---
title: "Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.21407v1"
permalink: "https://feed.astralor.com/items/itm_590c31d1324598eff18958e5"
published: "2026-07-23T15:06:57.000Z"
score: 84
channel: "ai"
tags: ["SOLPS-ITER","Uncertainty Estimation","Cycle-Consistent Neural Surrogate","Tokamak","Digital twin","Conditional U-Net"]
---

# Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas

## AI commentary

> This paper is worth attention because it demonstrates how advanced AI techniques (cycle consistency, uncertainty quantification) directly address a core computational bottleneck in nuclear fusion. For fusion researchers and engineers, it offers a practical tool to speed up simulations by orders of magnitude without sacrificing accuracy; for AI practitioners, it provides an excellent case of self-supervised learning and inverse problem solving in scientific applications.

## At a glance

This study develops a cycle-consistent neural surrogate for tokamak edge plasmas, combining a conditional U-Net forward model with an optimization-based inverse method built on the frozen forward network, enabling fast prediction from control parameters to 2D plasma-state fields, parameter recovery, and uncertainty estimation, with inference speeds 5-6 orders of magnitude faster than traditional simulations.

## Key points

- The forward model is a conditional U-Net mapping five control parameters to 2D plasma-state fields, achieving normalized RMSE below 2.6% and Pearson correlation above 0.95 for all fields.
- The inverse method uses cycle-consistency regularization for self-supervised quality checking without ground-truth labels, boosting average cyclical R² from 0.59 to 0.99, and recovering all five control parameters with Pearson r≥0.97.
- An ensemble of MLPs predicts electron temperature and density profiles with uncertainty estimates, flagging regions needing more simulations.
- With about 4×10⁶ parameters, the model produces full 2D predictions in milliseconds, 5-6 orders of magnitude faster than SOLPS-ITER, enabling real-time control, parameter scans, and digital twins.

## Full text

The boundary and divertor plasma govern how a tokamak exhausts power and particles, setting heat fluxes, target conditions, and the onset of detachment. Predicting these quantities is essential for operating current and future devices, but edge simulations that resolve them are too slow for parameter scans, optimization, or real-time control. Machine-learning surrogates offer a fast alternative, yet most are forward-only: they cannot recover input parameters from observations or assess the reliability of their predictions. We introduce a cycle-consistent neural surrogate for edge plasmas, combining a conditional U-Net forward model with an optimization-based inverse method built on the frozen forward network. The forward model maps five control parameters to two-dimensional plasma-state fields on the SOLPS-ITER mesh; the inverse method enforces consistency between forward and inverse predictions, a self-supervised quality check needing no ground-truth labels at inference. An ensemble of multilayer perceptrons also predicts electron temperature and density profiles at the outboard midplane and divertor targets, with uncertainty estimates that flag where more simulations are needed. The forward model achieves normalized root-mean-square errors below 2.6% and Pearson correlations above 0.95 for all fields. Cycle-consistency regularization raises the average cyclical $R^2$ from 0.59 to 0.99 without degrading forward accuracy and enables recovery of the core fueling rate; all five control parameters are recovered with Pearson $r\ge0.97$. A $k$-d tree warm start yields a database completion rate above 95%, versus roughly 30% outright failures when cold-started. With about $4\times10^6$ parameters, the model produces full 2D predictions in milliseconds, five to six orders of magnitude faster than SOLPS-ITER, enabling real-time control, parameter scans, uncertainty analysis, and digital twins.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_590c31d1324598eff18958e5