Machine learning prediction of methane, nitrogen, and natural gas mixture viscosities under normal and harsh conditions.

The accurate estimation of gas viscosity remains a pivotal concern for petroleum engineers, exerting substantial influence on the modeling efficacy of natural gas operations. Due to their time-consuming and costly nature, experimental measurements of gas viscosity are challenging. Data-based machine learning (ML) techniques afford a resourceful and less exhausting substitution, aiding research and industry at gas modeling that is incredible to reach in the laboratory. Statistical approaches were used to analyze the experimental data before applying machine learning. Seven machine learning techniques specifically Linear Regression, random forest (RF), decision trees, gradient boosting, K-nearest neighbors, Nu support vector regression (NuSVR), and artificial neural network (ANN) were applied for the prediction of methane (CH4), nitrogen (N2), and natural gas mixture viscosities. More than 4304 datasets from real experimental data utilizing pressure, temperature, and gas density were employed for developing ML models. Furthermore, three novel correlations have developed for the viscosity of CH4, N2, and composite gas using ANN. Results revealed that models and anticipated correlations predicted methane, nitrogen, and natural gas mixture viscosities with high precision. Results designated that the ANN, RF, and gradient Boosting models have performed better with a coefficient of determination (R2) of 0.99 for testing data sets of methane, nitrogen, and natural gas mixture viscosities. However, linear regression and NuSVR have performed poorly with a coefficient of determination (R2) of 0.07 and - 0.01 respectively for testing data sets of nitrogen viscosity. Such machine learning models offer the industry and research a cost-effective and fast tool for accurately approximating the viscosities of methane, nitrogen, and gas mixture under normal and harsh conditions.

A New Interpretation of Gas Viscosity for Flow through Micro-Capillaries and Pores.

The Hagen-Poiseuille equation for gas flow had never been derived theoretically; it is rather a simple analogy of the same for liquid flow, and "gas viscosity" is a measure for overall resistance to flow. In this work, experimental flow data for different gases through capillaries and porous media, reported in literature by different groups, including those measured and treated by Knudsen are treated with Hagen-Poiseuille equation, but taking "gas viscosity" as an adjustable parameter. It is found that, at constant temperature, there exists an unambiguous relation between the viscosity (µ) of a given gas, and the product of average pressure (Pav ) and capillary diameter (D). In addition, for Pav * D < 0.01, a universal linear relation exists between µ/M0.5 (where M is molecular mass) for different gases and the parameter Pav * D. The new interpretation of gas viscosity avoids the differentiation of regimes into "Knudsen" and "viscous" flow as it is frequently done in literature. The concept can be applied to obtain a reliable data base for gas viscosities in different fields of applications, for example in microfluidic systems or the analysis of pore size distributions of filters and membranes by gas flow porometry.

Measurement and Evaluation of the Gas Density and Viscosity of Pure Gases and Mixtures Using a Micro-Cantilever Beam.

Measurement of gas density and viscosity was conducted using a micro-cantilever beam. In parallel, the validity of the proposed modeling approach was evaluated. This study also aimed to widen the database of the gases on which the model development of the micro-cantilever beams is based. The density and viscosity of gases are orders of magnitude lower than liquids. For this reason, the use of a very sensitive sensor is essential. In this study, a micro-cantilever beam from the field of atomic force microscopy was used. Although the current cantilever was designed to work with thermal activation, in the current investigation, it was activated with an electromagnetic force. The deflection of the cantilever beam was detected by an integrated piezo-resistive sensor. Six pure gases and sixteen mixtures of them in ambient conditions were investigated. The outcome of the investigation showed that the current cantilever beam had a sensitivity of 240 Hz/(kg/m³), while the accuracy of the determined gas density and viscosity in ambient conditions reached ±1.5% and ±2.0%, respectively.