An official website of the United States government
Numerical simulation is a commonly employed technique for studying carbon dioxide (CO2) storage processes in porous media, particularly saline aquifers. It enables the representation of diverse trapping mechanisms and the assessment of CO2 retention capacity within the subsurface. The intricate physicochemical phenomena involved necessitate the incorporation of multiphase flow, accurate depiction of fluid and rock properties, and their interactions. Among these factors, geochemical reaction rates and mechanisms are pivotal for successful CO2 trapping in carbonate reactive rocks. However, research on kinetic parameters and the influence of lithology on CO2 storage remains limited. This limitation is partly due to the challenges faced in laboratory experiments, where the time scale of the reactions and the lack of in situ conditions hinder accurate measurement of mineral reaction rates. This study employs proxy models constructed using response surfaces calibrated with simulation results to address uncertainties associated with geochemical reactions. Monte Carlo simulation is utilized to explore a broader range of parameters and identify influential factors affecting CO2 mineralization. The findings indicate that an open database containing kinetic parameters can support uncertainty assessment. Additionally, the proxy models effectively represent objective functions related to CO2 injectivity and mineralization, with calcite dissolution playing a predominant role. pH, calcite concentration, and CO2 injection rate significantly impact dolomite precipitation, while quartz content remains unaffected.
more »
« less
Parametric Study of CO2 Sequestration in Deep Saline Aquifers Using Data-Driven Models
Abstract Large-scale geo-sequestration of anthropogenic carbon dioxide (CO2) is one of the most promising methods to mitigate the effects of climate change without significant stress on the current energy infrastructure. However, the successful implementation of CO2 sequestration projects in suitable geological formations, such as deep saline aquifers and depleted hydrocarbon reservoirs, is contingent upon the optimal selection of decision parameters constrained by several key uncertainty parameters. This study performs an in-depth parametric analysis of different CO2 injection scenarios (water-alternating gas, continuous, intermittent) for aquifers with varying petrophysical properties. The petrophysical properties evaluated in this study include aquifer permeability, porosity, relative permeability, critical gas saturation, and others. Based on the extensive data collected from the literature, we generated a large set of simulated data for different operating conditions and geological settings, which is used to formulate a proxy model using different machine learning methods. The injection is run for 25 years with 275 years of post-injection monitoring. The results demonstrated the effectiveness of the machine learning models in predicting the CO2 trapping mechanism with a negligible prediction error while ensuring a low computational time. Each model demonstrated acceptable accuracy (R2 >0.93), with the XGBoost model showing the best accuracy with an R2 value of 0.999, 0.995, and 0.985 for predicting the dissolved, trapped, and mobile phase CO2. Finally, a feature importance analysis is conducted to understand the effect of different petrophysical properties on CO2 trapping mechanisms. The WAG process exhibited a higher CO2 dissolution than the continuous or intermittent CO2 injection process. The porosity and permeability are the most influential features for predicting the fate of the injected CO2. The results from this study show that the data-driven proxy models can be used as a computationally efficient alternative to optimize CO2 sequestration operations in deep saline aquifers effectively.
