基于空间邻近点与极限学习机的大坝位移缺失数据补齐  

Missing Data Complement of Dam Displacement Based on Spatial Proximity Points and Extreme Learning Machine

作　　者：邱德俊 周洋 仲静文 贾玉豪 QIU Dejun;ZHOU Yang;ZHONG Jingwen;JIA Yuhao(Nanjing Water Planning and Designing Institute Co.,Ltd.,Nanjing 210022,Jiangsu,China;College of Water Conservancy and Hydropower Engineering,Hohai University,Nanjing 210098,Jiangsu,China;Shanghai Investigation,Design&Research Institute Co.,Ltd.,Shanghai 200335,China)

机构地区：[1]南京市水利规划设计院股份有限公司,江苏南京210022 [2]河海大学水利水电学院,江苏南京210098 [3]上海勘测设计研究院有限公司,上海200335

出　　处：《水力发电》2021年第12期98-101,共4页Water Power

基　　金：国家自然科学基金面上项目(51579085)。

摘　　要：准确补齐大坝位移缺失数据,对科学有效地分析大坝变形监测数据和准确可靠地评价大坝安全性态至关重要。从大坝的整体性和连贯性角度出发,构建了一种基于空间邻近点和极限学习机的大坝位移缺失数据补齐方法。同时将目标测点的空间邻近点测值和大坝变形统计模型分量作为大坝位移缺失数据的影响因子,利用输出权重优化的极限学习机算法对缺失数据进行补齐。以某碾压混凝土重力坝为例,利用基于空间邻近点和极限学习机的方法对大坝位移缺失数据补齐,同时与利用BP神经网络算法和基于大坝变形统计模型分量的补齐结果进行对比,结果表明,基于空间邻近点和极限学习机的方法补齐精度更高。Accurately complementing the missing data of dam displacement is very important for analyzing the monitoring data of dam deformation effectively and evaluating the safety behavior of dam accurately.From the point of view of dam integrity and coherence,a method of dam displacement missing data complement based on spatial proximity points and limit learning machine is constructed.The spatial adjacent point measurements of the target measurement points and the statistical model components of dam deformation are taken as the influence factors of dam displacement missing data,and the extreme learning machine algorithm optimized by the output weight is used to complement the missing data.Taking a roller compacted concrete gravity dam as an example,the dam displacement missing data is complemented by using the spatial proximity point and extreme learning machine model,and the results are compared with those based on the back propagation neural network algorithm and the component of the dam deformation statistical model.The comparison shows that the method based on the spatial proximity point and the extreme learning machine has higher complement accuracy.

关 键 词：数据缺失 数据补齐 大坝变形 空间邻近点 极限学习机 

分 类 号：TV698.1[水利工程—水利水电工程]