基于诱发因素响应与DE-SVM模型的滑坡位移预测  

Displacement Prediction of Landslide Based on the Response of Inducing Factors and DE-SVM Model

作　　者：江南 苏爱军[1] JIANG Nan;SU Aijun(Three Gorges Research Center for Geo hazards,Ministry of Education,China University of Geosciences（Wuhan）,Wuhan 430074,China)

机构地区：[1]中国地质大学(武汉)教育部长江三峡库区地质灾害研究中心,湖北武汉430074

出　　处：《安全与环境工程》2018年第5期44-49,54,共7页Safety and Environmental Engineering

基　　金：教育部长江三峡库区地质灾害研究中心开放基金项目(TGRC2001409)

摘　　要：三峡库区滑坡的稳定性直接影响到库区沿岸居民的生命及财产安全,而对滑坡变形的预测对于地质灾害防治有着重要意义。基于三峡库区涉水滑坡多存在阶跃式的变形特征和位移监测数据,采用差分进化(DE)算法对支持向量机(SVM)模型参数进行全局优化,并利用优化后的差分进化-支持向量机(DE-SVM)模型对滑坡位移监测数据进行训练,最终实现了对滑坡位移量的预测。以三峡库区典型滑坡——奉节县藕塘滑坡为例,运用DESVM模型并考虑库水位和降雨作为该滑坡的诱发因素,对该滑坡的变形及其发展趋势进行了预测研究。结果表明:基于诱发因素响应的DE-SVM模型对藕塘滑坡变形的预测结果与该滑坡的实际变形特征基本一致,表明该模型的预测精度高,对三峡库区涉水滑坡位移的预测具有重要的参考价值。The stability of landslide in the reservoir area directly affects the safety and property of the resi dents along the reservoir area. The prediction of landslide deformation has a great significance for the pre vention and the control of geological disasters. Based on the deformation characteristics and displacement monitoring data of water related landslides in the Three Gorges reservoir area,this paper uses the differen tial evolution algorithm to optimize the parameters of SVM,trains the displacement monitoring data with the optimized support vector machine model, and realizes the purpose of predicting the displacement of the landslide. Moreover,taking the typical landslide in reservoir area,the Outang landslide in Fengjie County as an example,using Differential Evolution Support Vector Machine （DE SVM） model and considering reset voir water level and rainfall as inducing factors of the landslide,the paper predicts the deformation and de velopment trend of the landslide. The results show that the prediction results of the DE SVM model based on the inducing factor response are relatively consistent with the actual deformation characteristics of the Qutang landslide,indicating that the model predicts the deformation of the Outang landslide with high pre cision. It has an important reference value in the displacement prediction of the Three Gorges reservoir are a.

关 键 词：滑坡 位移预测 诱发因素响应 差分进化支持-向量机模型 三峡库区 

分 类 号：X45[环境科学与工程—灾害防治] P642.22[天文地球—工程地质学]