监测信息不完整状态下的大坝整体性态监控方法  

Monitoring method of dam integrity in state of incomplete monitoring information

作　　者：姜振翔 陈辉 陈鲁皖 JIANG Zhenxiang;CHEN Hui;CHEN Luwan(School of Hydraulic & Ecological Engineering,Nanchang Institute of Technology,Nanchang 330099,China)

机构地区：[1]南昌工程学院水利与生态工程学院,江西南昌330099

出　　处：《南昌工程学院学报》2021年第3期1-6,共6页Journal of Nanchang Institute of Technology

基　　金：江西省教育厅科学技术研究项目(GJJ190970)。

摘　　要：针对大坝原始监测数据不完整、难以融合多测点数据诊断大坝整体性态问题,提出采用概率主成分分析(PPCA)对测值不完整的原始监测数据进行重构,提取表征大坝整体性态的概率主成分(PPC),并建立了PPC监控模型。算例表明,PPCA能够有效地从不完整监测数据中提取关键信息,PPC监控模型能够高精度地实时监控大坝整体性态。In view of the problem that the original monitoring data of the dam is incomplete and it is difficult to fuse the multi-point data to diagnose the dam integrity,the probabilistic principal component analysis(PPCA)is proposed to reconstruct the original monitoring data with incomplete measurement values,extract the probabilistic principal component(PPC)representing the dam integrity,and establish the PPC monitoring model.The example shows that PPCA can effectively extract the key information from the incomplete monitoring data,and PPC monitoring model can monitor the dam integrity in real time with high accuracy.

关 键 词：概率主成分分析 关键信息 数据缺失 性态监控 

分 类 号：TV698[水利工程—水利水电工程]