基于Bootstrap方法和LSSVM模型的滑坡位移区间预测  

Interval Prediction of Landslide Displacement Based on Bootstrap Method and LSSVM Model

作　　者：林平 李有鹏 谭彬 LIN Ping;LI Youpeng;TAN Bin(School of Smart City,Chongqing Jiaotong University,Chongqing 400074,China)

机构地区：[1]重庆交通大学智慧城市学院,重庆400074

出　　处：《测绘与空间地理信息》2024年第9期48-51,56,共5页Geomatics & Spatial Information Technology

基　　金：重庆市研究生教育优质课程建设计划项目(JDYZ2019009);重庆交通大学研究生课程思政示范项目(KCSZ2021010)资助。

摘　　要：针对滑坡预测中的不确定性和传统点位移预测无法对预测结果的可靠程度进行有效描述这一问题,本文提出了一种区间预测方法,通过Bootstrap方法和最小二乘支持向量机来构建滑坡位移预测模型。并以三峡库区秭归县树坪滑坡为例,选取ZG85监测点在2007年1月—2012年12月期间的位移、降水量及库水位数据进行研究。研究结果表明,滑坡位移区间预测模型相比于传统的确定性点预测模型,能够同时兼顾准确性和可靠性,为滑坡灾害预测提供了一种新思路。In view of the uncertainty in landslide prediction and the fact that traditional point displacement prediction cannot effectively describe the reliability of prediction results,a landslide displacement interval prediction model based on Bootstrap method and LSSVM model is proposed.Taking Shuping landslide in Zigui county of the Three Gorges Reservoir area as an example,data of ZG85 monitoring point including displacement,precipitation,and reservoir water level from January 2007 to December 2012 are selected for study.The results show that compared with the traditional deterministic point prediction model,the interval prediction model of landslide displacement can give consideration to both accuracy and reliability,providing a new idea for landslide disaster prediction.

关 键 词：滑坡 不确定性 区间预测 BOOTSTRAP方法 最小二乘支持向量机 

分 类 号：P25[天文地球—测绘科学与技术] TB22[天文地球—大地测量学与测量工程]