Copyright © the author(s). This work is licensed under a Creative Commons Attribution 4.0 International License. Improved Oil and Gas Recovery DOI: 10.14800/IOGR.1354 Received January 5, 2025; revised May 10, 2025; accepted July 23, 2025. *Corresponding author: dike.chukwuebuka@futo.edu.ng 1 Enhanced Gas Condensate Recovery by Injection of Produced Hydrocarbon Gas Anthony Ogbaegbe Chikwe, Valentine Igwe, Chukwuebuka Francis Dike*, Federal University of Technology Owerri, Owerri, Nigeria; Gift Aghaulor, Covenant University, Otta, Nigeria; Jude Emeka Odo, Federal University of Technology Owerri, Owerri, Nigeria Abstract Natural gas has become an increasingly favored energy source, meeting more than a quarter of the global energy demand. Among natural gas resources, gas condensate is distinct in that it exhibits liquid separation as pressure diminishes over the reservoir's lifespan, a phenomenon that can markedly influence reservoir productivity. To counteract this and rejuvenate well productivity, chemical and mechanical interventions are implemented. While these methods can yield some measure of success, there is a pressing need for more proactive measures that can preemptively address this issue and reduce operational interruptions. Currently, a significant challenge within the oil industry pertains to attaining high hydrocarbon recovery in gas condensate fields, predominantly due to pressure complications and the presence of active aquifers within these reservoirs. This research investigates various enhanced oil recovery (EOR) methodologies, with a particular emphasis on the enhanced recovery from gas condensate reservoirs through the injection of produced hydrocarbon gas. An alternative methodology was developed and verified, demonstrating efficacy and computational efficiency. Utilizing an inverted five-spot model in a gas condensate reservoir simulation, independent parameters were established, encompassing bottom- hole pressure for production wells and gas injection rate for injection wells. Through the application of a Box- Behnken experimental design, a spectrum of parameter combinations was generated to conduct simulation experiments, yielding outcomes such as cumulative oil production and cumulative gas production. Subsequently, response surface models were constructed using response surface methodology, revealing a dependable correlation between cumulative oil production and cumulative gas production with independent parameters. To ascertain the optimal solution, a multi-objective genetic algorithm was utilized, with the aim of maximizing cumulative oil production and minimizing cumulative gas production. The analysis yielded a singular optimal solution, representing the optimal set of operating conditions for enhanced gas condensate recovery. The findings indicate that the amalgamation of proxy models and optimization algorithms can substantially assist in identifying optimal operating conditions during gas condensate reservoir simulation. This approach not only reduces computational expenses and time demands but also enhances the efficacy of gas condensate recovery. Introduction The global demand for fossil fuels is experiencing significant growth (Kerunwa et al. 2024) due to population growth and development. In Nigeria, the power sector has been unbundled, and the country is facing energy poverty compounded by low oil prices and volatility. To reduce costs and increase productivity, oil companies are seeking ways to address these challenges. Nigeria possesses substantial proven reserves of natural gas, estimated at 180 trillion standard cubic feet, ranking it ninth globally and the largest in Africa (Central Intelligence Agency 2014). The natural gas in Nigeria can be categorized as associated or non-associated (Elehinafe et al. 2022), with a relatively equal distribution ratio between the two. Associated gas refers to gas found together