针对尾矿库溃坝风险评价过程中存在模糊性及不确定性等问题，将尾矿库溃坝作为一个整体并选取合适评价因子，结合改进的层次分析法及云模型理论，建立了适用于尾矿库的溃坝风险模糊评价模型。算例表明，依据各评价因子的赋值标准生成相应的云模型，能够充分的表征评价过程中模糊性、随机性与离散性的特点，经改进的层次分析法计算得到各评价因子的权重后，按照最大隶属度识别准则进行综合评判。最终结果与实际情况相符，进一步佐证该评价模型的正确性和有效性，为尾矿库溃坝风险评价提供了一条新途径。

Targeting at the fuzziness and uncertainty faced in the process of tailings dam risk evaluation, tailings dam risk was taken as a whole and appropriate evaluation factors were selected, in combination with the improved AHP and cloud model theory, a model suitable for tailings dam risk fuzzy evaluation was established. Numerical examples showed that, the corresponding cloud model generated based on the standard of each evaluation factor assignment could fully characterize the fuzziness, randomness and discreteness in the process of evaluation. After calculating the weight of each evaluation factor using the improved AHP, comprehensive evaluation was conducted according to the principle of maximum membership degree. The result was consistent with the actual situation，which further evidenced the correctness and validity of the evaluation model, and provides a new way for tailings dam risk evaluation.

国家自然科学基金项目（51409139;51569014）；江西省教育厅科学技术研究项目（GJJ14223）；南昌大学研究生创新专项资金项目（cx2015074）