中文摘要：

泥石流危险度的正确评估是泥石流灾害治理和工程措施的基础，但泥石流危险度评价是一个极其复杂的不确定性问题。基于集对分析理论，探讨了泥石流危险度的集对分析改进新方法，即基于样本评价指标实测值与讨论泥石流标准危险度类别间的接近程度，判别集对同异反，进而应用模糊联系度的概念挖掘样本与讨论类别相邻类别的集对同异反有效信息，以综合定量分析样本的危险度，从而提高评价结果的可靠性和准确性。实例应用及与其他方法对比结果表明，基于改进的集对分析方法来评价泥石流危险度是有效可行的，可以取得理想的结果，且操作简单。

英文摘要：

The dangerous degree evaluation of debris flow is a foundation for the disaster control and engineering measures. However, the evaluation of dangerous degree of debris flow is a very complicate and uncertainty problem. An improved model for the dangerous degree evaluation using the theory of set pair analysis （SPA） was discussed in this paper. The basic principle is that the identical - different - contrary is defined by an approaching degree between the measured index values of sample and corresponding index of standard sets of the dangerous degree. A model for data mining in discrepancy coefficient using the concept of the fuzzy degree of connection was investigated during the procedure of SPA. And the useful data was mined between the sample and adjacent discussed grade. Then the relative membership degree to each grade was calculated, and the dangerous degree of debris flow was predicted quantitatively. The results from the application to practical example and comparisons with other methods are fairly good. And they also show that the improved method of SPA used to assess dangerous degree of debris flow is feasible and effective, and easy to operate. The proposed novel method also can be used to evaluate the other uncertainty engineering problems.