Abstract
CO2 Storage as a technology is one that has been widely applied for the achievement of a greener world. It is considered as the only technology that mitigates global warming whilst, yet, allowing for the continuous use of fossil fuels which power our world today. Trapping efficiency of CO2 in saline aquifers depends on engineering design strategies and aquifer parameters. The range of plausible factors that could affect the efficiency creates the need for an optimization approach to reduce the risk and uncertainties associated with the process.
This paper presents an uncertainty and optimization model for the residual and solubility trapping mechanisms in CO2 storage processes using the design of experiments and response surface methodology for representative aquifer models at field scale. Extensive numerical experiments were conducted using a commercial simulator and an optimum set of conditions was suggested for site based storage activities using the desirability concept. The factors for optimization in this study were broadly classified into aquifer and design parameters based on the controllability or non-controllability of selected factors. Aquifer parameters selected are the rock heterogeneity quantified using the Dykstra Parson coefficient, spatial continuity of properties estimated using the correlation lengths, temperature, mean permeability, anisotropy, residual gas saturation while the design parameters selected are the well type, injection strategy and completion intervals. These factors were selected based on literature search, intuitions and discussions. The aquifer model was kept constant for accurate analysis. The objective functions are the residual trapping coefficient (RTC) and the solubility trapping coefficient (STC).
The process for the optimization technique utilized was a three pronged approach involving screening, characterizing and optimizing the parameters. Screening of parameters was performed using the factorial methods, the characterizing was performed using the Box Behnken response surface methodology while the optimization was performed using desirability functions.
For results analysis, Pareto plots were constructed to show the effect of main factors and interaction factors on the objective functions, response surface plots were also made to show response effects. Results were validated with a commercial simulator. This study provides engineers with a better understanding of the most influential factors on trapping capacity during CO2 storage thus enabling them to make decisions with less risk and more certainty.
CCUS Diagram
Keywords:
flow in porous media, Artificial Intelligence, Fluid Dynamics, Upstream Oil & Gas, enhanced recovery, reservoir simulation, mechanism, objective function, coefficient, solubility
Subjects:
Reservoir Characterization, Reservoir Fluid Dynamics, Improved and Enhanced Recovery, Reservoir Simulation, Flow in porous media
Comments