Aquifer Water Abundance Evaluation Based on a Variable Weight Model.

Existing aquifer water abundance evaluation methods employ fixed and constant index weights and overlook variations within major influencing factors as well as interactions among multiple factors. We propose a water abundance index method based on a variable weight model that considers various water abundance influencing factors and their respective weights, and quantitatively determines the weights of the same factor across different state values. This paper establishes a comprehensive system of four categories and 16 subcategories of main influencing factors that describe the distribution pattern of water abundance, considering lithological characters, hydraulic characteristics, structural factors, and geophysical parameters that influence aquifer water abundance. The variable weight model is used along with the K-means clustering method in dynamic clustering to determine the variable weight interval thresholds for each index of the main influencing factor. Additionally, we investigated the construction of the state variable weight vector, determination of weight adjustment parameters, quantitative assessment of the interaction relationship, and relative importance of each main influencing factor in relation to aquifer water abundance. Finally, an evaluation method and mathematical model for the water abundance index based on the variable weight model are developed. This study provides a detailed description of the specific implementation steps of the water abundance index method based on the variable weight model, using the direct water-filled J₂z–^#2 coal seam roof fissured aquifer in the Xiaojihan coal mine as a case study. The evaluation results were then verified using drainage data from the working face. The findings indicate that a comparison with the traditional constant weight evaluation results demonstrates that the variable weight model better reflects the characteristics of aquifer heterogeneity and achieves higher prediction accuracy. [ABSTRACT FROM AUTHOR]