摘要
为进一步提高降雨型滑坡监测的准确性,解决传统降雨型滑坡监测中仅依赖雨量监测作为判断滑坡依据可能导致的漏报、误报等问题,提出了一种新的基于雨量相关的多变量选择方法来监测滑坡事件,首先通过定量分析现场各监测物理量与降雨量之间的相关关系对变量进行筛选,然后利用主成分分析方法对筛选结果进行降维,最后利用高斯混合模型进行滑坡监测。结果表明:基于雨量相关的多变量选择方法能够实现降雨型滑坡监测的多变量选择,为滑坡监测模型提供合理的输入变量。研究结果可为降雨型滑坡监测提供参考。
In order to improve the accuracy of rainfall-type landslide monitoring,and solve the problems of false alarms that may be caused by relying only on rainfall monitoring as the basis for determining landslides in traditional rainfall-type landslide monitoring,a new multivariate selection method based on rainfall correlation is proposed to monitor landslide events.Firstly,the variables are screened by quantitative analysis of the correlation between the monitored physical quantities on site and the rainfall,then the principal component analysis method is used to reduce the dimensionality of the screening results,and finally the Gaussian mixture model is used for landslide monitoring.The results show that the multivariate selection method based on rainfall correlation can realize the multivariate selection of rainfall-type landslide monitoring and provide reasonable input variables for the landslide monitoring model.The research results can provide reference for rainfall-type landslide monitoring.
作者
方迎潮
余东亮
周广
黄鹏
李虎
郑杰
FANG Yingchao;YU Dongliang;ZHOU Guang;HUANG Peng;LI Hu;ZHENG Jie(PipeChinaSouthWest Pipeline Company,Chengdu 610041;Sichuan Institute of Geological Engineering Investigation Group Co.,Ltd.,Chengdu 610072)
出处
《四川地质学报》
2024年第1期149-153,166,共6页
Acta Geologica Sichuan
关键词
降雨型滑坡
多变量
相关分析
滑坡监测
rainfall-type landslide
multivariate
correlation analysis
landslide monitoring
