基于GM-GWO-SVR模型的斜坡形变预测  

Forecasting the Surface Deformation of Slope Based on GM-GWO-SVR Model

作　　者：丁德民 向莉 徐晨希 徐元进[4] DING Demin;XIANG Li;XU Chenxi;XU Yuanjin(Wuhan Comprehensive Transportation Research Institute Co.,Ltd.,Wuhan 430015,China;Changsha Natural Resources Comprehensive Survey Center,China Geological Survey,Ningxiang 410600,China;School of Geography and Information Engineering,China University of Geosciences(Wuhan),Wuhan 430074,China;School of Earth Resources,China University of Geosciences(Wuhan),Wuhan 430074,China)

机构地区：[1]武汉综合交通研究院有限公司,湖北武汉430015 [2]中国地质调查局长沙自然资源综合调查中心,湖南宁乡410600 [3]中国地质大学(武汉)地理与信息工程学院,湖北武汉430074 [4]中国地质大学(武汉)资源学院,湖北武汉430074

出　　处：《云南师范大学学报（自然科学版）》2024年第1期35-40,共6页Journal of Yunnan Normal University:Natural Sciences Edition

基　　金：湖北省交通运输厅科技资助项目(2022-11-4-8)。

摘　　要：选取湖北省秭归县屈家坪斜坡作为研究区,使用56期Sentinel-1数据,采用SBAS-InSAR技术提取斜坡形变信息,分析发现斜坡呈现三处明显负形变,与降雨集中时段相吻合.在此基础上,建立了非等距GM(1,2)、GM-SVR、GM-GWO-SVR预测模型,并对研究区进行形变预测,预测结果经MAE、RMSE、MAPE和SSE四个指标评估,结果表明GM-GWO-SVR模型的预测效果最佳.This article selects the Qujiaping slope in Zigui County,Hubei Province as the study area,and extracts the information of slope deformation by SBAS-InSAR technology,using Sentinel-1 data of 56 periods.An analysis shows that the slope exhibits three obvious negative deformations,which are consistent with the concentrated rainfall periods.So non-equidistant GM(1,2)model,GM-SVR model and GM-GWO-SVR model are established,and deformation prediction is carried out.The prediction results are evaluated by four indicators(MAE,RMSE,MAPE,and SSE).The results show that the GM-GWO-SVR model is better than the other two models,and is very effective.

关 键 词：形变预测 GM-GWO-SVR 斜坡 SBAS-InSAR 

分 类 号：P694[天文地球—地质学] P237