Simulation of polymer flooding

Bondor, P.L., Hirasaki, G.J., Tham, M.J., 1972. Mathematical simulation of polymer flooding in... 

Masuda, Y., Tang, K., Miyazawa, M., Tanaka, S., 1992. ID simulation of polymer flooding including the viscoelastic effect of polymer solution. SPEREE (May), 247-252. 

ZEITO, G. A., "Three Dimensional Numerical Simulation of Polymer Flooding in Homogeneous and Heterogeneous Systems", SPE 2186, 43rd Ann. Fall Conf. of SPE, Houston, Texas, Sept. 27-Oct. 2, 1968. 

Bondor, P. L., Hirasaki, G. J., and Tham, M. J. Mathematical Simulation of Polymer Flooding in Complex Reservoirs , SPE 3524 paper presented at the 46th Annual Fall Meeting of SPE of AIME, New Orleans, Oct. 3-6, 1971. 

In this chapter, we have discussed the rheological behaviour of non-Newtonian polymer solutions in flow through porous media. The practical objective of this is to establish the apparent viscosity versus flow rate vs. Q or vs. 7pjn) expressions that can be used in polymer simulations in reservoirs. How this information is used in the numerical simulation of polymer flooding is discussed in more detail in Chapter 8. Much work has been reported on the in-situ rheology of inelastic polymers such as xanthan and flexible coil synthetic polymers such as HPAM, PEO etc. 

Earlier parts of this book have discussed the various aspects of polymer structure, stability, solution behaviour, in-situ rheology and transport in porous media that are relevant to their ultimate task of improving oil recovery. In this chapter, an attempt is made to pull these strands together by describing the main mechanisms of polymer oil displacement processes in reservoir systems. For this purpose, the main multiphase flow equations that may be used in the design and simulation of polymer floods are developed, along with some simpler solutions for certain limiting cases. 

Naiki, M. (1979) Numerical simulation of polymer flooding including the effects of salinity. PhD Dissertation, University of Texas at Austin, TX. 

Zeito, G. A. (1968) Three dimensional numerical simulation of polymer flooding in homogeneous and heterogeneous systems. SPE 2186, Proceedings of the SPE 43rd Annual Fall Conference, Houston, TX, 27 September to 2 October 1968. 

Masuda, Y., K.-C. Tang, et al. (1992). "ID Simulation of Polymer Flooding Including the Viscoelastic Effect of Polymer Solution." SPE Reservoir Engineeiing(05). 

Sengul, M.M, eted. Simulation of Polymer Flood With a Propagating Hydraulic Fracture, paper SPE 13125 presented at the 1984 SPE... 

Zeito, G.A., 1968. Three-Dimensional Numerical Simulation of Polymer Flooding in Homogeneous and Heterogeneous Systems. Paper SPE 2186 presented at the SPE Annual Meeting, Houston, 29 September-2 Pctober. DPI 10.2118/2186-MS. 

H. Dakhlia. A simulation study of polymer flooding and surfactant flooding using horizontal wells. PhD thesis, Texas Univ, Austin, 1995. 

Although alkaline flooding only is not conducted as often as polymer flooding or surfactant flooding, alkaline injection is conducted together with surfactant and polymer injection. Simulation of alkaline flooding is very difficult because of complex chemical reactions. These complex reactions include at least the following ... 

The effects of adsorption on the propagation of polymer slugs through oil reservoirs determine the efficiency of polymer flooding for increasing oil recovery. However, up to now the adsorption laws introduced in polymer flood simulation models were oversimplified due to a lack of knowledge. 

K and n obtained from steady shear experiments do not describe flow in porous rocks. This finding has significant impact on simulation of polymer injection rates in potential polymer floods and permeability modification projects. Injection rates predicted using Blake-Kozeny models will be less than can be attained in practice. Therefore, it is necessary to determine... 

Ma Chunhua, Zheng Hao, Song Kaoping, et al. Remaining Oil Distribution Laws and Numerical Simulation after Polymer Flooding in the West Block of Beierdong Area[J]. Journal of Oil and Gas Technology (J. JPI),... 

The main objective of this chapter is to include the above phenomena in a suitable transport equation to describe the flow of polymer species through porous media. It has been found that terms describing these effects may be included in generalised convection-dispersion equations which appear to give a satisfactory macroscopic description of the processes in that they reproduce the main features observed in laboratory core flood experiments. It is these single-phase transport equations which provide the basis for simulation of polymer transport through porous media in the multiphase... 

