<![CDATA[Carbon dioxide, CO2 emissions have risen precipitously over the last century, wreaking havoc on the atmosphere. Carbon Capture and Sequestration (CCS) techniques are being used to inject as much CO2 as possible and meet emission reduction targets with the fewest number of wells possible for economic reasons. However, CO2 injectivity is being reduced in sandstone formations due to significant CO2-brine-rock interactions in the form of salt precipitation and fines migration. The purpose of this project is to develop a regression model using linear regression and neural networks to correlate the combined effect of fines migration and salt precipitation on CO2 injectivity as a function of injection flow rates, brine salinities, particle sizes, and particle concentrations. Statistical analysis demonstrates that the neural network model has a reliable fit of 0.9882 in R Square and could be used to accurately predict the permeability changes expected during CO2 injection in sandstones.]]>
Copyrights © 2021
