Research On Settlement Monitoring Along Expressway Based On InSAR Technology

Expressway is an important part of China's transportation industry,and it is an important infrastructure related to the development of urbanization and residents' life.In the highways operation process,different degrees of settlement will happen along the highways.Serious land subsidence will cause a series of geological disasters,which will affect the safe operation of expressways and the safety of human lives and property.Therefore,it is very important to monitor the settlement along the highways in time,ensure the safe operation of the highways,prevent the occurrence of the settlement disaster along the highways,and repair the serious settlement areas.In this paper,two time series InSAR technologies PS-InSAR technology and SBAS technology and sentinel-1 data from July 2018 to December 2019 are used to monitor the settlement of expressways across Kunming,Yuxi and Qujing in Yunnan Province.This paper analyzes the application characteristics of the two data processing methods in the deformation monitoring of highways and the settlement of expressways in the study area;at the same time,overlay the precipitation,groundwater level data in the same period,the geological conditions,construction of the settlement area and the settlement sequence to analyze settlement causes;and the Least Squares Support Vector Machines model is used to predict the settlement of expressways,and contrast the prediction results of neural network model and grey model.The experimental results show that:1.The distribution results of settlement area by the two methods is basically the same,SBAS technology is more suitable for non-urban areas and vegetation-covered area;2.There are different degrees of settlement phenomena in Kunming-Mohan expressway,Hangzhou-Ruili expressway,Kunming Ring Expressway,Chongqing-Kunming Expressway and Beijing-Kunming expressway;3.The settlement along highways is closely related to the geological,construction conditions of the area,rainfall,underground water level and highway dynamic load;4.Compared with the other two models,the Least Squares Support Vector Machines model has higher accuracy and is more suitable for time series settlement prediction of expressways.In this paper,time series InSAR technology is used to monitor the settlement and predict the settlement along expressways,which can provide a theoretical reference for the prevention and prediction of Expressway settlement in the study area.