Randomized Tensor Decomposition for Large-Scale Data Assimilation Problems for Carbon Dioxide Sequestration

Liu, Mingliang; Grana, Dario; Mukerji, Tapan

doi:10.1007/s11004-022-10005-1

Cited by 4 publications

(3 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach seamlessly integrates with SVGD through automatic differentiation. Alternative potential solutions include principal component analysis (Vo & Durlofsky, 2015) and randomized tensor decomposition (M. Liu et al., 2022), but in those methods, we need to manually derive the required gradient terms via adjoint methods. In this study, we used a pre‐trained 1D autoencoder based on initial samples obtained from geostatistical simulations and then perform the SVGD‐AE inversion trace‐by‐trace in 1D.…”

Section: Discussionmentioning

confidence: 99%

“…Another family of optimization methods consists of sequential Monte Carlo methods, including ensemble-based methods and particle filters, in which the distribution of model is represented by an ensemble of model realizations (or particles). M. Liu and Grana (2018) and M. Liu et al (2022) developed a stochastic seismic inversion method based on ensemble-smoother with multiple data assimilation (ES-MDA) and applied it to subsurface characterization. Yardim and Gerstoft (2012) applied particle filter and smoother to track non-volcanic tremor.…”

Section: Introductionmentioning

confidence: 99%

“…M. Liu and Grana (2018) and M. Liu et al. (2022) developed a stochastic seismic inversion method based on ensemble‐smoother with multiple data assimilation (ES‐MDA) and applied it to subsurface characterization. Yardim and Gerstoft (2012) applied particle filter and smoother to track non‐volcanic tremor.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Geostatistical Inversion for Subsurface Characterization Using Stein Variational Gradient Descent With Autoencoder Neural Network: An Application to Geologic Carbon Sequestration

Liu,

Grana,

Mukerji

2024

JGR Solid Earth

View full text Add to dashboard Cite

Geophysical subsurface characterization plays a key role in the success of geologic carbon sequestration (GCS). While deterministic inversion methods are commonly used due to their computational efficiency, they often fail to adequately quantify the model uncertainty, which is essential for informed decision‐making and risk mitigation in GCS projects. In this study, we propose the SVGD‐AE method, a novel geostatistical inversion approach that integrates geophysical data with prior geological knowledge to estimate subsurface properties. SVGD‐AE combines Stein Variational Gradient Descent (SVGD) for sampling high‐dimensional distributions with an autoencoder (AE) neural network for re‐parameterizing reservoir models, aiming to accurately preserve geologic characteristics of reservoir models derived from prior knowledge. Through a synthetic example of pre‐stack seismic inversion, we demonstrate that the SVGD‐AE method outperforms traditional probabilistic methods, particularly in inverse problems with complex posterior distributions. Then, we apply the SVGD‐AE method to the Illinois Basin—Decatur Project (IBDP), a large‐scale CO2 storage initiative in Decatur, Illinois, USA. The resulting petrophysical models with quantified uncertainty enhance our understanding of subsurface properties and have broad implications for the feasibility, decision making, and long‐term safety of CO2 storage at the IBDP.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Geostatistical Inversion for Subsurface Characterization Using Stein Variational Gradient Descent With Autoencoder Neural Network: An Application to Geologic Carbon Sequestration

Liu,

Grana,

Mukerji

2024

JGR Solid Earth

View full text Add to dashboard Cite

show abstract

Stochastic pix2vid: A new spatiotemporal deep learning method for image-to-video synthesis in geologic CO$$_2$$ storage prediction

Morales,

Torres-Verdín,

Pyrcz

2024

Comput Geosci

View full text Add to dashboard Cite

Bayesian Ensemble Kalman Filter for Gaussian Mixture Models

Gryvill,

Grana,

Tjelmeland

2024

Math Geosci

View full text Add to dashboard Cite

Inverse theory and data assimilation methods are commonly used in earth and environmental science studies to predict unknown variables, such as the physical properties of underground rocks, from a set of measured geophysical data, like geophysical seismic or electromagnetic data. A new Bayesian approach based on the ensemble Kalman filter using Gaussian mixture models is presented to overcome the assumption of Gaussian distribution of the unknown variables commonly used in the data assimilation literature and to generalize the algorithm to inverse problems with multimodal probability distributions. In applications of subsurface characterization, the multimodality of the unknown variables is generally due to the presence of different rock types, also known as geological facies. In the proposed method, the weights of the Gaussian mixture model represent the facies proportions, and they follow a Markov chain model. The proposed Bayesian model generates the unknown model parameters conditioned on measured data using a Markov chain Monte Carlo sampler. The validity of the method is demonstrated on a data assimilation problem where the goal is to estimate the posterior distribution of the unknown rock density from a set of repeated measurements of acoustic wave velocity measured at different times. The proposed method provides accurate estimates with efficient computational times.

show abstract

Randomized Tensor Decomposition for Large-Scale Data Assimilation Problems for Carbon Dioxide Sequestration

Cited by 4 publications

References 53 publications

Geostatistical Inversion for Subsurface Characterization Using Stein Variational Gradient Descent With Autoencoder Neural Network: An Application to Geologic Carbon Sequestration

Geostatistical Inversion for Subsurface Characterization Using Stein Variational Gradient Descent With Autoencoder Neural Network: An Application to Geologic Carbon Sequestration

Stochastic pix2vid: A new spatiotemporal deep learning method for image-to-video synthesis in geologic CO$$_2$$ storage prediction

Bayesian Ensemble Kalman Filter for Gaussian Mixture Models

Contact Info

Product

Resources

About