基于集对分析的岩溶塌陷危险性预测评价

Risk prediction and evaluation of the karst collapse based on the set pair mechanism analysis

为准确、有效地预测评价岩溶塌陷的危险性,运用集对分析原理建立岩溶塌陷危险性预测模型。选择岩性、岩溶发育程度、地形地貌、地质构造、土层厚度、土层岩性、地下水位距基岩面距离、地下水位变幅、地下水径流强度、地表水入渗、人工抽水强度和其他人类工程活动12个指标为岩溶塌陷主要影响因素,并将塌陷危险性划分为5个等级,用定量或定性的方法对指标进行赋值,构建其分级标准。运用集对分析方法,将影响因素实测值分别与5个危险性等级评价标准组成集对,计算其联系度,最后根据联系数的排序结果,判定岩溶塌陷危险性等级。利用该方法对贵州定扒地区岩溶塌陷危险性进行预测,结果表明,该模型评价结果与实际情况相符。

The paper aims at introducing a risk prediction and evaluation model for the karst-strata collapse based on the study of the theory of set pair analysis. We have engaged in and investigated and explored the 12 factors that the collapse may get involved with according to the developmental condition of the karst collapse in Guizhou as well as the currently available research findings to our sampling research target. The said 12 influential factors,may all contribute to the constituent embodiment for establishing the risk prediction and evaluation system for the karst collapse,namely,the lithological character of bedrock,the developmental condition of the karst,the landform,the geophysical structure,the soil thickness,the soil lithology,the distance from the underground water level to the bedrock surface,the amplitude of the variations of the underground water sources,the run-off intensity of the underground water,the infiltration situation of the surface water,the artificial pumping intensity and all the human engineering interference activities. And,in terms of the types of all the indexical factors,the evaluation can also be conducted in a quantitative or qualitative way,for example,the risk liabilities can be divided into 5 grades,that is,the stable grade（ 1）,the nearly stable（ 2）,the less stable grade（ 3）,the vulnerable grade to the collapse one（ 4）,the extremely vulnerable grade to the grade at the brink to collapse（ 5）. And,then,it would be possible to adopt the set pair analysis method,that is to say,all the data values of each evaluation index to be calculated can be paired up with one of the corresponding five risk grades,so as to establish the expression formula of its connection degree,with the ultimately determined risk grades of karst collapse serving as the basis of the sequence of connection numbers. And,with the model developed to be used in the risk prediction and evaluation of the karst collapse in Dingpa region,Guizhou,the evaluation results we have gained prove that the region is extremely vulnerable to collapse,which conforms to the earlier prediction of the actual situation of such collapse in the region. Moreover,the research results also indicate that the method we have developed is feasible for the risk prediction of the actual regional karst collapse. Therefore,the method we have developed can be expected to provide a scientific reference for the prevention and control of the local karst collapse.