with mailto:dike.chukwuebuka@futo.edu.ng Improved Oil and Gas Recovery 2 oil in a reservoir, while non-associated gas refers to gas found in reservoirs without oil. Gas condensate reservoirs are valuable sources of hydrocarbons, consisting of a mixture of natural gas and liquid condensate. Optimizing the recovery of gas condensate is crucial for economic reasons, leading to increased interest in enhanced recovery techniques. One such method gaining attention is the injection of produced hydrocarbon gas, which offers potential benefits in terms of improved recovery and reservoir performance. Gas condensate reservoirs represent new challenges for the petroleum industry, and it is important to approach them with caution, as applying conventional gas/oil system knowledge without careful examination may lead to incorrect conclusions and practices. This study focuses on enhanced recovery from gas condensate reservoirs through the injection of produced hydrocarbon gas. It covers various aspects, including gas condensate properties, composition, condensate banking, economic considerations, mitigation strategies, challenges, and future directions. Understanding how condensate accumulation impacts productivity and liquid phase composition configuration is crucial for optimizing production strategies, mitigating the effects of condensate accumulation and enhancing the overall retrieval of gas and condensate. Gas Condensate In a gas condensate reservoir, the initial state of the reservoir fluids is primarily in the form of gas phase. As the reservoir undergoes primary production and the pressure within the reservoir decreases, the gas undergoes a phase change, and liquid condensate starts to accumulate. This accumulation is particularly significant in areas such as fractures and the well bottom hole, especially when the in-situ reservoir pressure falls below the dew-point pressure. At this point, the gas starts to condense into liquid form, forming what is known as condensate. However, the condensate does not immediately flow; instead, it accumulates until it reaches a critical saturation level. Generally, three distinct zones can be identified from the wellbore to the reservoir boundary, each with varying concentrations of condensate and gas phases. • Mobile gas and mobile condensate region: This zone is located near the wellbore and consists of both gas and condensate that can freely move within the formation. • Transition zone: This zone encompasses both mobile gas and immobile oil. It acts as a transition area between the mobile gas and condensate region and the gas phase zone without condensate dropout. • Gas phase zone: This zone is characterized by the absence of condensate dropout and primarily contains gas (Penuela and Civian 2000). The existence of trapped condensate in the reservoir has a notable consequence of leaving a substantial volume of high-quality oil unrecovered. This trapped condensate impedes the flow of gas towards the wellbore, causing a decline in gas production. These observations have been corroborated by various studies conducted by researchers such as Moses and Donohoe (1987), Li and Firoozabadi (2000), Pope et al. (2000). Their findings provide valuable insights into the impact of trapped condensate on oil and gas production in gas condensate reservoirs. Figure 1 depicts a gas field in a retrograde state where the temperature exceeds the critical point temperature. The curved lines on the diagram represent the transitions between different phases of the fluid as it moves from the reservoir (indicated by the vertical green line), undergoes cooling while ascending through the wellbore, and eventually reaches the separator. Retrograde condensation in a gas condensate reservoir starts near the wellbore and gradually spreads outward in a radial pattern as the pressure decreases. Improved Oil and Gas Recovery 3 Figure 1—Phase diagram