Simulation Study of In-Situ Polymer Rheology in a Radial Flow Experiment
Not peer reviewed
MetadataShow full item record
The global energy demand is continuously growing, and even though renewable energy sources are becoming increasingly important, oil remains the leading energy fuel to this date, accounting for one third of the consumption.  This indicates that the demand for oil will still be an important energy source in the years to come. The majority of the conventional fields have already been produced, which means that the industry is moving over into a new phase of more complex oil recovery methods.  The amount of heavy oil resources worldwide is estimated to 3.396 billion barrels, and these reservoirs demand more complex recovery methods to be successful. These unconventional recovery methods are expensive and face technological challenges.  Enhanced oil recovery methods are a necessity to optimize the production of heavy oils, and polymer flooding is one of the most widely used EOR methods for unconventional reservoirs.  Polymer flooding is an especially efficient recovery method for heterogeneous reservoirs or heavy crude oil reservoirs with an unfavorable mobility ratio between the oil and the displacing fluid. The aim of this recovery method is to increase the macroscopic sweep efficiency by increasing the viscosity of the injection fluid, and hereby make the mobility ratio more favorable to avoid viscous fingering. Polymers can additionally be used to plug of high permeability zones in a reservoir.  On the other hand, polymer flooding faces multiple challenges concerning field applications, in example low injectivity, degradation, retention, and the potentially high expenses.  Polymer injectivity is an important factor to study, because better understanding may increase the implementation of polymer flooding for EOR. The majority of injectivity studies found in literature today are performed in linear cores, but the flow in a radial core may have great differences from a linear core flood. The most important difference is that linear core floods are performed under steady-state conditions, while radial core floods are under an unsteady-state pressure regime.  In actual field operations, the latter situation is most common and thereby most relevant to study. Experiments in radial cores exert radial flow and are generally better replicates for fluid flow in field operations. Viscosity measurements of synthetic polymers obtained from viscometers has shown to differ from the apparent viscosity in porous media because of its viscoelastic behavior and the elongational flow. This implies that the in-situ polymer rheology is important to investigate further in order to minimize the uncertainties concerning polymer flooding in enhanced oil recovery (EOR) operations. Linear core experiments may overestimate the shear thickening behavior, which in turn can lead to an underestimated polymer injectivity.  This will certainly affect the economical calculations for the EOR-project and may lead to polymer flooding being disposed as an efficient recovery method for the project. The experiment in this thesis was performed at the Centre for Integrated Petroleum Research (CIPR) with the objective of analyzing the in-situ rheology of partially hydrolyzed polyacrylamide (HPAM) by history matching the pressure data collected from fluid flow experiments in a radial core sample. The experimental data, which made the basis for this thesis, was collected from water- and polymer flooding experiments in a radial Bentheimer disc. Automatic history matching was performed to find the permeability field and to estimate the in-situ polymer rheology. The in-situ rheology was estimated for four HPAM solutions with different polymer concentrations, to investigate how the concentration affects the rheological behavior. Analysis of the individual pressure ports additionally contributed to the fluid flow characterization. The tool used to do the history matching was MRST by SINTEF by using the Ensemble Kalman Filter (EnKF), developed at the University of Bergen. The results showed a permeability change in the core sample during the experiment. Both shear thinning and shear thickening behavior was observed for the polymer solutions. There was observed both rate and concentration dependency of the rheological behavior. A distinct shear thickening behavior at the higher injection rates was observed, which is consistent with the existing literature, where this behavior is assigned to the viscoelasticity of HPAM.  Significant shear thinning behavior was observed for the polymer solutions of the highest concentrations. The onset of shear thickening showed shifting towards higher velocities for increasing injection rates. The bulk and in-situ viscosity had similar tangential lines, although different viscosity values.
PublisherThe University of Bergen
Copyright the Author. All rights reserved