基于EMD滤波和云模型的大坝安全监控指标拟定  

Establishing of Security Index for Dam Monitoring Based on Cloud Model and EMD Filtering

作　　者：赵鲲鹏[1,2,3] 梁嘉琛 仇建春[1,2,3] 杨景文[1,2,3] 曹睿哲[1] 

机构地区：[1]河海大学水利水电学院,江苏南京210098 [2]河海大学水文水资源与水利工程科学国家重点实验室,江苏南京210098 [3]河海大学水资源高效利用与工程安全国家工程研究中心,江苏南京210098

出　　处：《人民黄河》2015年第10期120-122,127,共4页Yellow River

基　　金：国家自然科学基金重点项目(41323001;51139001);水利部公益性行业科研专项(201201038;201301061);江苏高校优势学科建设工程资助项目(水利工程)(YS11001);淮安市水利院士工作站资助项目

摘　　要：为克服现有大坝安全监控指标拟定方法的不足,利用云模型,采用数字特征熵来揭示异常值的随机性与模糊性,通过期望、熵和超熵构成的特定结构算法,将大坝安全与否的定性概念定量表示,以此拟定安全监控指标。由于大坝监测效应量影响因素复杂,监测数据常存在高频噪声,因此应对原始数据进行降噪处理。以福建某大坝水平位移为例,先用经验模态分解(EMD)滤波方法去噪,然后采用云模型进行安全监控指标拟定,并与典型的小概率法进行对比分析。结果表明,基于EMD去噪和云模型拟定的监控指标是合理准确的。A cloud model was proposed in order to overcome the limitations of the existing method to establish the security index for the dam monitoring such as the typical probability method and the confidence interval method. The cloud model adopted numeral characteristics of entropy to reveal the randomness and fuzziness of outliers. Through a particular structural algorithm constitute of expected number,entropy and hyper entropy,the cloud model could translate the qualitative concept of dam safety to quantitative expression,which was used to establish the security index for the dam monitoring. The monitoring data often contains high frequency noise because of the complex influence factors,so it is necessary to de-noise the original data. The horizontal displacement of a dam in Fujian Province was taken as the example to testify the method based on empirical mode decomposition( EMD) filtering method and cloud model,and then it was compared with the traditional probability method. The result indicates that the method based on EMD filtering method and cloud model is accurate and reasonable.

关 键 词：大坝 变形 监控指标 EMD滤波 云模型 

分 类 号：TV698.1[水利工程—水利水电工程]