of a retrograde-condensate gas (McCain 1990). Materials and Methods Materials. The materials utilized for this study are a typical gas condensate simulation model, design expert software and eclipse reservoir simulator. The reservoir simulation model is a representative of a real field reservoir while Eclipse reservoir simulator was used to run the reservoir simulation models to evaluate different hydrocarbon gas injection scenarios. Design expert software was used to generate parameter realization for conducting reservoir simulation runs. The reservoir simulation model consists of 9×9×4 grid blocks in the X, Y, and Z directions respectively. The dimensions in the x and y directions is 50 m while each layer in the z direction has dimensions of 10 m. The depth of the top of the reservoir is 2,070 m. The oil water contact and gas oil contacts for a gas condensate reservoir are the same and were set to be equal to 2,110 m. The net pay thickness of the reservoir is 40 m. An inverted five spot was modelled in the gas condensate reservoir such that an injection well was placed at the center while 4 producers were placed at the edges of the reservoir as shown in Figure 2. Methods. The Method comprises of development of input & output data, development of polynomial regression model, and validation & optimization of polynomial regression models. Development of Input and Output Data. The input data was developed using the Box-Behnken design approach in the Design Expert Software. The Box-Behnken method entailed the generation of parameter realization based on introduced minimum and maximum value depicted in Table 1. These generated input data from Design Expert Software was introduced into Eclipse Simulator Software for Simulation. The result of the eclipse simulation run yielded cumulative oi production and cumulative gas injection and cumulative gas production depicted in Table 2 Table 1—Minimum and Maximum values of input data. Factor Name Units Min. Max. Mean Std. Dev. Bottom-hole pressure of P1 X1 Bar 20.00 45.00 32.50 7.45 Bottom-hole pressure of P2 X2 Bar 20.00 45.00 32.50 7.45 Bottom-hole pressure of P3 X3 Bar 20.00 45.00 32.50 7.45 Bottom-hole pressure of P4 X4 Bar 20.00 45.00 32.50 7.45 Gas injection rate of I1 X5 M3/day 100.00 1000.00 550.00 268.33 Improved Oil and Gas Recovery 4 (a) Gas saturation (b)Oil saturation Figure 2—3D modeling of an inverted five spot. Development of Polynomial Regression Model. The developed input and output datasets were entered into Design Expert software from which analysis of variance (ANOVA) was conducted to determine the parameters that were significant to the model. Validation and Optimization of Polynomial Regression Models. The models were validated using cross plots and statistical error evaluation, while the optimization was carried out using polynomial regression models. The polynomial regression models were coded in a MATLAB script file so that an optimization algorithm within MATLAB’s global optimization toolbox can be used to run the models. The models were run using multi- objective genetic algorithm because two objective functions represented by two polynomial regression models were considered. Improved Oil and Gas Recovery 5 Table 2—Input and output data obtained BBD and reservoir simulation. Run A:X1 B:X2 C:X3 D:X4 E:X5 R1-CGP R2-COP R3-CGI 1 32.5 32.5 32.5 32.5 550 413.905 59.8002 1.825 2 20 45 32.5 32.5 550 452.963 66.4864 10.0375 3 32.5 32.5 32.5 32.5 550 433.875 63.188 10.0375 4 32.5 32.5 20 20 550 453.017 66.6856 10.0375 5 32.5 32.5 20 32.5 100 452.204 66.3689 1.825 6 20 32.5 32.5 20 550 453.016 66.6909 10.0375 7 20 32.5 32.5 32.5 1000 453.773 66.5757 18.25 8 32.5 32.5 32.5 20 100 452.204 66.3689 1.825 9 45 45 32.5 32.5 550 433.874 63.1921 10.0375 10 32.5 20 32.5 45 550 452.963 66.4896 10.0375 11 20 32.5 32.5 45 550 452.963 66.4828 10.0375 12 45 32.5 45 32.5 550 433.874 63.1908 10.0375 13 45 32.5 