基于最小熵理论和未确知测度理论的泥石流敏感性分析

为了合理确定各影响因子权重，客观分析泥石流敏感性，基于最小熵分析原理和未确知测度理论建立泥石流敏感性分析模型。选取金沙江流域26条典型泥石流沟，确定9个主要影响因子，利用最小熵分析原理分析各因子对泥石流系统贡献率及权重，然后基于未确知测度理论计算各泥石流沟的多指标综合评价尺度向量，最后根据置信度识别准则确定各泥石流沟敏感度水平。将分析结果与已有成果进行对比可知，仅有7条泥石流沟敏感性结果有所差异。将评价结果与现场实际调查情况对比分析可知，22条泥石流沟的敏感度水平与泥石流实际发育情况相符，仅有4条沟评价结果偏于“保守”，因此，利用所提出的模型进行泥石流敏感度分析是十分有效的，并可对泥石流的防治提供合理的建议，可以在实际工程中推广应用。

Debris flow susceptibility analysis based on theories of minimum entropy and uncertainty measurement

For the purpose of legitimately determining each major factor weight and objectively analyzing debris flow susceptibility, a model for analyzing debris flow susceptibility is established based on minimum entropy analysis and uncertainty measurement theory. Nine major factors of 26 typical debris flows along Jinsha River are selected. By utilizing the minimum entropy analysis, the contribution rate to debris flow system and weight of each major factor are obtained. And then, the multi-index comprehensive evaluation measure vector of each debris flow catchment is determined based on the uncertainty measurement theory. The final susceptibility degree is acquired by credible degree recognition criteria. By comparing the debris-flow susceptibility results with reference11], we find that only seven debris flow catchments are different. Comparing the susceptibility results with field observations, 22 of 26 debris flow catchments agrees very well with real status of debris flow; and the susceptibilities of four debris flow catchments are conservative. Thus, the proposed model is shown to be quite effective in predicting debris flow susceptibility, which can be applied to actual engineering; and the analyzing results can provide rational suggestion for debris flow prevention.