Thermodynamic models are usually employed to predict formation condition of hydrates. However, these thermodynamic models usually require a large amount of calculations to approach phase equilibrium. Additionally, parameters included in the thermodynamic model need to be calibrated based on the experimental data, which leads to high uncertainties in the predicted results. With the rapid development of artificial intelligence (AI), machine learning as one of sub-discipline has been developed and been widely applied in various research area. In this work, machine learning was innovatively employed to predict the formation condition of natural gas hydrates to overcome the high computation cost and low accuracy. Three data-driven models, Random Forest (RF), Naive Bayes (NB), Support Vector Regression (SVR) were tentatively used to determine the formation condition of hydrate formed by pure and mixed gases. Experimental data reported in previous work were taken to train and test the machine learning models. As a representative thermodynamic model the Chen–Guo (C-G) model was used to analyze the computational efficiency and accuracy of machine learning models. The comparison of results predicted by C-G model and machine learning models with the experimental data indicated that the RF model performed better than the NB and SVR models on both computation speed and accuracy. According to the experimental data, the average AADP calculated by the C-G model is 7.62 times that calculated by the RF model. Meanwhile, the average time costed by the C-G model is 75.65 times that by the RF model. Compared with the other two machine learning models, the RF model is expected to be used in predicting the formation condition of natural gas hydrate under field conditions.

No abstract

Conventional techniques for hydrate production may cause the deconstruction of hydrate, changing the geomechanical stresses of the reservoir, which could trigger the subsidence of the seafloor. A new method for replacing CH4 from the hydrate lattice by CO2, without damaging the mechanical structure of sediment, has been proposed. This approach can achieve both the objectives of long-term CO2 sequestration and the safe production of CH4 from hydrates. By coupling the Chen-Guo model into Tough+Hydrate V1.5, an updated simulator CO2-EGHRSim V.10 (CO2 Enhanced Gas Hydrate Recovery simulator) was developed in this work to describe the replacing processes of CH4 from the hydrate lattice by CO2 and to evaluate the storage potential of CO2 and the recovery efficiency of CH4 from the hydrate-bearing reservoirs. The developed simulator was verified using measured data obtained from laboratory experiments. The verification suggested that CO2-EGHRSim performed well in predicting the replacing processes of CH4 with CO2. The simulator was applied to calculate the CO2 storage potential combined with the CH4 recovery from hydrates at the site of Iġnik Sikumi on the North Slope of Alaska. The simulated results indicated that the CO2–CH4 exchange mostly occurred inside the gas plume, and the CO2 hydrate was only present around the production well. The simulated CO2 storage ratio was 0.58, and the CH4 recovery efficiency was 25.95%.