more »
« less
- Award ID(s):
- 2436996
- PAR ID:
- 10577232
- Publisher / Repository:
- SPE
- Date Published:
- Format(s):
- Medium: X
- Location:
- Palo Alto, California, USA
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
This study introduces an innovative physics-informed machine learning approach to predict the cumulative oil production, CO2 stored, and CO2 adsorbed during CO2 injection. The developed model can be utilized to maximize both oil recovery and CO2 storage in naturally fractured reservoirs (NDR). While existing research has largely focused on maximizing oil recovery and net present value (NPV) through CO2 injection, our methodology integrates comprehensive physics data to fine-tune gas injection processes. We consider a wide span of reservoir characteristics in five-spot injection pattern. We developed dual-permeability, naturally fractured numerical compositional models incorporating sorption to simulate flow mechanisms during injection accurately. Using Langmuir's adsorption dynamics in our simulation integrates well designs and injection strategies. Data from these simulations were trained using Artificial Neural Networks (ANN) to create a robust production prediction proxy model for different gas injection designs. The ANN model demonstrates high accuracy, reflecting its potential for optimizing oil production and carbon storage. The developed model facilitated the screening and optimization process for the considered system as it provides the optimized design in a few seconds compared to days. This study not only highlights the cost-efficiency of using machine learning algorithms in optimization processes but also deepens our understanding of their potential applications in enhanced oil recovery (EOR) and carbon capture and storage (CCUS). The developed models can be used to provide new insights into the interactions between fluid compositions and the complex physical phenomena of sorption, underscoring the transformative potential of integrating advanced machine learning with traditional petroleum engineering techniques.more » « less
-
Abstract Dissolution trapping is one of the most dominant mechanisms for CO2 storage in subsurface porous media saturated with brine. The CO2 dissolution rate and overall fluid flow dynamics in subsurface formations can vary significantly based on permeability variation. Although some numerical simulations have focused on these factors, detailed flow behavior analysis under nonuniform permeability distribution needs further study. For this purpose, we conduct simulations on the flow behavior of CO2-dissolved brine in two different heterogeneous media. The spatial permeability variations in the cell enable the analysis of complex subsurface storage phenomena, such as changes in finger morphology and preferential dissolution path. Finally, the amount of CO2 dissolved was compared between each case, based on which we draw informed conclusions about CO2 storage sites. The results demonstrated a preferential movement of CO2-dissolved regions toward high permeability regions, whereas a poor sweep efficiency was observed due to minimum dissolution in areas with lower permeability. Furthermore, simulation results also reveal uneven CO2 concentration inside the convective fingers. This study provides fundamental insight into the change in flow behavior at heterogeneous regions, which could be translated into saline aquifer conditions. The proposed workflow in this study could be extended further to analyze complex heterogeneous storage systems at different flow regimes.more » « less
-
Abstract Geological carbon dioxide (CO2) sequestration (GCS) in deep saline aquifers is a promising solution to mitigate the impact of anthropogenic CO2emissions on global climate change. CO2dissolved in formation water increases the solution density and can lead to miscible density‐driven downward convection, which significantly accelerates the dissolution trapping of injected CO2. Experimental studies on miscible density‐driven convection have been limited. In the laboratory, we found an empirical linear correlation between reflected green light intensity and solute concentration, which enabledin situmeasurements of solute concentrations in the spatial and temporal domains and consequently the mass flux across the top boundary of the porous medium. Using the novel experimental techniques, we determined the critical Rayleigh‐Darcy number and critical time scales for the onset of density‐driven instability and convective dissolution. This is the first study to determine these critical system parameters using laboratory experiments.more » « less
-
SUMMARY Geological Carbon Storage (GCS) is one of the most viable climate-change mitigating net-negative CO2-emission technologies for large-scale CO2 sequestration. However, subsurface complexities and reservoir heterogeneity demand a systematic approach to uncertainty quantification to ensure both containment and conformance, as well as to optimize operations. As a step toward a digital twin for monitoring and control of underground storage, we introduce a new machine-learning-based data-assimilation framework validated on realistic numerical simulations. The proposed digital shadow combines simulation-based inference (SBI) with a novel neural adaptation of a recently developed nonlinear ensemble filtering technique. To characterize the posterior distribution of CO2 plume states (saturation and pressure) conditioned on multimodal time-lapse data, consisting of imaged surface seismic and well-log data, a generic recursive scheme is employed, where neural networks are trained on simulated ensembles for the time-advanced state and observations. Once trained, the digital shadow infers the state as time-lapse field data become available. Unlike ensemble Kalman filtering, corrections to predicted states are computed via a learned nonlinear prior-to-posterior mapping that supports non-Gaussian statistics and nonlinear models for the dynamics and observations. Training and inference are facilitated by the combined use of conditional invertible neural networks and bespoke physics-based summary statistics. Starting with a probabilistic permeability model derived from a baseline seismic survey, the digital shadow is validated against unseen simulated ground-truth time-lapse data. Results show that injection-site-specific uncertainty in permeability can be incorporated into state uncertainty, and the highest reconstruction quality is achieved when conditioning on both seismic and wellbore data. Despite incomplete permeability knowledge, the digital shadow accurately tracks the subsurface state throughout a realistic CO2 injection project. This work establishes the first proof-of-concept for an uncertainty-aware, scalable digital shadow, laying the foundation for a digital twin to optimize underground storage operations.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.2118/218906-MS
