Weak Supervision Multigeophysical Inversion for CO2 Saturation Imaging

Authors:

Abstract
In CO2 sequestration projects, multi-physics inversion has been widely used to reconstruct various geophysical properties (such as velocity and conductivity). However, the saturation of CO2 is not feasible through partial differential equations (PDEs), which poses a challenge to traditional multi-physics inversion techniques in directly inverting CO2 saturation from geophysical measurements. Typically, a rock physics model is constructed to tackle this challenge. Nevertheless, the construction of such a model is intricate due to the inherent complexities and uncertainties within subsurface geology and geophysical data. Data-driven inversion methods present an alternative solution, as they can connect CO2 saturation to geophysical measurements and directly learn their relationships from labeled data. Nonetheless, the efficacy of these methods hinges on extensive data labeling, incurring considerable costs. To address this challenge, we propose a novel data-driven technique for the inversion of multi-physics data called Weakly Supervised Multi-Geophysical Inversion (WS-MGI), which reduces the need for labeling and promises to be a more cost-effective solution. In particular, we focus on the multi-physics inversion problem from two geophysical data (electromagnetic (EM) and seismic) to CO2 saturation. By learning the local relationship between velocity and CO2 saturation at a few well logs, we construct the pseudo labels for CO2 saturation, thus enabling the data-driven inversion with few labels. We verify our method using synthetic data based on the Kimberlina storage reservoir in California. Experiments show that compared to the supervised counterpart, our method can achieve similar inversion results with only 2% labels.
Funding Information
  • School of Data Science and Society at the University of North Carolina at Chapel Hill. (Faculty start-up Grant)
  • Los Alamos National Laboratory (20210542MFR)
  • Hydrocarbons and Geothermal Energy Office (Science-Informed Machine Learning to Accelerate Re)

This publication has 67 references indexed in Scilit: