Numerical Simulation and Optimization of Gas Mobility Control Techniques During CO2 Sequestration in Cranfield

by Xueying Lu and Mohammad Lotfollahi

CO₂ sequestration in subsurface often suffers from poor volumetric sweep efficiency due to low gas viscosity, low gas density, and formation heterogeneity. This study investigates CO₂ mobility control techniques of Water Alternating Gas (WAG) and Surfactant (or Nanoparticle) Alternating Gas (SAG) to increase CO₂ storage capacity in Cranfield via field-scale simulations and optimization.

UT in-house parallel compositional reservoir simulator, IPARS, can model both compositional flow and geomechanics. IPARS incorporates advanced three-phase relative permeability (UTKR3P) and hysteresis (UTHYST) models [2] with foam models [3] to more accurately simulate WAG and SAG processes . The hysteretic relative permeability model captures local capillary trapping during cyclic injection of liquid and gas. Foam-assisted CO₂ mobility control technique is examined to investigate eminent level of CO₂ capillary trapping by applying a foam model.

Figure 1. Comparison of injection schedules for continuous CO₂ injection, water alternating gas injection, and surfactant alternating gas injection cases.

Figure 1 shows three designed injection schemes for field scale CO₂ sequestration in Cranfield. Same amount of CO₂ is injected at the end of each injection scheme. The simulations were run on Stampede2 with 64 cores for each case.

Figure 2-5 show CO₂ saturation profiles in the reservoir with a view from the bottom during the continuous CO₂ injection, WAG without hysteresis modeling, WAG with hysteresis modeling and SAG, respectively. It is clearly demonstrated that including the observed hysteresis in alternating injection processes significantly increases the performance of these processes. Alternative injection of water or surfactant with CO₂ can suppress gas gravity override and reduce gas channeling through high permeable streaks. During SAG process, foam is generated in the high permeability streaks and upper layers with higher CO₂ flow rates and diverts the CO₂ flow into low permeability regions and bottom layers, leading to more efficient areal and vertical sweep efficiency. In this study, CO₂ storage volume increased by 20.1% and 40.2% of total CO₂ injection volume – compared to continuous CO₂ injection- during four-cycle WAG and SAG processes, respectively.

Figure 2. CO₂ saturation during continuous CO₂ injection, view from the bottom.

Figure 3. CO₂ saturation during water-alternating-gas injection without hysteresis modeling, view from the bottom

Figure 4. CO₂ saturation during water-alternating-gas injection with hysteresis modeling, view from the bottom.

Figure 5. CO₂ saturation during surfactant-alternating-gas injection, view from the bottom.

SAG processes is optimized for injection schedule using UT optimization toolbox with Genetic Algorithm [4]. The number of cycles and length of each cycle are optimized with an objective function equals to cumulative CO₂ storage volume in 20 years. Figure 6 shows the evolution of the objective function. A trend of increase and convergence is achieved in 10 generations. Optimized SAG achieved 15% more CO₂ storage volume and 60% less surfactant and water consumption comparing to basecase SAG.

Figure 6. Evolution of the objective function using the genetic algorithm optimization.

This work on numerical simulation and optimization of gas mobility control techniques during CO₂ sequestration is currently in preparation [1] and a talk was recently given on the topic at the 2017 SIAM Geosciences Conference in Erlangen, Germany. Our simulation results suggest that foam has promising applications in increasing CO₂ storage volume during field scale CO2 sequestrations. The results also demonstrated the capacities of our well developed toolsets for screening and designing field scale CO₂ sequestration operations. In the near future, the team will work on sensitivity analysis of optimization parameters, and foam flow modeling in a coupled compositional- geomechanics framework.

References

[1] Lu, X., Lotfollahi M., Ganis B., Min B., & Wheeler, M. F., Numerical simulation and optimization of gas mobility control techniques during CO2 sequestration in Cranfield, in preparation, 2017

[2] Beygi, M. R., Delshad, M., Pudugramam, V. S., Pope, G. A., & Wheeler, M. F. (2015). Novel three-phase compositional relative permeability and three-phase hysteresis models. SPE Journal, 20(01), 21-34.

[3] Lotfollahi, M., Kim, I., Beygi, M. R., Worthen, A. J., Huh, C., Johnston, K. P., Wheeler, M. F. & DiCarlo, D. A. (2017). Foam Generation Hysteresis in Porous Media: Experiments and New Insights. Transport in Porous Media, 116(2), 687-703.

[4] Min, B., Wheeler, M. F., & Sun, A. Y. (2017). Parallel Multiobjective Optimization for the Coupled Compositional/Geomechanical Modeling of Pulse Testing. In SPE Reservoir Simulation Conference. Society of Petroleum Engineers.