加权优化的D-S证据理论在大坝安全评价中的应用

Application of Weight Optimized D-S Evidence Theory in Dam Safety Evaluation

针对传统D-S证据理论存在鲁棒性差与"一票否决"等问题,提出采用加权融合法优化D-S证据理论,以提高其在大坝安全评价应用中的可靠性。首先对原始监测数据进行标准化及归一化处理;然后用欧氏距离确定权重,利用各监测项目间的内在联系进行D-S证据融合;最后根据加权融合结果对西溪水库大坝进行安全评价。结果表明,大坝整体的安全与较安全指标概率赋值达0.527,远大于较不安全指标0.049,大坝处于安全状态;坝基渗流较不安全概率赋值达0.492,存在一定的渗流安全隐患,融合评价结果与工程实际相符合,说明该方法可如实反映大坝的安全状况。

In order to solve the problem that the traditional D-S evidence theory has drawbacks of poor robustness and‘one-vote veto＇,a weight fusion method was proposed to optimize the D-S evidence theory for improving the reliability of application in dam safety evaluation.First of all,it standardized and normalized the original monitoring data.Then the Euclidean distance was used to determine the weight and the data could be integrated by considering inherent relationship between the monitoring projects.Finally,the safety evaluation of Xixi reservoir was carried out according to the result of the weight fusion.The results show that the probability assignment of safety index is 0.527,which is far greater than0.049,and the dam is in safe state.Probability assignment of relatively unsafe index of dam foundation seepage reaches0.492,which shows that there is some hidden danger of seepage.The whole result of fusion evaluation accords with the engineering practice and it indicates that the method can scientifically reflect the safety of the dam.