Scholarlink

<p>Waterflooding is crucial for China’s oil and gas industry, but it induces dynamic heterogeneity in reservoir properties, which commercial simulators fail to model accurately. To address this, we propose a multi-scale simulation method incorporating time-varying absolute permeability (k) and relative permeability (k<sub>r</sub>) driven by surface flux. An improved multi-scale finite volume (IMsFV) method solves pressure equations on multi-scale grids, enhancing computational efficiency. SPE10 benchmark validation shows 95.07% reduction in total simulation time and 98.19% in linear solver time versus the fully implicit method, with errors &lt; 5%. Unlike commercial simulators neglecting dynamic heterogeneity, this approach captures opposing mechanisms: dynamic k exacerbates water channeling, reducing recovery, while dynamic k<sub>r</sub> enhances fluid mobility and reduces residual oil saturation. Crucially, dynamic k<sub>r</sub>’s positive effect dominates during high water-cut stages, ultimately improving recovery. Sensitivity analysis confirms: (1) Dynamic heterogeneity primarily benefits mid-high water-cut stages, reducing water cut and increasing oil production; (2) At 99% water cut, it enhances recovery by 28.88–32.87% across permeabilities (400–2000 md), with greater gains in higher-permeability reservoirs; (3) Under increasing injection rates (50 → 200&#xa0;m<sup>3</sup>/day), recovery gains amplify from 18.07 to 45.49%. This approach provides an efficient, accurate tool for predicting remaining oil in high-water-cut reservoirs.</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

An efficient multiscale simulation framework integrating dynamic heterogeneity for accurate waterflooding prediction

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月2日

查看全文
  • Li Wu,
  • Junqiang Wang,
  • Deli Jia,
  • Jiqun Zhang,
  • Ruichao Zhang,
  • Yiqun Yan,
  • Yanfang Yin,
  • Shuoliang Wang

摘要

Waterflooding is crucial for China’s oil and gas industry, but it induces dynamic heterogeneity in reservoir properties, which commercial simulators fail to model accurately. To address this, we propose a multi-scale simulation method incorporating time-varying absolute permeability (k) and relative permeability (kr) driven by surface flux. An improved multi-scale finite volume (IMsFV) method solves pressure equations on multi-scale grids, enhancing computational efficiency. SPE10 benchmark validation shows 95.07% reduction in total simulation time and 98.19% in linear solver time versus the fully implicit method, with errors < 5%. Unlike commercial simulators neglecting dynamic heterogeneity, this approach captures opposing mechanisms: dynamic k exacerbates water channeling, reducing recovery, while dynamic kr enhances fluid mobility and reduces residual oil saturation. Crucially, dynamic kr’s positive effect dominates during high water-cut stages, ultimately improving recovery. Sensitivity analysis confirms: (1) Dynamic heterogeneity primarily benefits mid-high water-cut stages, reducing water cut and increasing oil production; (2) At 99% water cut, it enhances recovery by 28.88–32.87% across permeabilities (400–2000 md), with greater gains in higher-permeability reservoirs; (3) Under increasing injection rates (50 → 200 m3/day), recovery gains amplify from 18.07 to 45.49%. This approach provides an efficient, accurate tool for predicting remaining oil in high-water-cut reservoirs.