基于改进FIM-未确知测度的尾矿库风险评价模型及应用

Safety evaluation of tailings pond based on FIM-optimization unascertained measure

为了准确评估尾矿库运行安全风险状况,预防溃坝事故的发生,基于因子交互作用矩阵(FIM)和未确知测度理论构建了金属矿山尾矿库的安全性等级评价模型。结合实际情况从防洪强度指标、运行稳定性、结构破坏、尾矿输送与回水、安全管理和环境破坏6个方面选取影响尾矿库稳定性的16个因素指标,构建了各评价指标的未确知测度函数。该模型针对尾矿库危险评价过程的复杂性和指标的不确定性,根据实际情况分别对其进行定性和定量分析,利用FIM方法计算各评价指标的权重,并通过盲数理论对定性指标进行处理,运用兰氏距离函数对等级归属判定进行优化,最后得出最小距离为0.857,对应尾矿库安全等级为Ⅱ级(良好)。研究表明,该评价模型能够综合考虑影响尾矿库安全的各种因素,有效降低了尾矿库危险性评价中主观因素的影响,可为尾矿库的安全评价提供理论依据和可行性参考。

This paper is aimed to accurately assess the operational risk status of tailings ponds and prevent dam break accidents,building a comprehensive evaluation model of unascertained measures of metal mine tailings ponds based on the weighting method of interaction theory,from the engineering technology,safety management and environmental impact of the tailings ponds. In view of this purpose,we firstly clarified the necessity of regular safety assessment of tailings ponds,and analyzed the current status of safety research of tailings ponds. In view of the ambiguity and uncertainty of the tailings pond evaluation indicators,the "Technical Regulations for Tailings Tank Safety"and other documents,we have considered the tailings pond storage system,flood drainage system and backwater system. Large factor indicators,derived six key categories of flood discharge intensity indicators, operational stability, structural damage, tailings transportation and backwater,safety management and environmental damage,and selected 16 evaluation indicators to build the tailings pond risk Evaluation System. In order to consider the influence of index factors,we used the interaction matrix determine the weight of each index. In addition,in order to optimize the unascertained measure model and avoid subjective influence during the evaluation process,we used the blind number theory process the qualitative indicators to generate a blind number measure function. We also introduced the Lang distance function to optimize the membership judgment of the risk level. Finally,we calculated the influence intensity of each index and the risk level of the tailing pond to belong to level Ⅱ. The above research results prove that using the FIM-optimized unascertained measurement model can fully consider the various factors that affect the safety of the tailings pond and the interaction between the indicators,reducing the impact of subjective factors in the evaluation process. It can scientifically evaluate the safety of tailings ponds. In addition,the model also provides a feasibility reference and theoretical basis for the safety management and decisionmaking of the tailings pond. Through the results of the evaluation,the supervision and management of the tailings pond can be strengthened in a targeted manner.