Permeability evolution of uranium reservoirs under multi-field coupling (gas-temperature-force-seepage) and artificial intelligence multi-parameter target optimization study DOI

Y.Z. Chen,

Qizhi Wang, Yuan Wei

и другие.

Physics of Fluids, Год журнала: 2025, Номер 37(3)

Опубликована: Март 1, 2025

In insitu leaching uranium mining, insufficient permeability of the rock formation is a primary bottleneck limiting resource extraction. This study focuses on sandstone-type mine in Xinjiang, where coupled gas-temperature-force-seepage model was developed using finite discrete element method. The rock's damage mechanism and evolution during blasting for enhancement were systematically analyzed. First, three-factor, five-level orthogonal numerical experiment designed to quantify effects uncoupling coefficient, short delay time, pressure rise time fracture network expansion. optimal parameter combinations reservoir modification identified, regulatory geostress sequence connectivity elucidated. Second, effect injection reservoir's range flow distribution analyzed, providing theoretical support optimization parameters. Finally, an framework combining machine learning genetic algorithms introduced further enhance rate. accurately predicts rate optimizes eXtreme Gradient Boosting model. results show that maximum 3.2491 × 10−4 m3 s−1 can be achieved under various combinations, demonstrating robustness broad applicability framework. provides additional insights into reservoirs.

Язык: Английский

Permeability evolution of uranium reservoirs under multi-field coupling (gas-temperature-force-seepage) and artificial intelligence multi-parameter target optimization study DOI

Y.Z. Chen,

Qizhi Wang, Yuan Wei

и другие.

Physics of Fluids, Год журнала: 2025, Номер 37(3)

Опубликована: Март 1, 2025

In insitu leaching uranium mining, insufficient permeability of the rock formation is a primary bottleneck limiting resource extraction. This study focuses on sandstone-type mine in Xinjiang, where coupled gas-temperature-force-seepage model was developed using finite discrete element method. The rock's damage mechanism and evolution during blasting for enhancement were systematically analyzed. First, three-factor, five-level orthogonal numerical experiment designed to quantify effects uncoupling coefficient, short delay time, pressure rise time fracture network expansion. optimal parameter combinations reservoir modification identified, regulatory geostress sequence connectivity elucidated. Second, effect injection reservoir's range flow distribution analyzed, providing theoretical support optimization parameters. Finally, an framework combining machine learning genetic algorithms introduced further enhance rate. accurately predicts rate optimizes eXtreme Gradient Boosting model. results show that maximum 3.2491 × 10−4 m3 s−1 can be achieved under various combinations, demonstrating robustness broad applicability framework. provides additional insights into reservoirs.

Язык: Английский

Процитировано

1