On the applicability of simplified models to assess pressure evolution during CO2 injection into multiple site systems
Abstract
Energy systems models show that the availability of large scale CCS is a key sensitivity for the costs of meeting stringent climate change targets. The inclusion of large, regional CO2 storage into energy systems models requires rapid evaluation of the storage capacity for different injection scenarios at multiple sites. Currently, energy systems models do not incorporate limitations to storage imposed by injectivity or pressurization in the reservoir. To this aim, the use of simplified analytical solutions poses a significant advantage over full 3D numerical simulations because they provide a rapid assessment of the pressure response and the plume evolution. In the case of multiple well systems the pressure build-up is often evaluated as the superposition of the analytical solutions for pressure associated to each individual well. However, the non-linearity of the multiphase flow makes the applicability of the superposition principle uncertain, and is some cases invalid.
In this work we investigate the error associated with the use of superposed analytical solutions by means of multiphase numerical simulations of CO2 injection into a simplified homogeneous reservoir. We separately analyze the error associated with the accuracy of analytical solutions and the error associated with the application of the superposition principle for different configurations of number and location of wells. We find that simple superposition of the pressure responses overestimates the pressure build-up as it neglects the presence of the multiple CO2 plumes, which locally increase the mobility. This error, which increases with the number of injecting wells and with shorter distances, is small for early times but then increases with the square root of time. As an example, in the case of continuous injection into 9 wells located in a Cartesian pattern, the pressure build-up evaluated by superposition is overestimated by the 30% compared to the numerical solution at a dimensionless time (square root of wells distance divided by hydraulic diffusivity) equal to 100.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFMMR53A0101D
- Keywords:
-
- 1822 Geomechanics;
- HYDROLOGYDE: 1858 Rocks: chemical properties;
- HYDROLOGYDE: 5114 Permeability and porosity;
- PHYSICAL PROPERTIES OF ROCKSDE: 5139 Transport properties;
- PHYSICAL PROPERTIES OF ROCKS