32.5 20 550 452.963 66.4827 10.0375 14 20 32.5 32.5 32.5 100 452.204 66.3688 1.825 15 32.5 45 32.5 20 550 452.963 66.4895 10.0375 16 45 32.5 32.5 32.5 100 433.06 63.0101 1.825 17 20 32.5 45 32.5 550 452.963 66.4895 10.0375 18 32.5 32.5 32.5 20 1000 453.773 66.5757 18.25 19 45 20 32.5 32.5 550 452.963 66.4864 10.0375 20 45 32.5 20 32.5 550 452.963 66.4896 10.0375 21 32.5 45 45 32.5 550 433.874 63.1941 10.0375 22 32.5 32.5 45 32.5 100 433.06 63.0101 1.825 23 32.5 45 32.5 32.5 100 433.06 63.01 1.825 24 32.5 32.5 32.5 45 100 433.06 63.01 1.825 25 32.5 32.5 32.5 45 1000 434.689 63.3832 18.25 26 32.5 32.5 20 32.5 1000 453.773 66.5757 18.25 27 32.5 45 32.5 32.5 1000 434.689 63.3832 18.25 28 32.5 20 32.5 32.5 100 452.204 66.3689 1.825 29 32.5 20 20 32.5 550 453.016 66.6909 10.0375 30 32.5 32.5 20 45 550 452.963 66.4864 10.0375 31 32.5 32.5 45 32.5 1000 434.689 63.3832 18.25 32 32.5 32.5 32.5 32.5 550 433.875 63.188 10.0375 33 32.5 32.5 45 45 550 433.874 63.1921 10.0375 34 32.5 20 32.5 20 550 453.017 66.6804 10.0375 35 32.5 20 45 32.5 550 452.963 66.4828 10.0375 36 20 32.5 20 32.5 550 453.017 66.6804 10.0375 37 32.5 20 32.5 32.5 1000 453.773 66.5757 18.25 38 32.5 32.5 32.5 32.5 550 433.875 63.188 10.0375 39 32.5 45 32.5 45 550 433.874 63.1908 10.0375 40 32.5 32.5 32.5 32.5 550 433.875 63.188 10.0375 41 45 32.5 32.5 32.5 1000 434.689 63.3832 18.25 42 45 32.5 32.5 45 550 433.874 63.1941 10.0375 43 20 20 32.5 32.5 550 453.017 66.6856 10.0375 44 32.5 32.5 32.5 32.5 550 433.875 63.188 10.0375 45 32.5 32.5 45 20 550 452.963 66.4863 10.0375 46 32.5 45 20 32.5 550 452.963 66.4828 10.0375 Improved Oil and Gas Recovery 6 Result and Discussion Analysis of Variance. Tables 3 to 5 present the ANOVA results for cumulative gas production, cumulative oil production, and cumulative gas injection, respectively. It is deducible that the models and their corresponding terms are statistically significant. Consequently, the model is deemed accurate and suitable for predictive purposes. Eqs. 1 to 3 illustrate polynomial regression models delineating the correlations between cumulative gas production, cumulative oil production, and cumulative gas injection with the independent variables, (such as gas injection rate and bottom-hole pressure, respectively. Table 3—ANOVA for cumulative gas production. Source Sum of Squares df Mean Square F-value p-value Model 4353.92 16 272.12 13.03 < 0.0001 significant A-X1 571.88 1 571.88 27.39 < 0.0001 B-X2 571.88 1 571.88 27.39 < 0.0001 C-X3 571.88 1 571.88 27.39 < 0.0001 D-X4 571.88 1 571.88 27.39 < 0.0001 E-X5 10.23 1 10.23 0.4898 0.4896 AB 90.58 1 90.58 4.34 0.0462 AC 90.58 1 90.58 4.34 0.0462 AD 90.59 1 90.59 4.34 0.0462 BC 90.59 1 90.59 4.34 0.0462 BD 90.58 1 90.58 4.34 0.0462 CD 90.58 1 90.58 4.34 0.0462 A² 680.27 1 680.27 32.58 < 0.0001 B² 680.27 1 680.27 32.58 < 0.0001 C² 680.27 1 680.27 32.58 < 0.0001 D² 680.27 1 680.27 32.58 < 0.0001 E² 143.60 1 143.60 6.88 0.0138 Residual 605.54 29 20.88 Lack of Fit 273.20 24 11.38 0.1713 0.9988 not significant Pure Error 332.34 5 66.47 Cor Total 4959.46 45 Improved Oil and Gas Recovery 7 Table 1—ANOVA for cumulative oil production. Source Sum of Squares df Mean Square F-value p-value Model 133.73 16 8.36 14.60 < 0.0001 significant A-X1 18.13 1 18.13 31.66 < 0.0001 B-X2 18.13 1 18.13 31.66 < 0.0001 C-X3 18.13 1 18.13 31.66 < 0.0001 D-X4 18.13 1 18.13 31.66 < 0.0001 E-X5 0.3364 1 0.3364 0.5875 0.4496 AB 2.39 1 2.39 4.18 0.0500 AC 2.41 1 2.41 4.22 0.0491 AD 2.37 1 2.37 4.14 0.0510 BC 2.37 1 2.37 4.14 0.0510 BD 2.41 1 2.41 4.22 0.0491 CD 2.39 1 2.39 4.18 0.0500 A² 20.82 1 20.82 36.36 < 0.0001 B² 20.82 1 20.82 36.36 < 0.0001 C² 20.82 1 20.82 36.36 < 0.0001 D² 20.82 1 20.82 36.36 < 0.0001 E² 3.88 1 3.88 6.77 0.0144 Residual 16.61 29 0.5726 Lack of Fit 7.04 24 0.2934 0.1534 0.9994 not significant Pure Error 9.56 5 1.91 Cor Total 150.34 45 Table 5—ANOVA for cumulative gas injection. Source Sum of Squares df Mean