Font Size: a A A

Research On Monitoring And Prediction Of Land Subsidence In The Districts Of Lanzhou City

Posted on:2022-04-01Degree:MasterType:Thesis
Country:ChinaCandidate:R R LiFull Text:PDF
GTID:2480306341462724Subject:Traffic mapping information technology
Abstract/Summary:PDF Full Text Request
In recent years,land subsidence has occurred frequently in the districts of Lanzhou City.The land subsidence and the secondary disasters caused by it have the characteristics of diffcult monitoring and prevention.Timely monitoring and forecasting of land subsidence in the districts of Lanzhou City can not only guarantee the safe operation of the city and the safety of people's lives and property,but also provide urban planning departments and land subsidence supervision departments with current and high-precision reference data.Based on the comparison of various methods of land subsidence monitoring and prediction,SBAS-InSAR technology is used to monitor the land subsidence in the districts of Lanzhou city,and LSTM is used to predict the the land subsidence in the districts of Lanzhou city.First,the SBAS-InSAR technology was used to monitor the land subsidence in the districts of Lanzhou city from 2017-02-19 to 2020-10-01 using 54 Sentinel-1A data.Then combined with statistical analysis,standard deviation ellipse method,time series analysis and other methods to analyze the characteristics of land subsidence and space-time changes in the districts of Lanzhou City,and focus on the analysis of the subsidence status of typical subsidence areas.At the same time,the second-class settlement monitoring data was used to verify the accuracy of the SBAS-InSAR technology settlement monitoring results.Finally,uses the time series cumulative settlement obtained by SBAS-InSAR monitoring as the data source,uses the LSTM model to predict the next 20 periods cumulative settlement of the10 points in the typical settlement area,uses the LSTM model and the Kriging interpolation to predict the next20 periods cumulative settlement of the typical settlement area.And using the absolute error,mean absolute error,root mean square error and mean absolute percentage error to analyze the settlement prediction accuracy.Get the following conclusions:(1)From February 19,2017 to October 1,2020,the annual average subsidence rate in Lanzhou urban area is the largest-41mm/a,and the cumulative subsidence is the largest-151.4mm.About 53% of the urban areas have low severity of land subsidence,but there are still 0.06% of areas where the severity of land subsidence is moderate.On the whole,the central city is relatively stable as a whole,and the areas with larger settlements are mainly distributed in some villages at the foothills of the north and south,along the railway,construction area,logistics park.(2)From February 19,2017 to October 1,2020,the land subsidence center in the districts of Lanzhou city gradually moved to the northwest,and the land subsidence area showed a trend of first decline and then rise,and finally showed a downward trend.The land subsidence has obvious directionality in the northwest-southeast direction.And land subsidence continues to develop in a clockwise direction.This is mainly related to the transfer of large-scale construction areas in the urban area of Lanzhou City.(3)The accuracy of the monitoring results of the SBAS-InSAR technology is verified by using the second-class subsidence monitoring data.The results show that the error of the SBAS-InSAR monitoring results is 4.15 mm and less than 10 mm,and SBAS-InSAR can be used to monitor the land subsidence in the urban area of Lanzhou.(4)From February 19,2017 to October 1,2020,the areas with large land subsidence in the districts of Lanzhou city are mainly distributed in Baidaoping Village,Donggang Street,Bali Town,Lanxi Locomotive Depot,Shajingyi Street,Liuquan Town Street and Jiuzhou East Ring Road.The land subsidence supervision department should strengthen the supervision of the above seven aspects.The large settlement in these areas is mainly related to the local loess soil type,engineering construction and railway line operation.(5)From the point and area prediction results,the constructed LSTM neural network settlement prediction model has high accuracy and can be used for the prediction of land subsidence in the districts of Lanzhou City.From the point prediction results,the predicted cumulative settlement of each point in Baidaoping Village is still increasing,and the maximum cumulative settlement is as high as 125.2mm.From the area prediction results,In the next 20 periods,the size and scope of the cumulative settlement value of the predicted area are still being further expanded.The land subsidence supervision department should pay attention to the subsidence in this area.The area prediction results can not only visually see the changes in the cumulative amount,but also clearly see the changes in the settlement range in each cumulative settlement area.The area settlement prediction model constructed in this article is a good settlement prediction forecast means.Human activities have caused major changes on the surface,and a series of changes on the same surface will in turn affect human activities.Obtaining the land subsidence in the districts of Lanzhou city in time and predicting the future subsidence of typical subsidence areas can provide relevant data support for urban geological disaster prevention and control departments,urban planning and layout departments,and construction and construction departments.This is of great significance for maintaining the safe operation of the city and ensuring the safety of people's lives and property.
Keywords/Search Tags:land subsidence monitoring, land subsidence prediction, SBAS-InSAR technology, LSTM, the districts of Lanzhou city
PDF Full Text Request
Related items