基于含水层渗透系数研究的云-Markov模型：建立与应用

渗透系数的精度对地下水流和溶质运移有重要影响，传统方法在计算渗透系数的过程中均存在一些局限性。本文建立了一种新的云-Markov模型对渗透系数进行预测：利用云模型中的多条件多规则不确定推理技术，根据样品的粒径分布对渗透系数进行预测，并对其进行误差分析；在此基础上利用权Markov链对预测误差的随机性进行模拟，进而根据此模拟值对云模型的预测结果进行校正，将校正后的预测值作为云-Markov模型最终的计算结果输出，即完成对一个样品渗透系数的预测。将该模型应用于华北平原典型区冲洪积扇含水层参数研究，计算结果表明：与渗透系数的实测值相比，云模型的误差相对数介于0.996~1.178间，通过权Markov误差校正后，云-Markov模型的误差相对数为1.021~1.134，预测精度较最初的云模型有了一定的提高。故与传统模型相比，云-Markov模型基本可以应用于含水层渗透系数的计算。

A Cloud-Markov model based on aquifer hydraulic conductivity Research: Establishment and Application

Accurate estimation of hydraulic conductivity is great importance in the analysis of groundwater quantity assessment. Although many methods can be selected to determine hydraulic conductivity of sediments, each method has faults that limit application. A new model Cloud-Markov was developed to estimate hydraulic conductivity in this paper. The Cloud-Markov model mainly included two steps: first, the cloud model was employed to predict hydraulic conductivity of sediment according to its grain size distribution; and then the weight Markov chain was used to simulate random of prediction error distribution. The simulated value of weight Markov chain was employed to correct the prediction results of cloud model. Finally, the correction prediction result was outputted as the finally calculation result of Cloud-Markov model. This model was applied in Luancheng aquifer deposit in the southeast Hutuo River alluvial-pluvial fan in the North China Plain (NCP), the results indicated that: compared to hydraulic conductivity measured by permeameter test, the relative number of error was 0.996~1.178, through the weight Markov method error correction, the relative number of error decreased to 1.021~1.134, the prediction accuracy of Cloud-Markov model was higher than Cloud model. So compared to other methods, the Cloud-Markov model could be successfully used to estimate hydraulic conductivity.