Square F-value p-value Model 1079.90 6 179.98 107.66 < 0.0001 significant A-X1 0.0000 1 0.0000 0.0000 1.0000 B-X2 0.0000 1 0.0000 0.0000 1.0000 C-X3 0.0000 1 0.0000 0.0000 1.0000 D-X4 0.0000 1 0.0000 0.0000 1.0000 E-X5 1079.12 1 1079.12 645.52 < 0.0001 A² 0.7820 1 0.7820 0.4678 0.4981 Residual 65.20 39 1.67 Lack of Fit 8.99 34 0.2645 0.0235 1.0000 not significant Pure Error 56.20 5 11.24 Cor Total 1145.10 45 Improved Oil and Gas Recovery 8 CGP = 543.521 + −1.18159 ∗ X1 + −1.1816 ∗ X2 + −1.1816 ∗ X3 + −1.1816 ∗ X4 + −0.0202577 ∗ X5 + −0.030456 ∗ X1 ∗ X2 + −0.0304548 ∗ X1 ∗ X3 + −0.0304574 ∗ X1 ∗ X3 + −0.0304573 ∗ X2 ∗ X3 + −0.0304548 ∗ X2 ∗ X4 + −0.0304559 ∗ X3 ∗ X4 + 0.0565043 ∗ X12 + 0.0565043 ∗ X22 + 0.0565043 ∗ X32 + 0.0565043 ∗ X42 + 2.00312e − 05 ∗ X52 ,.............................(1) COP = 84.9002 + −0.244973 ∗ X1 + −0.244975 ∗ X2 + −0.244974 ∗ X3 + −0.244976 ∗ X4 + −0.0032982 ∗ X5 + −0.00495211 ∗ X1 ∗ X2 + −0.00497278 ∗ X1 ∗ X3 + −0.00492893 ∗ X1 ∗ X4 + −0.00492904 ∗ X2 ∗ X3 + −0.00497267 ∗ X2 ∗ X4 + −0.00495202 ∗ X3 ∗ X4 + 0.00988563 ∗ X12 + 0.00988566 ∗ X22 + 0.00988566 ∗ X32 + 0.00988559 ∗ X42 + 3.29129e − 06 ∗ X52 ,...................................................................................................................................(2) 𝐶𝐺𝐼 = 1.5768 + −0.11388 ∗ 𝑋1 + 9.48346𝑒 − 17 ∗ 𝑋2 + −5.19117𝑒 − 18 ∗ 𝑋3 + −3.24107𝑒 − 19 ∗ 𝑋4 + 0.01825 ∗ 𝑋5 + 0.001752 ∗ 𝑋12 ,......................................................................(3) Polynomial Regression Model. The validation process of the models was rigorously conducted through the utilization of cross plots and a comprehensive statistical error analysis. Figures 3 through 5 present detailed cross plots that correspond to cumulative gas production, cumulative oil production, and cumulative gas injection, respectively. In Figure 3, it is evident that the predicted cumulative gas production achieved an impressive regression value of 87.81% when compared to the actual cumulative gas production data. This indicates a strong correlation between the predicted and actual values, underscoring the accuracy of the model in this aspect. Similarly, Figure 4 illustrates the cross plot for cumulative oil production, where the predicted values exhibited a regression of 88.91% against the actual cumulative oil production figures. This high degree of correlation further substantiates the reliability of the model in forecasting oil production trends. Moreover, Figure 5 depicts the cross plot for cumulative gas injection, revealing that the predicted values had a regression of 94.31% with respect to the actual cumulative gas injection data. This exceptionally high regression value suggests that the model is highly effective in capturing the dynamics of gas injection processes. Figure 3—Cross plots for cumulative gas production. y = 0.8778x + 54.287 R² = 0.8781 415 420 425 430 435 440 445 450 455 460 410 415 420 425 430 435 440 445 450 455 460 P re d ic te d C G P Actual CGP Improved Oil and Gas Recovery 9 Figure 4—Cross plots for cumulative oil production. Figure 5—Cross plots for cumulative gas injection. Collectively, these cross plots and their corresponding statistical analyses provide compelling evidence that the developed polynomial regression models are not only valid but also robust enough to be employed for predictive and optimization studies. The high regression values across all three categories-cumulative gas production, cumulative oil production, and cumulative gas injection-demonstrate the models' capability to accurately represent real-world scenarios. This validation is crucial for ensuring that any predictions or optimizations derived from these models will be based on a solid foundation of empirical data and analytical rigor. Therefore, it is reasonable to conclude that these models can be confidently utilized in further studies to enhance understanding and performance in