The Data Fusion Labeler (dFL): Challenges and Solutions to Data Harmonization, Labeling, and Provenance in Fusion Energy

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Michoski, Craig, Waller, Matthew, Sammuli, Brian, Li, Zeyu, Nakkina, Tapan Ganatma, Nazikian, Raffi, Smith, Sterling, Orozco, David, Kuang, Dongyang, Foltin, Martin, Olofsson, Erik, Fredrickson, Mike, Louis-Jeune, Jerry, Hatch, David R., Oliver, Todd A., Clark, Mitchell, Louis, Steph-Yves
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912705429897216
author Michoski, Craig
Waller, Matthew
Sammuli, Brian
Li, Zeyu
Nakkina, Tapan Ganatma
Nazikian, Raffi
Smith, Sterling
Orozco, David
Kuang, Dongyang
Foltin, Martin
Olofsson, Erik
Fredrickson, Mike
Louis-Jeune, Jerry
Hatch, David R.
Oliver, Todd A.
Clark, Mitchell
Louis, Steph-Yves
author_facet Michoski, Craig
Waller, Matthew
Sammuli, Brian
Li, Zeyu
Nakkina, Tapan Ganatma
Nazikian, Raffi
Smith, Sterling
Orozco, David
Kuang, Dongyang
Foltin, Martin
Olofsson, Erik
Fredrickson, Mike
Louis-Jeune, Jerry
Hatch, David R.
Oliver, Todd A.
Clark, Mitchell
Louis, Steph-Yves
contents Fusion energy research increasingly depends on the ability to integrate heterogeneous, multimodal datasets from high-resolution diagnostics, control systems, and multiscale simulations. The sheer volume and complexity of these datasets demand the development of new tools capable of systematically harmonizing and extracting knowledge across diverse modalities. The Data Fusion Labeler (dFL) is introduced as a unified workflow instrument that performs uncertainty-aware data harmonization, schema-compliant data fusion, and provenance-rich manual and automated labeling at scale. By embedding alignment, normalization, and labeling within a reproducible, operator-order-aware framework, dFL reduces time-to-analysis by greater than 50X (e.g., enabling >200 shots/hour to be consistently labeled rather than a handful per day), enhances label (and subsequently training) quality, and enables cross-device comparability. Case studies from DIII-D demonstrate its application to automated ELM detection and confinement regime classification, illustrating its potential as a core component of data-driven discovery, model validation, and real-time control in future burning plasma devices.
format Preprint
id arxiv_https___arxiv_org_abs_2511_09725
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Data Fusion Labeler (dFL): Challenges and Solutions to Data Harmonization, Labeling, and Provenance in Fusion Energy
Michoski, Craig
Waller, Matthew
Sammuli, Brian
Li, Zeyu
Nakkina, Tapan Ganatma
Nazikian, Raffi
Smith, Sterling
Orozco, David
Kuang, Dongyang
Foltin, Martin
Olofsson, Erik
Fredrickson, Mike
Louis-Jeune, Jerry
Hatch, David R.
Oliver, Todd A.
Clark, Mitchell
Louis, Steph-Yves
Plasma Physics
Machine Learning
Data Analysis, Statistics and Probability
I.2
Fusion energy research increasingly depends on the ability to integrate heterogeneous, multimodal datasets from high-resolution diagnostics, control systems, and multiscale simulations. The sheer volume and complexity of these datasets demand the development of new tools capable of systematically harmonizing and extracting knowledge across diverse modalities. The Data Fusion Labeler (dFL) is introduced as a unified workflow instrument that performs uncertainty-aware data harmonization, schema-compliant data fusion, and provenance-rich manual and automated labeling at scale. By embedding alignment, normalization, and labeling within a reproducible, operator-order-aware framework, dFL reduces time-to-analysis by greater than 50X (e.g., enabling >200 shots/hour to be consistently labeled rather than a handful per day), enhances label (and subsequently training) quality, and enables cross-device comparability. Case studies from DIII-D demonstrate its application to automated ELM detection and confinement regime classification, illustrating its potential as a core component of data-driven discovery, model validation, and real-time control in future burning plasma devices.
title The Data Fusion Labeler (dFL): Challenges and Solutions to Data Harmonization, Labeling, and Provenance in Fusion Energy
topic Plasma Physics
Machine Learning
Data Analysis, Statistics and Probability
I.2
url https://arxiv.org/abs/2511.09725