Machine Learning Predictive Models to Estimate the Minimum Miscibility Pressure of CO2-Oil System

One of the most critical parameters of the CO2 injection (for EOR purposes) is the Minimum Miscibility Pressure MMP. The determination of this parameter is crucial for the success of the operation. Different experimental, analytical, and statistical technics are used to predict the MMP. Nevertheless, experimental technics are costly and tedious, while correlations are used for specific reservoir conditions. Based on that, the purpose of this paper is to build machine learning models aiming to predict the MMP efficiently and in broad-based reservoir conditions. Two ML models are proposed for both pure CO2 and non-pure CO2 injection. An important amount of data collected from literature is used in this work. The ANN and SVR-GA models have shown enhanced performance comparing to existing correlations in literature for both the pure and non-pure models, with a coefficient of R2 0.98, 0.93 and 0.96, 0.93 respectively, which confirms that the proposed models are reliable and ready to use.

Modified Method of Characteristics for Generating EOR Oil Recovery Curves

The paper describes a fast and approximate 1D simulation algorithm for calculating the percent recovery that can be obtained from an oil reservoir if gas injection is carried out at a pressure lower than the minimum miscibility pressure. The algorithm is based on the Method of Characteristics. While a conventional 1D reservoir simulation of a gas injection scenario may take minutes or even hours, the proposed algorithm allows a full evaluation of the recovery to be completed within seconds. To make the method numerically robust, a number of approximations were needed. The result is an extremely fast algorithm that not only provides a good estimate of the recovery obtained by gas injection, but also gives a good visualization of how the gas displaces the oil.

Minimum Miscibility Pressure of Gas Injection in Unconventional Reservoirs

Unconvnetional reservoirs are predominantly consisted of nanoscale pores. The strong confinement effect within nanopores imposes significant deviations to the confined fluid phase behavior. Minimum miscibility pressure (MMP) in unconventional reservoirs, as a parameter highly related to the phase behavior of confined fluids, is inevitably affected by the nanoscale confinement. The objective of this work is to investigate the impact of nanoscale confinement on MMP of unconventional reservoir fluids and to recognize a reliable theoretical approach to determine the MMP values in unconventional reservoirs. A modified Peng-Robinson equation of state (PR EOS) applicable for confined fluid characterization is applied to perform the EOS simulation of the vanishing interfacial tension (VIT) experiments. The MMP of a binary mixture at bulk and 50 nm are obtained via the VIT simulation. Meanwhile, the multiple mixing cell (MMC) algorithm coupled with the modified PR EOS is applied to compute the MMP for the same binary system. Comparison of the calculated results to the experimental values recognize that the MMC approach has higher accuracy in determining the MMP of confined fluid systems. Moreover, this approach is then applied to predict the MMP values of both Bakken and Eagle Ford oil at different pore sizes with various injected gases. Results demonstrate that the nanoscale confinement causes drastic suppression to the MMP of unconventional reservoir fluids and the suppression rate increases with decreasing pore size. The drastic suppression of MMP is highly favorable for the miscible gas injection EOR in unconventional reservoirs.

The Effect of Formation Water Salinity on the Minimum Miscibility Pressure of CO2-Crude Oil for Y Oilfield

CO2 miscible flooding is an important technology for enhancing oil recovery and greenhouse gas storage in the world. As a tertiary recovery technology, it is usually applied after water flooding. Therefore, the actual reservoirs usually contain a lot of injected water in addition to connate water. The salinity of these formation waters varies from place to place. CO2 is an acid gas. After it is injected into the reservoir, it easily reacts with formation water and rock and affects the physical properties of the reservoir. However, no research results have been reported whether this reaction affects the minimum miscibility pressure (MMP) of CO2-crude oil, a key parameter determining miscible flooding in formation water. Based on CO2-formation water–rock interaction experiments, this paper uses the core flooding method to measure the CO2-crude oil MMP under different salinity in formation water. Results show that CO2 causes a formation water pH decrease from 7.4 to 6.5 due to its dissolution in formation water. At the same time, CO2 reacts with formation water, albite, potassium feldspar, and carbonate minerals in the cores to generate silicate and carbonate precipitates, which could migrate to the pore throat together with the released clay particles. Overall, CO2 increased core porosity by 5.63% and reduced core permeability by 7.43%. In addition, when the salinity of formation water in cores was 0, 4,767, and 6,778 mg/L, the MMP of CO2-crude oil was 20.58, 19.85, and 19.32 MPa, respectively. In other words, the MMP of CO2-crude oil decreased with the increase of salinity of formation water.

Determination of Minimum Miscibility Pressure of CO2–Oil System: A Molecular Dynamics Study

CO2 enhanced oil recovery (CO2-EOR) has become significantly crucial to the petroleum industry, in particular, CO2 miscible flooding can greatly improve the efficiency of EOR. Minimum miscibility pressure (MMP) is a vital factor affecting CO2 flooding, which determines the yield and economic benefit of oil recovery. Therefore, it is important to predict this property for a successful field development plan. In this study, a novel model based on molecular dynamics to determine MMP was developed. The model characterized a miscible state by calculating the ratio of CO2 and crude oil atoms that pass through the initial interface. The whole process was not affected by other external objective factors. We compared our model with several famous empirical correlations, and obtained satisfactory results—the relative errors were 8.53% and 13.71% for the two equations derived from our model. Furthermore, we found the MMPs predicted by different reference materials (i.e., CO2/crude oil) were approximately linear (R2 = 0.955). We also confirmed the linear relationship between MMP and reservoir temperature (TR). The correlation coefficient was about 0.15 MPa/K in the present study.

