No abstract

No abstract

Stable dispersions of superparamagnetic nanoparticles that are already in use in biomedicine as image-enhancing agents, also have potential use in subsurface applications. Surface-coated nanoparticles are capable of flowing through micron-size pores across long distances in a reservoir, with modest retention in rock. These particles change the magnetic permeability of the flooded region, and thus can be used to enhance images of the flood. In this paper, we model the propagation of a "ferrofluid" slug in a reservoir and its response to a crosswell magnetic tomography system. This approach to monitoring fluid movement within a reservoir is built on established electromagnetic (EM) conductivity monitoring technology. In this work, however, we investigate the contrast between injected and resident fluids when they have different magnetic permeabilities. Specifically, we highlight the magnetic response at low-frequency (less than 100 Hz) to the magnetic excitations generated by a vertical magnetic dipole source positioned at the injection well. At these frequencies, the induction effect is small, the casing effect is manageable, the crosswell response originates purely from the magnetic contrast in the formation, and changes in fluid conductivities are irrelevant. The sensitivity of the measurements to the magnetic slug is highest when the slug is closest to the source or receivers and lower when the slug is midway in the interwell region. At low frequencies (below 100 Hz in our assumed model), the magnetic response of the ferrofluid slug is largely independent of frequency. As expected for the conductive slug, the sensitivity of the inductive measurements is negligible at low frequencies while significant levels of detectability result at frequencies above 100 Hz. We demonstrate sensitivity to the vertical boundaries of the slug by shifting the vertical position of the excitation source relative to the magnetic slug. The slug geometry plays a key role in determining the magnetic response. Given a fixed volume of ferrofluid and considering several candidate formations for test pilot design, there is an optimum layer thickness which results in the maximum magnetic response. Hydrodynamic dispersion of the slug has negligible effect on the magnetic response during early stages of the waterflood. As the slug travels farther into the formation, however, dispersion reduces the concentration of nanoparticles and the spatial contributions of the magnetic measurements are more diffuse. We illustrate how these low frequency excitation behaviors are consistent with the quasi-static magnetic dipole physics. The fact that the progress of the magnetic slug can be detected at very early stages of the flood, that the traveling slug's vertical boundaries can be identified at low frequencies, and that the magnetic nanoparticles can be sensed well before the actual arrival of the slug at the observer well, provides significant value of using the magnetic contrast agents in crosswell EM tomography.

To accurately predict production in compactible reservoirs, we must use coupled models of fluid flow and mechanical deformation. Staggered-in-time loose coupling of flow and deformation via a high-level numerical interface that repeatedly calls first flow and then mechanics allows us to leverage the decades of work put into individual flow and mechanics simulators while still capturing realistic coupled physics. These two processes are often naturally modeled using different time stepping schemes and different spatial grids-flow should only model the reservoir, whereas mechanics requires a grid that extends to the earth's surface for overburden loading and may extend further than the reservoir in the lateral directions. Although spatial and temporal variability between flow and mechanics can be difficult to accommodate with full coupling, it is easily handled via loose coupling. We calculate the total stress by adding pore pressures to the effective rock stress. In turn, changes in volume strain induce updates to porosity and permeability and, hence, dynamically alter the flow solution during simulation. Incorporating the resulting time-dependent pressures, saturations, and porosities (from coupled flow and mechanics) into Gassmann's equations results in seismic wave velocities and densities that can differ markedly from those calculated from flow alone. In a synthetic numerical experiment based on Belridge field, California, incorporation of coupled flow and mechanical deformation into time-lapse calculations produces compressional wave velocities that differ markedly from those produced by flow alone. In fact, it is the closing of the pores themselves (reduction in permeability) in this example which has the greatest impact on fluid pressures and saturations and, hence, elastic wave parameters such as velocity.

No abstract