首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于集成学习的阶跃型滑坡阶跃点判别分析
引用本文:杨光辉,简文星,张树坡,付智勇.基于集成学习的阶跃型滑坡阶跃点判别分析[J].中国地质灾害与防治学报,2019,30(4):1-8.
作者姓名:杨光辉  简文星  张树坡  付智勇
作者单位:中国地质大学(武汉)工程学院
基金项目:国家自然科学基金项目:三峡库区典型顺向岸坡库水与降雨联合作用失稳观测与理论解析(41272306)
摘    要:针对阶跃型滑坡阶跃点识别和预测难的问题,提出了一种基于聚类分析和集成学习的阶跃型滑坡阶跃点识别和判别模型。以三峡库区八字门滑坡ZG110钻孔2010年4月至2016年12月80个滑坡位移、库水位和降雨数据为例,通过聚类分析方法识别滑坡累积位移-时间曲线中的阶跃点和平稳点,并利用K均值聚类分析检验分类结果的准确性。基于灰色关联确定了滑坡位移的最佳诱发因素,结合随机森林模型建立阶跃型滑坡阶跃点判别模型并利用八字门滑坡ZG111钻孔验证该模型的准确性。模型阶跃点和平稳点的识别准确率均达90%以上,表明该方法在阶跃型滑坡识别中具有较好的适用性,可为阶跃型滑坡的预测提供参考。

关 键 词:阶跃型滑坡  阶跃点  聚类分析  集成学习  预测

Discrimination step points of step-landslides based on ensemble learning
YANG Guanghui,JIAN Wenxing,ZHANG Shupo,FU Zhiyong.Discrimination step points of step-landslides based on ensemble learning[J].The Chinese Journal of Geological Hazard and Control,2019,30(4):1-8.
Authors:YANG Guanghui  JIAN Wenxing  ZHANG Shupo  FU Zhiyong
Institution:(Faculty of Engineering,China University of Geosciences,Wuhan,Hubei430074,China)
Abstract:Aiming at the difficulty in predicting and identifying step points of step-landslides,a method for identifying and distinguishing step points of step-landslides based on clustering analysis and ensemble learning is proposed.With displacement,reservoir water level and rainfall data of Bazimen landslide from April 2010 to December 2016,the step points and stable points were identified by cluster analysis method in the cumulative displacement-time curve of landslide,and the accuracy of classification results was tested by K-means cluster analysis.The grey correlation method was used to determine the suitable influencing factors and the random forest model was adopted to establish the discriminant model of step-landslide,the accuracy of the method was verified by the drill ZG111 of Bazimen landslide.The accuracy rate of identifying of the step points and stationary points is more than 90%,which indicates that the proposed method has a good ability in identifying and prediction of step-landslide.
Keywords:step-landslide  step point  cluster analysis  ensemble learning  predict
本文献已被 CNKI 维普 等数据库收录!
点击此处可从《中国地质灾害与防治学报》浏览原始摘要信息
点击此处可从《中国地质灾害与防治学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号