考虑诱发因素影响滞后性的库岸滑坡位移预测  

Displacement prediction of reservoir landslide considering lag effect of inducing factors

作　　者：陈亮青 邹宗兴[1] 苑谊[2] 王艳昆 CHEN Liangqing;ZOU Zongxing;Yuan Yi;WANG Yankun(Three Gorges Research Center for Geo - hazards, China University of Geosciences, Wuhan 430074, China;Department of Land and Resources of Hubei Province, 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 [2]湖北省国土资源厅,湖北武汉430074

出　　处：《人民长江》2018年第10期60-65,共6页Yangtze River

基　　金：国家自然科学基金项目(41502290)

摘　　要：针对诱发因素对于滑坡位移变形的滞后影响,采用平均影响值法(MIV)对不同滞后期诱发因素进行筛选,然后结合广义回归神经网络(GRNN)建立了MIV-GRNN滑坡位移混合预测模型。以三峡库区具有代表性的树坪滑坡为例,将滑坡位移时间序列分解为趋势项和周期项,运用多项式和MIV-GRNN模型分别预测趋势项和周期项位移。分析结果表明:MIV-GRNN模型可以较好地反映诱发因素对滑坡位移滞后性的影响,与传统预测模型相比最大相对误差减少了11.2%。In view of the lag effect of inducing factors on displacement and deformation of landslides,the Mean Impact Value（ MIV） method is used to screen the inducing factors for different lag phases,and then the hybrid MIV-GRNN model is set to predict the landslide displacement combined with the Generalized Regression Neural Network（ GRNN）. Taking Shuping Landslide in Three Gorges Reservoir area as an example,the time series of landslide displacement is decomposed into trend term and periodic term,and the polynomial and MIV-GRNN models are used to predict the trend term and periodic term displacement respectively. The results show that MIV-GRNN model can well reflect the lag influences of inducing factors on landslide displacement. Compared with the traditional prediction model,the maximum relative error is reduced by 11. 2%.

关 键 词：位移预测 滞后性 诱发因素 树坪滑坡 三峡库区 

分 类 号：P642[天文地球—工程地质学]