摘要
大坝位移监测资料分析大多以单一测点得到的数据建立模型,这种单测点模型无法反映大坝作为一个整体各点位移之间的相互联系,也就不能真实地反映坝体的整体安全性态。为此,将多个同源测点联合考虑,通过测点之间的相关性对目标测点建模,再通过该模型推求未布置测点位置的大坝位移值,为推求大坝整体单向位移创造条件。经实例验证,这种新模型在对目标测点的位移预测和补全缺失数据两方面相对常规的单测点模型效果更佳,而且在对未知点的位移值进行预测方面也具有良好效果。
Aiming at the problem of that most of the analyses on dam deformation monitoring data based on establishing model with the data obtained from single measuring point, which cannot fully reflect the correlation among the displacements of all the points of the dam body, and then cannot really reflect the integral safety behavior of dam body as it is an integrity, several homologous measuring points are jointly considered and the target measuring points are modeled through the correlation among all the measuring points, and then the displacement values of those locations without measuring points are derived through this model, so as to create the condition for deriving the integral unidirectional displacement of the dam. Through the verification of the actual case concerned, the effect of this kind of new model is better in the aspects of predicting displacement and completing missing data if compared with those from the conventional single measuring point model, while it has better effect in the aspect of predicting the displacement value of the location without measuring point as well.
作者
季威
刘晓青
林潮宁
陶园
JI Wei;LIU Xiaoqing;LIN Chaoning;TAO Yuan(College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, Jiangsu, China)
出处
《水利水电技术》
北大核心
2019年第2期112-117,共6页
Water Resources and Hydropower Engineering
基金
国家重点研发计划"非常规条件下大坝极限承载能力与破坏模式"(2016YFC0401601)
关键词
多测点模型
大坝安全监测
同源测点
BP神经网络
位移
multiple measuring points model
safety monitoring of dam
homologous measuring points
BP neural network
dis placement
