TCV Tokamak Plasma Turbulence Data Analyzed with Machine Learning

Cross-field particle transport data from the boundary region of magnetically confined fusion plasmas on the Tokamak à Configuration Variable (TCV). The dataset was created by Han, W.; Golfinopoulos, T.; Terry, J.L.; Marmar, E.S. and last updated on 2026-05-14. It likely contains results from a machine learning blob-tracking approach applied to Gas Puff Imaging data for plasmas with varying triangularity.

Use Cases

• Modeling cross-field particle transport based on blob dynamics described in the study
• Analyzing the relationship between plasma triangularity and transport rates based on the experimental conditions mentioned
• Training machine learning models for blob detection and tracking in Gas Puff Imaging data
• Validating simulation results from KN1D, GBS, and SOLPS-ITER against experimental observations
• Investigating correlations between blob count, area, and radial speed under different plasma conditions

Strengths

• Data is derived from experiments on the Tokamak à Configuration Variable (TCV), a recognized fusion research device
• Analysis includes results for four distinct plasma triangularity values (+0.38, +0.15, −0.14, and −0.26)
• Methodology employs a novel machine learning blob-tracking approach, suggesting a modern analytical technique

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• Description metadata is limited; actual data quality requires manual inspection after download

Provenance

Source

Plasma Science and Fusion Center Dataverse

Collection Method

Machine learning blob-tracking applied to Gas Puff Imaging (GPI) data from TCV tokamak experiments

Freshness

Last updated 2026-05-14 20:21:37