|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
A predictive model (PM) has been developed for water and polymer flooding for use in screening studies and feasibility analysis. The PM requires only a traction of the computing time of numerical simulators, but includes most of the phenomena which affect flood performance. This is accomplished by combining a two-dimensional cross-sectional model (using vertical equilibrium), with areal sweep correlations, and injectivity functions which gives results (oil rate versus time) which are suitable for economic analysis.
The paper presents validations of the PM against simulator, analytical, and field results. A brief sensitivity analysis concludes the paper.
In making decisions about water or polymer flooding prospective reservoirs, some means of estimating future performance is necessary. Many methods have been developed for this purpose, ranging from empirical correlations and screening guides to numerical reservoir simulators.
Waterflood and polymer flood prediction methods can be grouped into four broad categories: "binary" , screening guides, empirical methods, simple analytical methods, and numerical models which require the use of a computer.
The simplest prediction tool for polymer flooding is the "binary" screening guide. These guides give extreme values of reservoir parameters that are considered acceptable for a "successful" polymer flood. They are useful as a guide to some polymer flood. They are useful as a guide to some of the important parameters affecting performance. Their main limitations are that they do not consider the composite effect of all variables, and offer no indication of economic feasibility.
Empirical correlations of waterflood performance have been generated through statistical analysis. These correlations are usually based on reservoir parameters, but do not specifically consider heterogeneities or other factors which may be unique to a particular reservoir. Most do not provide results suitable for economic analysis. provide results suitable for economic analysis. There are several widely-used analytical waterflood performance prediction methods that do not require the use of a computer. These techniques are more reservoir-specific than empirical methods, since they account for actual reservoir and fluid properties to a greater extent. Their primary disadvantages lie in the assumptions made in their formulation. One of the most limiting assumptions common to all analytic methods is that no crossflow occurs, or that individual layers are not influenced by their position in the reservoir or by other layers.
Analytical and graphical methods based on fractional flow, theory for predicting polymer flood behavior have limited application and give incomplete results for purposes of economic analysis. For continuous polymer injection (and other simplifying assumptions), these methods show the effects of mobility reduction and adsorption on the displacement process which is helpful in checking complex models. They are not sufficient for three-dimensional field cases since the effects of finite slug, areal sweep, heterogeneity, and injectivity are absent.
Sophisticated numerical models can account for almost all of the mechanisms known to affect water and polymer flooding. The accuracy of their results is limited primarily by the accuracy and availability of data characterizing the reservoir and the injected fluid. These models have limited application as screening tools because of their extensive data requirements. For preliminary screening, the cost of computing time and data gathering would often be prohibitive. Numerical models are best suited for detailed reservoir analysis and project development work.
Abstract The paper discusses the feasibility study approach of polymer flooding enhanced oil recovery. This work is focused on understanding and quantifying key aspects of polymer flooding and design parameter optimization case. A synthetic reservoir simulation model was employed for the study. The first stage is to identify and understand key factors that have most significant impact to polymer flooding response. There are eight parameters that are considered in the analysis, such as polymer concentration, polymer thermal degradation, polymer injection duration, and polymer-rock properties (adsorption, residual resistance factor, etc.). The impact of each parameter to oil recovery response was sensitized with its low, mid, and high values. The difference of high to low oil recovery output for all parameters was ranked to determine their significance levels. The top three parameters obtained from the sensitivity analysis are polymer injection duration, thermal degradation, and polymer concentration. Sensitivity cases of polymer injectivity and thermal degradation effects were covered in this work. The second stage is to determine optimum design parameters of polymer flooding. The most significant parameters from the sensitivity analysis results were considered for further optimization. Three parameters that were selected for design optimization include polymer injection duration, polymer concentration, and well spacing. An optimization workflow with simplex algorithm is linked with a reservoir simulator to generate optimization cases by varying values of optimized parameters. The optimization iteration stops when the maximum value of the objective function, which is the net revenue, is reached. The optimization cycle was done for rock permeability of 500 md and 1000 md. For a low rock permeability reservoir, the well spacing should be short and a lower polymer concentration is sufficient to provide a good response, in addition to avoiding potential injectivity problem. There should be minimum injectivity problem for reservoir with permeability above 1000 md. It is very important to apply polymer thermal degradation in the simulation model to avoid an optimistic performance prediction. The sensitivity analysis results provide a good understanding on the significance impact of parameters controlling polymer injection response and potential challenges. The optimization approach used in the study aids in investigating many optimization scenario within a short period of time.
Sabirov, Denis Galievich (Gazpromneft STC) | Demenev, Roman Aleksandrovich (Gazpromneft STC) | Isakov, Kirill Dmitrievich (Gazpromneft STC) | Ilyasov, Ilnur Rustamovich (Messoyakhaneftegaz) | Orlov, Alexander Gennadievich (Messoyakhaneftegaz) | Glushchenko, Nikolay Aleksandrovich (Messoyakhaneftegaz)
Abstract Most of the Russian oil fields consist of the complex reservoirs and it is required to apply secondary reservoir development methods, such as waterflooding, in order to increase reservoir development efficiency. However, for highly heterogeneous reservoir with viscous oil, "classical" waterflooding is not enough and there is a need to use enhanced oil recovery methods, one of which is polymer flooding. The prospects polymer solutions injection have been proved in different fields worldwide, including the East-Messoyakhskoye field at the PK1-3 reservoir with high-viscosity oil. At this field, polymer flooding pilots were carried out and taking into account the obtained field data, the geological and dynamic model were updated, which helped to improve the process physics understanding and evaluate the possibility of sweep efficiency increase during project implementation. This paper describes the challenges, difficulties, applied approaches, results and experience obtained in reservoir simulation of polymer flooding.
The complete paper presents a new three-phase relative permeability model for use in chemical-flooding simulators. A model that has been widely used in chemical-flooding simulators for decades has numerical discontinuities that are not physical in nature and that can lead to oscillations in the numerical simulations. The proposed model is simpler, has fewer parameters, and requires fewer experimental data to determine the relative permeability parameters compared with the original model. Two- and three-phase relative permeability measurements at low interfacial tension (IFT) have been published previously, and microemulsion relative permeability models have been proposed in the literature as well. But none of these can model the microemulsion phase across different phase-behavior environments, from oil-in-water, to the middle phase, to water-in-oil emulsions.