Design And Implementation Of The Landslide Monitoring And Warning Cloud Platform

With the launch of various earth observation satellites,the era of remote sensing big data has come,which provides massive data sources for landslide disaster monitoring.At the same time,various Internet technologies and deep learning algorithms bring forth the new,providing intelligent algorithms for deformation prediction of landslide disasters.Modern spatial earth observation technologies(satellite remote sensing,aerial remote sensing,GNSS,etc.)have been widely used in geological disasters research,including early identification of disaster bodies,pre-disaster monitoring and early warning,post-disaster emergency response,recovery and reconstruction,etc.,providing important information for the research,decision-making and planning of government.However,how to efficiently manage data and use massive data to prevent and control landslide disaster has become an important problem we are facing now.To address the above problems,this paper carried out a research on landslide disaster prediction and early warning by using multi-source data.The main research contents and achievements of this paper are as follows:(1)In order to solve the problems of low accuracy of traditional reservoir area landslide displacement prediction methods and difficulty in obtaining reliable long time series prediction results,this paper proposes the method of decomposing long time series displacement into trend term and period term based on In SAR deformation monitoring data,and each sub-term is predicted separately.Taking Guobu Landslide as an example,the time series decomposition method was used to decompose the displacement into trend term and periodic term,and the rainfall and reservoir water level changes were selected as the influencing factors.The LSTM,BP neural network and SARIMA models were used to forecast and analyze the periodic term,the LSTM model had the best prediction effect.The root mean square error RMSE of the total displacement predicted value was 3.21 mm,and the correlation coefficient R was 0.986.The results show that the combined prediction method based on In SAR monitoring results and LSTM model can effectively reflect the influence of induced factors on the change of landslide displacement in the reservoir area,improve the accuracy of In SAR deformation prediction results,and this method has a good application prospect in the landslide prediction in the reservoir area.(2)Aiming at the problems of storage and management of massive landslide monitoring data,which are difficult to be fully utilized in disaster prevention and control work,this paper proposes a combination of point and surface monitoring platforms to achieve monitoring and identification of landslide areas and prediction and early warning work,and designs a landslide disaster monitoring and early warning platform that combines computer,database and Web GIS technical with multi-source landslide monitoring data as the core.The platform mainly consists of four modules: spatial data results display and analysis,historical data query,time-series displacement prediction,and landslide early warning.The platform can provide a variety of functions such as 3D earth browsing,spatial data query,spatial measurement,spatial data analysis,management of historical data,prediction of long time series displacement data and early warning information query and release.(3)Taking Longyangxia reservoir area landslide as an example,a landslide disaster monitoring and early warning cloud platform integrating 3D data display,monitoring data query,multi-source monitoring data analysis,prediction and warning is built,which can realize storage,management,processing and analysis of all types of landslide geological disaster historical data in Longyangxia reservoir area.The landslide monitoring and early warning cloud platform has stronger applicability compared with traditional landslide disaster investigation and monitoring means,and can provide data service platform for early warning and prediction as well as prevention and control prevention of landslide disasters in the reservoir area nationwide,and try our best to reduce the affected population and economic loss in the reservoir area,which is of great significance to the prevention and control of landslide disasters in the reservoir area.