1887

Abstract

Summary

Underground bio-methanation (UBM) technology offers a promising approach for carbon utilization and sequestration, as well as large-scale underground energy storage. This study is the first to apply machine learning methods to forecast CO conversion ratios in UBM, using extensive numerical simulation data. Nine key input parameters were sampled via Latin Hypercube Sampling, generating 3,000 simulation cases used to train and evaluate three classical models: Random Forest (RF), Least Squares Boosting (LSBoost), and Feedforward Neural Network (FNN). The results show that the FNN model yields the highest predictive accuracy, as indicated by the highest coefficient of determination (R2) on the test set, and the lowest mean absolute error (MAE) and root mean square error (RMSE) values. Further analysis of varying sample sizes (1500 to 3000) using FNN demonstrates that increasing the training data significantly improves model performance, with a notable enhancement at 3000 samples (test R2 = 0.990, MAE = 0.017, RMSE = 0.029). These findings underscore the effectiveness of FNN in rapidly and accurately predicting CO conversion ratios, thereby supporting optimized site selection and operational management in UBM processes.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202521245
2025-10-27
2026-01-15

  • From This Site

    /content/papers/10.3997/2214-4609.202521245
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Maior, C. B. S., Silva, S. P., Lins, I. D., Moura, M. J., and Droguett, E. L. [2025]. Machine learning for hydrogen storage applications: An exploration of the current literature. International Journal of Hydrogen Energy, 150219.
     [Google Scholar]
  2. Strobel, G., Hagemann, B., Huppertz, T. M., and Ganzer, L. [2020]. Underground bio-methanation: Concept and potential. Renewable and Sustainable Energy Reviews, 123, 109747.
     [Google Scholar]
  3. Wu, L., Hou, Z. M., Luo, Z. F., Fang, Y. L., Huang, L. C., Wu, X. N., Chen, Q. J., and Wang, Q. C. [2024]. Impacts of microbial interactions on underground hydrogen storage in porous media: A comprehensive review of experimental, numerical, and field studies. Petroleum Science, 21(6), 4067–4099.
     [Google Scholar]
  4. Wu, L., Hou, Z., and Luo, Z. [2025]. Impacts of microbial competition on underground bio-methanation of hydrogen and carbon dioxide: Insights from biogeochemical simulations. Renewable Energy, 123458.
     [Google Scholar]
  5. Xiong, Y., Hou, Z., Xie, H., Zhao, J., Tan, X., and Luo, J. [2023]. Microbial-mediated CO2 methanation and renewable natural gas storage in depleted petroleum reservoirs: A review of biogeochemical mechanism and perspective. Gondwana Research, 122, 184–198.
     [Google Scholar]
  6. Zhang, T., Hou, Z., Li, X., Chen, Q., Wang, Q., Lüddeke, C., Wu, L., Wu, X. N., and Sun, W. [2025]. A novel multivariable prognostic approach for PEMFC degradation and remaining useful life prediction using random forest and temporal convolutional network. Applied Energy, 385, 125569.
     [Google Scholar]
/content/papers/10.3997/2214-4609.202521245
Loading
Machine Learning-Based Prediction of CO2 Conversion Ratio in Underground Bio-Methanation
Conference Proceedings 2025, 1 (2025); https://doi.org/10.3997/2214-4609.202521245
/content/papers/10.3997/2214-4609.202521245
/content/papers/10.3997/2214-4609.202521245
Loading

Data & Media loading...

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error