like tungsten while maintaining accuracy in radiation and transport predictions.
• Charge exchange with edge neutrals is unlikely to affect the total radiated power within the ITER core during high-performance operation, based on SOLPS-ITER post-processing with Aurora.
• Aurora enables high-performance, in-memory 1.5D simulations that reduce runtime by over 10× compared to disk-I/O intensive methods, especially beneficial for transport coefficient inference.
• The ImpRad module in OMFIT allows consistent experimental analysis and transport inference across multiple devices using a unified physics model based on Aurora.
• Aurora successfully interfaces with experimental data and simulation outputs from EFIT, GACODE, SOLPS-ITER, FIDASIM, and KN1D, enabling cross-code validation and spectroscopic analysis.
• The package supports automated parsing and interpolation of ADAS ADF15 and ADF04 files, enabling accurate emissivity and cooling coefficient calculations for fusion-relevant ions.