ABSTRACT The location and water conductivity characteristics of groundwater runoff channels in karst areas significantly influence the spatial distribution of groundwater. Currently, few studies have utilized mathematical algorithms to analyse the locations of these channels and their conductivity characteristics, particularly with groundwater level distribution as the objective function. This paper fully leverages the principle of water flowing to lower elevations and employs an optimized gradient descent method to identify the flowpaths. Additionally, a hybrid population optimization algorithm is used to determine the water conductivity parameters of the identified channels. The accuracy and applicability of the proposed path identification and parameter identification methods are validated through five preset cases. Finally, the proposed methods are applied in the Pingnan mining area, yielding results that align with on-site geophysical explorations and pumping tests. In conclusion, the proposed methods can provide valuable guidance for groundwater resource management and disaster mitigation in karst regions.
Read full abstract