the respective fields of gas production, oil production, and gas injection. Optimization of Polynomial Regression Models. Table 6 showed the multi-objective optimization results. The models were run using multi-objective genetic algorithm because two objective functions represented by two y = 0.8891x + 7.2052 R² = 0.8893 60 61 62 63 64 65 66 67 68 59 60 61 62 63 64 65 66 67 68 P re d ic te d C O P Actual COP y = 0.9429x + 0.5621 R² = 0.9431 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 p re d ic te d C G I Actual CGI Improved Oil and Gas Recovery 10 polynomial regression models were considered. The optimization was geared towards get the highest cumulative oil production and the least cumulative water production. As shown in Table 6, the X1, X2, X3, X4 and X5 values of serial number 16 recorded the highest cumulative oil production and the least cumulative water production showing the role of optimization in production capacity enhancement. Figure 6 shows the Pareto front. As observed from the Pareto front increase in cumulative oil production yielded a reduction in cumulative gas production, and vice versa Table 6—Multi-objective optimization results. S/N x1 x2 x3 x4 x5 y1 y2 1 44.99835 44.12597 44.99238 35.92743 617.1253 60.95589 407.6993 2 44.99914 44.70789 44.98837 42.14048 617.3741 60.09117 410.8003 3 44.99983 44.99617 44.99998 44.9608 618.6007 59.94088 413.5667 4 44.99923 44.06912 44.99925 36.44121 616.9327 60.8564 407.7916 5 44.99925 44.00692 44.99938 36.77325 616.8554 60.79503 407.8758 6 44.99986 44.53442 44.99991 38.65928 617.8937 60.48542 408.5261 7 44.99872 44.7249 44.9956 40.62113 618.5155 60.23432 409.6339 8 44.99922 44.74702 44.99191 43.2595 618.0383 60.01514 411.8113 9 44.99843 44.42491 44.99319 37.02922 616.9287 60.74936 407.9331 10 44.9991 44.2125 44.99207 37.99829 617.9439 60.58704 408.2783 11 44.99889 44.06207 44.99943 35.48444 617.0295 61.04561 407.6294 12 44.99877 44.28185 44.99814 40.18627 617.3646 60.28689 409.3936 13 44.99881 44.92972 44.99796 42.72717 617.6583 60.04498 411.2694 14 44.99984 44.98566 44.99992 44.14489 617.9883 59.96714 412.6583 15 44.99979 44.5243 44.98589 39.22763 617.5508 60.4049 408.8248 16 45 43.98145 45 34.39897 616.8373 61.28123 407.5679 17 44.99575 44.56349 44.99753 41.31478 617.4664 60.16476 410.1453 18 44.99801 44.93166 44.99919 41.7078 618.3014 60.1253 410.4034 Figure 6—Pareto front showing optimal solutions. Improved Oil and Gas Recovery 11 Conclusions This study concentrated on enhancing gas condensate recovery by identifying optimal operational parameters through the construction of surrogate models and optimization algorithms. Conventional methodologies for pinpointing optimal conditions, which typically involve trial and error or direct optimization via reservoir simulation, can be both computationally demanding and time-intensive. In contrast, the methodology employed in this study, which incorporates surrogate models and an optimization algorithm, has been shown to yield optimal outcomes with reduced computational expenditure. Utilizing an inverted five-spot model and a comprehensive parameter set generated by a Box-Behnken experimental design, a sequence of simulation experiments was executed in this study. The resultant responses, namely cumulative oil production and cumulative gas production, were ascertained. Response surface models were constructed employing response surface methodology, and these models exhibited a satisfactory correlation between cumulative oil production, cumulative gas production, and the independent variables. Validation of the models confirmed their reliability for optimization endeavors. A multi-objective genetic algorithm was implemented to ascertain a singular optimal solution that maximizes cumulative oil production and minimizes cumulative gas production. The identified optimal solution represents the optimal combination of bottom-hole pressure for