Scholarlink

<p>This study presents an innovative approach for estimating permeability (K), a key reservoir property that influences fluid flow in natural gas hydrate (NGH) systems, which is essential for optimizing gas production from hydrocarbon reservoirs. In the NGH system, permeability is often significantly reduced due to the accumulation of hydrates within pore spaces, making the accurate estimation of permeability critical for evaluating reservoir quality and production. In this study, empirical correlations, regression analysis (RA), and artificial neural networks (ANNs) are integrated to enhance prediction accuracy. Comprehensive well-log datasets, including nuclear magnetic resonance (NMR), gamma ray (GR), P-wave sonic velocity, bulk density, and resistivity, were utilized to identify gas hydrate-bearing intervals, with a particular emphasis on NMR data for K estimation. The study evaluates the predictive efficacy of these models through absolute average relative error (AARE), normalized mean square error (NMSE), root mean square error (RMSE), and correlation coefficient (<i>R</i><sup>2</sup>). The ANN model demonstrates superior performance, accurately predicting <i>K</i> values ranging from 0.01 to 100 mD in the gas hydrate zone (GHZ) at depths of 300–325&#xa0;m below the seafloor (mbsf). For this study, the ANN model was trained solely on a single well dataset and still produced consistent permeability estimates, indicating its reliability for NGH assessment in data-scarce areas. This work provides novel insights by integrating advanced computational techniques for permeability prediction, strengthening the foundation for developing efficient production strategies in NGH resource exploitation. The proposed methodology offers a precise, data-driven solution for predicting permeability. It holds the potential for broader applications in similar geological settings, advancing the understanding and exploitation of gas hydrates.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel statistical and soft computing technique for permeability prediction in the offshore Krishna–Godavari basin, NGHP-02, India

  • 期刊论文

  • 发表日期：

    2026年1月6日

查看全文
  • Pradeep Kumar Shukla,
  • Tabish Rahman,
  • Vikram Vishal

摘要

This study presents an innovative approach for estimating permeability (K), a key reservoir property that influences fluid flow in natural gas hydrate (NGH) systems, which is essential for optimizing gas production from hydrocarbon reservoirs. In the NGH system, permeability is often significantly reduced due to the accumulation of hydrates within pore spaces, making the accurate estimation of permeability critical for evaluating reservoir quality and production. In this study, empirical correlations, regression analysis (RA), and artificial neural networks (ANNs) are integrated to enhance prediction accuracy. Comprehensive well-log datasets, including nuclear magnetic resonance (NMR), gamma ray (GR), P-wave sonic velocity, bulk density, and resistivity, were utilized to identify gas hydrate-bearing intervals, with a particular emphasis on NMR data for K estimation. The study evaluates the predictive efficacy of these models through absolute average relative error (AARE), normalized mean square error (NMSE), root mean square error (RMSE), and correlation coefficient (R2). The ANN model demonstrates superior performance, accurately predicting K values ranging from 0.01 to 100 mD in the gas hydrate zone (GHZ) at depths of 300–325 m below the seafloor (mbsf). For this study, the ANN model was trained solely on a single well dataset and still produced consistent permeability estimates, indicating its reliability for NGH assessment in data-scarce areas. This work provides novel insights by integrating advanced computational techniques for permeability prediction, strengthening the foundation for developing efficient production strategies in NGH resource exploitation. The proposed methodology offers a precise, data-driven solution for predicting permeability. It holds the potential for broader applications in similar geological settings, advancing the understanding and exploitation of gas hydrates.