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

2025-11-12

AI Summary

Addressing the challenges of heterogeneous, multimodal data integration—spanning diagnostics, control, and multiscale simulation—in nuclear fusion research, including poor cross-device interoperability, inefficient manual annotation, and lack of provenance, this paper proposes an operator-order-aware, reproducible data fusion and annotation framework. The framework integrates temporal-spatial alignment, cross-platform normalization, schema-compliant fusion, uncertainty quantification, and scalable (semi-)automatic annotation. It enables standardized, provenance-rich fusion and annotation of data across disparate fusion devices. Deployed on the DIII-D tokamak, it supports second-level annotation of over 200 plasma discharges per hour, directly applied to ELM detection and confinement-mode classification; model training quality improves significantly. Analysis turnaround time is reduced by over 50×, enabling high-throughput, uncertainty-aware, and reproducible data-driven discovery, physics-informed model validation, and real-time closed-loop control.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web

Abstract

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.

Problem

• Harmonizing heterogeneous multimodal fusion energy datasets
• Enabling scalable uncertainty-aware data fusion and labeling
• Reducing analysis time while improving cross-device comparability

Innovation

• Uncertainty-aware data harmonization and fusion
• Provenance-rich manual and automated labeling
• Reproducible framework reducing analysis time 50X