In simulating field polymer floods, an important issue which may have to be considered is the thermal effect of injecting fairly cool water (say at 70-100°F) into hotter reservoir systems. This may cause a temperature gradient across the flooded zone which is cool at the injectors and hot in the main bulk of the reservoir. Since certain important polymer properties depend on temperature (e.g. the solution viscosity—and that of oil and reservoir brine—and the chemical degradation rate), simulation models should take this effect into account when it is important. In addition, the cooler zone close to the injector will affect the local apparent viscosity and hence the injectivity of the polymer solution. In order to perform such calculations, the simulator must include a heat balance equation from which temperature distributions are found. This has not been done on a routine basis until recently (Sorbie etal, 1982 Clifford and Sorbie, 1985 Scott etai, 1985 Scott et ai, 1987), and the effect of the temperature on polymer flooding efficiency will be discussed later in this chapter. 

In order to evaluate projects using the usual type of project economic analysis, it is important to have predicted recovery profiles for the continued waterflood and for a range of polymer flooding strategies. The only way of obtaining these data is to perform reservoir simulation calculations of the proposed projects. Although these calculations may be subject to many uncertainties and difficulties, they are currently the only available means of assessing projects of this type in a rational and consistent manner for a specific reservoir. For this reason, polymer flood simulation is discussed in some detail in this and other chapters. 

Preliminary evaluation of polymer flooding (2-D) Improved waterflood history match (2-D, 3-D) Simulation of tracer flood (2-D, 3-D)... 

Keeping the above remarks in mind, the stages in a full programme of simulation of a pilot polymer flood would include the following four steps preliminary 2-D evaluation simulations, improved waterflood history match, simulation of tracer flood (possibly) and further polymer simulations. Each of these stages is discussed in turn below. 

Further polymer simulations. When this stage is reached all of the preliminary polymer flood simulations should have been completed and a reasonable finer grid model of the polymer pilot area should be near completion. Some experimental data on the polymer may also be available for the particular application but this may not be extensive since laboratory work is usually much slower than simulation studies. The sequence of polymer flood calculations which should now be carried out is as follows ... 

In a polymer-augmented waterflood, polymer is injected continuously at the initial polymer concentration for a limited period. Reducing the polymer concentration systematically as more PV s are injected (as depicted in Fig. 5.1) is the most cost-effective method to conduct a flood. After sufficient polymer has been injected, the polymer slug is displaced through the reservoir by injecting water. Polymer selection, injected concentration, polymer volume injected, and the method of reducing the polymer concentration with PV s of fluid injected ai-e determined from data obtained in laboratory experiments and by simulating the polymer flood with computer models. i... 

The impact of polymer flooding on the displacement performance of a linear reservoir with noncommunicating layers can be predicted by several techniques. If a polymer-flood model, such as that used to generate the polymer-flood performance in Examples 5.7 through 5.10, is available, the displacement performance can be simulated easily when the pressure drop is constant. A similar calculation technique can be developed for constant iryection rate, but the computations are complex and are not included in this text. 

Polymer-Augmented Waterflood in Patterns or Other Arrangements of Production and Injection Wells. Estimation of polymer flood performance in pattern floods or other flooding programs where the flow is not linear requires the use of more complex models or reservoir simulators. An adaptation of a black-oil reservoir simulator is available in the public domain. Most reservoir software companies have polymer flood simulators of various degrees of complexity. Application of reservoir simulators is beyond the scope of this text. 

Maitin, B.K., Daboul., B. and Sohn, W.O. Numerical Simulation for Planning and Evaluation of Polymer Flood Process A Field Perform-anceAnalysrs, paperSPE 17631 presented atthel988SPBIntl.Meet-ing on Petroleum Engineering, Tianjin, Nov. 1-4. 

A computer simulation study on the impact of different parameters on the economics of polymer flooding indicated that the percentage of polymer which remains in the reservoir in immobile fluids and is not adsorbed ranges from 20 to 50 per cent of the injected volume, depending on reservoir conditions and the WOR at the economic limit of production (9). 

To simulate a polymer flood or a polymer bank in a layered, two-dimensional cross-section of a reservoir, a chain of points for the front of the polymer bank and another chain for the rear is defined for each layer of the model. Fig. 5 depicts such a model. Each layer has constant polymer... 

In this paper, we proceed to outline the basic architecture of a SL simulator with a polymer option. The physics of polymer flooding is the same as that being used in estabhshed FD codes. We discuss advantages and disadvantages of the formulation and present numerical experiments in ID, 2D, and 3D to illustrate our results. 

---