Stochastic Simulation of Leaching Range in In-situ Leaching Process Considering Uncertainty of Permeability Coefficient

Numerical simulation is an important method to determine the leaching range in the process of in-situ uranium mining, and the permeability coefficient is an important factor affecting the accurate calculation of leaching range by numerical simulation methods. However, the numerical model is difficult to completely describe the actual spatial distribution of permeability coefficients due to the complexity of the distribution of aquifer permeability. In order to analyze the influence of permeability coefficient uncertainty on leaching agent migration process and leaching range, according to the field condition test data of sandstone-type uranium deposits in Inner Mongolia, 200 random fields of permeability coefficient in accordance with the specific distribution were obtained using the Python geostatistical software package GSTools. All permeability coefficient random fields were applied to solute transport simulation, and Monte Carlo method was used to convert the parameter uncertainty of permeability coefficient into the uncertainty of model results. The results of simulation under homogeneous and heterogeneous permeability conditions were compared. The leaching agent in homogeneous conditions could not migrate outward and the leaching range is a regular circular shape. The direction of leaching agent migration is affected by the regional permeability between pumping and injection wells under heterogeneous conditions of permeability coefficients. When the permeability coefficient of the region between the pumping and injection wells is relatively higher, the leaching agent migrates only to the pumping wells and not outward. Conversely, a part of the leaching agent would spread along the preferential flow channels to the high-permeability region outside the pumping and injection wells region, resulting in a non-regular shape of contamination plume. Based on the analysis of the actual geological conditions, the complexity of its spatial distribution is ignored when the permeability coefficient is extended to the average value, and the influence of the random distribution of the permeability coefficient should be considered in the model. However, the leaching range obtained from different permeability coefficient random fields has obvious differences, indicating that the uncertainty of permeability coefficient leads to the uncertainty of simulation results, which can be quantified and analyzed using relevant statistical methods. The results of the quantitativ