基于克里格法的地下水流数值模拟模型的替代模型研究

以内蒙古乌拉特中旗金泉工业园区水源地为研究区,建立了研究区地下水流数值模拟模型.采用拉丁超立方方法得到输入(抽水量)输出(水位降深)数据集,运用克里格法建立了地下水流数值模拟模型的替代模型—克里格模型.将克里格模型输出结果与地下水流数值模拟模型输出数据集进行对比发现,克里格模型和地下水流数值模拟模型得到的地下水位降深均值的拟合平均相对误差为0.22%,地下水位降深剩余标准差的拟合平均相对误差为0.03%,拟合误差很小.表明克里格模型可以有效地替代地下水流数值模拟模型,这为大幅减小优化模型在优化迭代求解过程中多次调用模拟模型造成的巨大计算负荷提供一种有效的替代方法.

Surrogate model of numerieal simulation model of groundwater based on Kriging

Jinquan Industrial Park groundwater sources in Neimenggu was selected as the study area, and the groundwater numerical simulation model for this area was established. Latin hypercube samping was used to obtain input (pumping) and output (water level drawdown) data sets, and the Kriging was proposed to establish an surrogate model of the groundwater numerical model, namely Kriging model. Compared the outputs of kriging model with that of numerical simulation model of groundwater, the fitting mean relative error of mean groundwater level drawdown obtained by Kriging model and numerical simulation model of groundwater was 0.22%, and the fitting mean relative error of the remaining average relative standard deviation of groundwater level drawdown was 0.03%. So the fitting error was small, which indicated that the Kriging model can effectively replace groundwater numerical model. This study will provide an effective surrogate method to reduce computational load resulted from multiple invocation of the the numerical simulation model of groundwater in the processes of optimization solution by optimization model.