Modeling High-Pressure Methane Adsorption on Shales with a Simplified Local Density Model

Deep Learning Models for Predicting Gas Adsorption Capacity of Nanomaterials

Metal-organic frameworks (MOFs), a class of porous nanomaterials, have been widely used in gas adsorption-based applications due to their high porosities and chemical tunability. To facilitate the discovery of high-performance MOFs for different applications, a variety of machine learning models have been developed to predict the gas adsorption capacities of MOFs. Most of the predictive models are developed using traditional machine learning algorithms. However, the continuously increasing sizes of MOF datasets and the complicated relationships between MOFs and their gas adsorption capacities make deep learning a suitable candidate to handle such big data with increased computational power and accuracy. In this study, we developed models for predicting gas adsorption capacities of MOFs using two deep learning algorithms, multilayer perceptron (MLP) and long short-term memory (LSTM) networks, with a hypothetical set of about 130,000 structures of MOFs with methane and carbon dioxide adsorption data at different pressures. The models were evaluated using 10 iterations of 10-fold cross validations and 100 holdout validations. The MLP and LSTM models performed similarly with high prediction accuracy. The models for predicting gas adsorption at a higher pressure outperformed the models for predicting gas adsorption at a lower pressure. The deep learning models are more accurate than the random forest models reported in the literature, especially for predicting gas adsorption capacities at low pressures. Our results demonstrated that deep learning algorithms have a great potential to generate models that can accurately predict the gas adsorption capacities of MOFs.

Absolute adsorption and adsorbed volume modeling for supercritical methane adsorption on shale

Adsorbed methane significantly affects shale gas reservoir estimates and shale gas transport in shale formations. Hence, a practical model for accurately representing methane adsorption behavior at high-pressure and high-temperature in shale is imperative. In this study, a reliable mathematical framework that estimates the absolute adsorption directly from low-pressure excess adsorption data is applied to describe the excess methane adsorption data in literature. This method provides detailed information on the volume and density of adsorbed methane. The obtained results indicate that the extensively used supercritical Dubinin-Radushkevich model with constant adsorbed phase density underestimates absolute adsorption at high pressure. The adsorbed methane volume increases both the pressure and expands with the temperature. The adsorbed methane density reduces above 10 MPa, and approaches a steady value at high pressure. This study provides a novel method for estimating adsorbed shale gas, which is expected improve the prediction of shale gas in place and gas production.

Impacting Factors, Dynamic Process, and Correction of Adsorption Reduction in Shale Reservoir: A Case Study on Shale Samples from the Western Guizhou

Adsorption reduction occurring during isothermal experiment leads to the failure of telling the true adsorption capacity of shale reservoir. A correct understanding of this will be helpful in improving the accuracy of resource estimation and economic evaluation of shale gas reserves. Six shale samples were collected from the Permian Longtan Formation in the western Guizhou Province, China. Volumetric methane isotherm adsorption experiment and data processing were conducted in this research. The study investigates the effect of free space volume reduction (FSVR), excess adsorption amount conversion (EAAC), and blank test correction (BTC) on adsorption reduction, the understanding of the dynamic process of adsorption reduction, and the evaluation of the way of weakening and correcting this phenomenon. The conclusions are as follows. (1) Adsorption reduction does exist in the shale sample. The adsorption process of methane in the shale sample can be divided into the strong adsorption stage, approximate saturation stage, and adsorption reduction stage. (2) Shale adsorbing methane has a positive effect on the experimental adsorption amount. Comparatively, free space volume, excess adsorption amount, and blank test have negative effects. Adsorption reduction is the result of combined influence of positive and negative effects above. (3) At the first two stages of methane adsorption, the positive effect is greater than the negative effect, resulting in the hidden of adsorption reduction, and the experimental adsorption amount increases with the growth of experimental pressure. While at the adsorption reduction stage, the former effect is smaller than the latter, and their difference increases as the experimental pressure increases. It leads to the occurrence of adsorption reduction, and the phenomenon becomes increasingly obvious. (4) FSVR has the strongest impact on the weakening of adsorption reduction, followed by EAAC and BTC. The adsorption reduction in shale reservoir can be corrected effectively by BTC and EAAC.