Injection of CO2 to enhance oil recovery is widely used due to its multiple advantages such as mobilizing the oil and sequestration of carbon dioxide. Injection of CO2 can enhance oil recovery by reducing oil viscosity and improving overall fluid mobility. However, several problems are associated with CO2 injection such as viscous fingering, gravity override, and CO2 channeling that results in early gas breakthrough, low sweep efficiency, and low ultimate oil recovery. In this study, dual benefits of CO2 injection are presented: enhancing oil recovery and sequestering carbon dioxide. In this work, different scenarios of field scale simulation were conducted to evaluate oil recovery during CO2 injection, and the CMG (Computer Modeling Group) software package was used. Three main scenarios were examined which are CO2 injection into the reservoir, CO2 injection into the aquifer, and CO2 injection into the aquifer followed by waterflooding. Also, three well configurations were utilized—all injectors and producers are drilled vertically, all wells are drilled horizontally, and vertical injectors and horizontal producers are used. Therefore, the oil recovery profiles were examined for nine scenarios over a 20-year period. In all simulated models, CO2 injection was started at the residual oil saturation (Sor) conditions, to represent the cases of depleted oil reservoirs. The results indicated that the highest oil recovery of 73% of the original oil-in-place (OOIP) can be achieved by injecting CO2 into the reservoir, utilizing vertical injectors and producers. While injecting CO2 into aquifers can significantly enhance oil recovery by around 68–70% of the OOIP, using horizontal wells can provide more oil recovery (67.7%) than that using vertical wells (54.8%), in the same conditions. Moreover, around 7,928 tons of carbon dioxide can be sequestered in underground formations, on average. Finally, CO2 injection outperformed the conventional waterflooding, where 68 and 12% of the OOIP were obtained, respectively. Overall, injection of CO2 into the depleted reservoir can provide dual benefits of CO2 sequestration and improved oil recovery. CO2 can be injected into the water zone resulting in a slow release of CO2 which will reduce the fluid viscosity, enhance oil recovery, and reduce the greenhouse effect.

CO2 is used to swell the oil and increase its mobility by reducing the viscosity and reducing the IFT between the rock and oil. The problem of CO2 during the EOR process is the gravity override and the displacement efficiency. Foamed CO2 was introduced to overcome the displacement efficiency problem. The foam can be generated using different types of surfactants. The surfactants will not from a stable foam at HPHT conditions and the reservoir water salinity will impact the foam quality in addition to the surfactant adsorption to the carbonate rocks will reduce the surfactant concentration and will affect the generated foam. In this study we will investigate the use of slow release CO2 concept. The CO2 can be slowly released in the carbonate reservoir by injecting low reaction rate fluids (in this case citric acid) that will react with the rock after placement and this will generate CO2. Coreflooding test will be performed using carbonate core at 100°C saturated with crude oil. CMG will be used to simulate the process of CO2 generation by injecting the CO2 in an aquifer underlying the carbonate formation and after that allow the CO2 to release to the oil zone. The slow released CO2 from the aquifer will swell the oil and then the oil can be displaced by sea water. Coreflooding results showed that the slow release CO2 technique gave up to 42 % extra oil recovery from the oil in place after sea water flooding into the carbonate core at 100°C and 1100 psi back pressure. The extra oil recovery was due to the slow resealed CO2 and oil swelling and also due to the rock dissolution and IFT reduction. The simulation results by CMG showed that the slow release CO2 technique gave 70% oil recovery from the residual oil more than the sea water injection which confirmed the efficiency of the slow release technique in large scale.

Availability of large amounts of data helps in developing data-driven models using state of the art Artificial intelligence (AI) methodologies. Relative permeability is an important parameter used by reservoir engineers and are usually accurately obtained from laboratory experiments, which are relatively expensive. Therefore, AI can play an important role in developing models to predict relative permeability accurately without extensive lab procedures. Accordingly, this work presents application of two AI algorithms namely, Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). Furthermore, two novel mathematical correlations are extracted from the ANN model to predict relative permeability of oil/water in oil- and water-wet environments. The input data, obtained from literature, for the development of AI models include porosity, rock absolute permeability, initial water saturation, residual oil saturation, wettability index and water saturation. A customized workflow is applied to ensure proper data is fed into the AI models. In addition, a rigorous sensitivity analysis is performed within the framework of this workflow. This analysis involves running multiple realizations with varying number of neurons, resulting in various weights and bias for the ANN model. Moreover, ANFIS model is tuned using various cluster sizes to result in the most optimum value. Finally, the optimized ANN and ANFIS models are compared using the Root Mean Squared Error (RMSE) and correlation coefficient (R2) analysis when applied to a blind dataset comprising of more than 300 data points. The analysis illustrates that the ANN model is relatively better in predicting relative permeability values to both, oil, and water. On the other hand, analysis of the ANFIS model shows that it yields high error values when tested on unseen dataset. Also, unlike the ANN mode, it does not provide an actual mathematical correlation. This work presents alternate data-driven artificial intelligence models which will lead to quicker and cheaper relative permeability estimates.