云神经网络确定含水层参数研究

高效、精确的含水层参数求解方法一直是水文地质研究领域的重要研究内容之一。实践中通常利用非稳定流抽水试验资料通过配线法确定含水层参数,但是随着计算机应用的普及,已有人开发出几种在非稳定流试验条件下求解含水层水文地质参数的快速、精确的计算机智能优化算法。在此基础上尝试建立了云神经网络模型(Cloud Neural Net,CNN),并将其应用于石家庄市元氏县3个单孔非稳定流抽水试验,对承压含水层参数进行计算,模型计算结果与当地的水文地质条件较为符合,且比传统方法及单纯的人工神经网络模型所得结果更加精确。因此云神经网络模型为研究区地下水资源评价、地下水数值模拟以及溶质运移模拟提供了另一种重要手段。

Determination of Aquifer Parameters Using Cloud Neural Network

Efficient and accurate solutions for determination of aquifer parameters have been one of the most important research topics in hydrogeological research field. The fitting curve method is usually used to determine the aquifer parameters from unsteady pumping test. With the wide computer application, several rapid and accurate computer intelligence optimization algorithms were developed to determine the aquifer parameters under the conditions of unsteady flow. On this basis, the Cloud Neural Net (CNN) model was applied in this paper to calculate the hydraulic parameters of a confined aquifer in Yuanshi County of Shijiazhuang City based on 3 single-hole unsteady flow pumping tests. The model results were in accordance with the actual hydrogeological conditions, and more accurate compared with the results derived from the traditional method and simplified artificial neural net model. Thus CNN model establishes a good foundation for groundwater resources assessment, groundwater numerical simulation, as well as solute transportation simulation.