Abstract:Accurately predicting the deviation factor (Z-factor) of natural gas is crucial for the estimation of natural gas reserves, evaluation of gas reservoir recovery, and assessment of natural gas transport in pipelines. Traditional machine learning algorithms, such as Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Artificial Neural Network (ANN) and Bidirectional Long Short-Term Memory Neural Networks (BiLSTM), often lack accuracy and robustness in various situations due to their inability to generalize across different gas components and temperature-pressure conditions. To address this limitation, we propose a novel and efficient machine learning framework for predicting natural gas Z-factor. Our approach first utilizes a signal decomposition algorithm like Variational Mode Decomposition (VMD), Empirical Fourier Decomposition (EFD) and Ensemble Empirical Mode Decomposition (EEMD) to decouple the Z-factor into multiple components. Subsequently, traditional machine learning algorithms is employed to predict each decomposed Z-factor component, where combination of SVM and VMD achieved the best performance. Decoupling the Z-factors firstly and then predicting the decoupled components can significantly improve prediction accuracy of all traditional machine learning algorithms. We thoroughly evaluate the impact of the decoupling method and the number of decomposed components on the model's performance. Compared to traditional machine learning models without decomposition, our framework achieves an average correlation coefficient exceeding 0.99 and an average mean absolute percentage error below 0.83% on 10 datasets with different natural gas components, high temperatures, and pressures. These results indicate that hybrid model effectively learns the patterns of Z-factor variations and can be applied to the prediction of natural gas Z-factors under various conditions. This study significantly advances methodologies for predicting natural gas properties, offering a unified and robust solution for precise estimations, thereby benefiting the natural gas industry in resource estimation and reservoir management.

Dynamic Measurement and Prediction of Sulfur Hexafluoride Gas Weight Based on Non-Ideal Gas State Equation and Improved Random Forest

Application of Random Forest Machine Learning Models to Forecast Combustion Profile Parameters of a Natural Gas Spark Ignition Engine

A New Prediction Model of SF6 Alternative Gases in Electrical Strength

Prediction of the Electrical Strength and Boiling Temperature of the Substitutes for Greenhouse Gas SF₆ Using Neural Network and Random Forest

Leveraging Machine Learning To Predict the Atmospheric Lifetime and the Global Warming Potential of SF 6 Replacement Gases

Development and Application of Multi-Component Dynamic Gas Distribution Instrument Based on Mass Flow Controller

Research on Load Prediction of Low-Calorific Fuel Fired Gas Turbine Based on Data and Knowledge Hybrid Model

Fault Diagnosis of SF6-Insulated Equipment by Micro Gas Sensor Array

First Principles Based Computational Scheme for Designing New SF6 Replacements

Quantitative detection of multicomponent SF 6 decomposition products based on Fourier transform infrared spectroscopy combined with SCARS-DNN

Fault Prediction Using a Grey-Markov Model from the Dissolved Gases Contents in Transformer Oils

Integrating Dynamic State Information to Enhance Gas Concentration Modeling Accuracy

Temperature and Humidity Compensation for MOS Gas Sensor Based on Random Forests.

Effects of SF 6 decomposition components and concentrations on the discharge faults and insulation defects in GIS equipment.

Prediction of NOx Emissions in Thermal Power Plants Using a Dynamic Soft Sensor Based on Random Forest and Just-in-Time Learning Methods

Quantitative Detection of Sulphur Hexafluoride Decomposition Product with a Gas Sensor Array

Effects of SF6 decomposition components and concentrations on the discharge faults and insulation defects in GIS equipment

A Modified Model of Direct Estimation Method for Fine Fuel Moisture Content Prediction by Considering Crown Density

A Novel Efficient Model for Gas Compressibility Factor Based on GMDH Network

Decoupling and predicting natural gas deviation factor using machine learning methods

Multicomponent SF6 Decomposition Product Sensing with a Gas-Sensing Microchip