production wells and gas injection rate for injection wells. By implementing this optimal solution, more efficient gas condensate recovery can be realized. The project conclusively demonstrated that the amalgamation of surrogate models and optimization algorithms facilitates the determination of optimal operational parameters in gas condensate reservoir simulation. This approach diminishes computational costs and temporal demands, offering a valuable asset for reservoir engineers and operators. Recommendation Based on the findings and conclusions of this project, the following recommendations can be made for further studies and practical applications. • Expand the Scope. The current study focused on an inverted five-spot model in a gas condensate reservoir. Future research can explore different reservoir geometries, injection strategies, and field development scenarios to validate the effectiveness of the proposed approach across a broader range of scenarios. • Consider Additional Objective Functions. While the current study focused on maximizing COP and minimizing CGP as objective functions, other factors such as the economic viability, environmental impact, or resource utilization efficiency can be included as additional objectives for a more comprehensive analysis. • Experimental Validation. Conducting laboratory experiments or pilot tests to validate the results obtained from the proxy models and optimization algorithms can further enhance the reliability and practical applicability of the approach. • Field Application. Collaborate with industry partners to implement and field test the optimized operating conditions derived from this study. Real-world data and feedback can provide valuable insights into the effectiveness and practicality of the proposed approach. • Integration of Advanced Technologies. Explore the integration of advanced technologies such as artificial intelligence, machine learning, or advanced data analytics techniques to improve the accuracy and efficiency of proxy models and optimization algorithms. Conflicting Interests The author(s) declare that they have no conflicting interests. Improved Oil and Gas Recovery 12 References Central Intelligence Agency. 2014. The World Factbook 2014.https://www.cia.gov/library/publications/the-world- factbook/ (accessed 1 January 2025). Elehinafe, F.B., Nwizu, C.I., Odunlami, O.B., et al. 2022. Natural Gas Flaring in Nigeria Its Effects and Potential Alternatives- A Review. Journal of Ecological Engineering 23(8): 141-151. Kerunwa, A., Izuwa, N.C., Dike, C.F., et al. 2024. Review on the Utilization of Local ASP in the Niger-Delta for Enhanced Oil Recovery. Petroleum and Coal 66(1): 256-273. Li, D. and Firoozabadi, A. 2000. Asphaltene Deposition in Porous Media: Effects of pH and Pressure. Journal of Petroleum Science and Engineering 27(1): 95-105. McCain, W.D. 1990. The Properties of Petroleum Fluids. Tulsa, USA: PennWell Books. Moses, M.O., and Donohoe, R.J. 1987. Enhanced Oil Recovery. Upper Saddle River, New Jersey: Prentice Hall. Peñuela, G. and Civan, F. 2000. Predicting Microbial Plugging in Porous Media Using a Population Balance Model. Journal of Petroleum Science and Engineering 26(1): 209-219. Pope, G.A., Sepehrnoori, K., and Delshad, M. 2000. A Compositional Streamline Simulator for Three-Phase Three- Component Flow in Petroleum Reservoirs. SPE Reservoir Evaluation & Engineering 3(4): 344-355. Chukwuebuka Francis Dike is a Research Technologist at the Department of Petroleum Engineering, Federal University of Technology Owerri. He has research interest in Drilling Fluids Technology, Reservoir Engineering, Enhanced Oil Recovery and Flow Assurance. Dike Holds a bachelor’s degree and master’s degree in petroleum engineering from Federal University of Technology Owerri. https://www.cia.gov/library/publications/the-world-factbook/ https://www.cia.gov/library/publications/the-world-factbook/