A Simulation Study of the Effect of Injecting Carbon Dioxide with Nitrogen or Lean Gas on the Minimum Miscibility Pressure

The objective of this study is to determine the effects of the concentration of injected gases on recovery factors. CO2 has been used for flooding because it requires low injection pressures to achieve miscibility. However, the availability of CO2 is an issue. CO2 required for the process is not sufficient for the reservoir under consideration. Moreover, its benefit-cost ratio (b/c) represents another issue; higher volumes of CO2 increases the economic expenditures. An alternative is to inject lean gas or nitrogen along with CO2 to achieve better recoveries at optimum costs and suitable pressure. Slim tube simulation using a commercial simulator is utilized to measure the minimum miscibility pressure (MMP) of an injected gas mixture. The mixture contains CO2/N2 or CO2/lean gas. Using different concentrations in the injected mixture, an effect on the recovery factor is studied. The first 30 runs have 100% concentration of each individual gases i.e. N2, CO2 and lean gas. Based on these runs, the simulation model is validated using the co-relations present in the literature. Gas mixtures of CO2/N2 and CO2/lean gas were then simulated using the compositional model to test the effect on MMP by varying the concentration of each gas in the mixture. By changing the volumes of the gas in the injected mixture, we can find the optimum concentration of each component in the mixture in terms of MMP. From the results obtained through simulation, it can be deduced that higher percentages of CO2 in the mixture would result in reduced minimum miscibility pressure. The addition of a secondary slug to the injected CO2 fluid increases the pressure required to achieve miscibility. Of N2 and Lean gas, Lean gas provided better results as it showed low miscibility pressure responses compared to the same amount of N2 gas. For example, for a case, 50% CO2 and 50 %N2 or lean gas mixture, the MMP for the lean gas mixture is 3500 Psi, while for N2 mixture it was 4667 Psi. However, lean gas is expensive as compared to N2 and N2 is easily available. N2, if used in optimum concentration along with CO2 can produce greater recoveries keeping the process cost-effective while satisfying other constraints. CO2 is widely used for miscible injection, but it presents problems like costs, corrosion, and asphaltene deposition etc. The study can give an idea of the success of carrying out EOR through gas flooding by using N2 and CO2 to enhance recovery at low cost. N2 is easily available from air and it is cheap.

Analysis of Nitrogen Minimum Miscibility Pressure MMP and Its Impact on Instability of Asphaltene Aggregates - An Experimental Study

Minimum miscibility pressure (MMP) is a critical parameter when undergoing miscible gas injection operations for enhanced oil recovery (EOR). Miscibility has become a major term in designing the gas injection process. When the miscible gas contacts the reservoir oil, it causes changes in the basic oil properties, affecting reservoir oil composition and equilibrium conditions. Changes in conditions may also favor flocculation and deposition of organic solids, mainly asphaltene, which were previously in thermodynamic equilibrium. The main purpose of this study is to investigate how the most important parameters, such as oil temperature and oil viscosity, could affect the nitrogen (N2) MMP and the instability of asphaltene aggregation. Three sets of experiments were conducted: first, the determination of MMP was performed using a slim-tube packed with sand. The impact of crude oil viscosity using 32, 19, and 5.7 cp; and temperature using 32, 45, and 70 °C, were investigated. The results showed that the N2 MMP decreased when crude oil temperature increased. The temperature is inversely proportional to the N2 MMP due to the N2 remaining in a gaseous phase at the same conditions. In terms of viscosity, the MMP for N2 was found to decrease with the reduction in oil viscosity. Second, the effect of miscibility N2 injection pressure on asphaltene aggregation using 750 psi (below miscible pressure) and 1500 psi (at miscible pressure) was investigated using a specially designed filtration vessel. Various filter membrane pores sizes were placed inside the vessel to highlight the effect of asphaltene molecules on plugging the unconventional pore structure. The results demonstrated that increasing the pressure increased asphaltene weight percentage. The asphaltene weight percent was higher when using miscible injection pressure compared to immiscible injection pressure. Also, the asphaltene weight percentage increased when the pore size structure decreased. Finally, the visualization of asphaltene deposition over time was conducted, and the results showed that asphaltene particles started to precipitate after 2 hours. After 12 hours, the colloidal asphaltenes were fully precipitated.

Minimum Miscibility Pressure of CO2/Hydrocarbon Systems Within Nanopores

Minimum miscibility pressure (MMP), as a key parameter for the miscible gas injection enhanced oil recovery (EOR) in unconventional reservoirs, is affected by the dominance of nanoscale pores. The objective of this work is to investigate the impact of nanoscale confinement on MMP of CO2/hydrocarbon systems and to compare the accuracy of different theoretical approaches in calculating MMP of confined fluid systems. A modified PR EOS applicable for confined fluid characterization is applied to perform the EOS simulation of the vanishing interfacial tension (VIT) experiments. The MMP of multiple CO2/hydrocarbon systems at different pore sizes are obtained via the VIT simulations. Meanwhile, the multiple mixing cell (MMC) algorithm coupled with the same modified PR EOS is applied to compute the MMP for the same fluid systems. Comparison of these results to the experimental values recognize that the MMC approach has higher accuracy in determining the MMP of confined fluid systems. Moreover, nanoscale confinement results in the drastic suppression of MMP and the suppression rate increases with decreasing pore size. The drastic suppression of MMP is highly favorable for the miscible gas injection EOR in unconventional reservoirs.