Unsupervised learning monitors the carbon-dioxide plume in the subsurface carbon storage reservoir

作者：

Highlights：

• Developed a scalable workflow for visualizing the subsurface CO2 plume.

• Fourier and wavelet transforms are crucial for the visualization.

• Multi-level clustering is used to handle the data imbalance.

• Scores/indices confirm the reliability and consistency of visualization.

• ANOVA F-Test and Kendall’s Tau reveal the geophysical signatures.

摘要

•Developed a scalable workflow for visualizing the subsurface CO2 plume.•Fourier and wavelet transforms are crucial for the visualization.•Multi-level clustering is used to handle the data imbalance.•Scores/indices confirm the reliability and consistency of visualization.•ANOVA F-Test and Kendall’s Tau reveal the geophysical signatures.

论文关键词：Carbon sequestration,Carbon storage,Unsupervised learning,Statistical tests,Clustering,Visualization

论文评审过程：Received 10 May 2021, Revised 20 March 2022, Accepted 8 April 2022, Available online 12 April 2022, Version of Record 20 April 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117216