Research On Key Technology And Application Of Real-time Monitoring And Early Warning System Of Landslide

Geohazards monitoring and early warning is an important means of geohazards prevention and mitigation.Monitoring is the basis of early warning,and early warning is the purpose of monitoring.In recent years,scholars at home and abroad have conducted in-depth research on the methods and technical system of landslide monitoring and early warning,and achieved a lot of research results.But overall,the combination of geography and geology is not close enough,the monitoring and early warning model is difficult to fully consider the landslide deformation process and disaster mechanism,and it is difficult to obtain high early warning accuracy.The developed monitoring and early warning systems are also difficult to meet the real-time monitoring and early warning needs of tens of thousands of hidden danger points.The existing research results are difficult to effectively solve the tasks of “when might it happen” and “strive to achieve early warning three hours in advance”.How to improve the ability of landslide monitoring and early warning,we face many challenges: How to improve the accuracy of landslide early warning? How to apply the theoretical research results to the actual monitoring and early warning,and build a set of real-time monitoring and early warning system of landslide which can be applied in a large-scale and professional way?In order to solve these problems,this paper systematically summarizes the author's practical achievements in monitoring and early warning in the past 10 years,uses advanced technologies such as cloud computing and Internet of things to build a cloud platform for landslide monitoring and early warning,integrates and manages all kinds of data in the whole process of landslide geological disaster evolution,develops a parallel and efficient multi-source heterogeneous monitoring data aggregation platform,integrates multi-source heterogeneous real-time monitoring data,forming a multi-dimensional “Sky-Space-Ground” monitoring data center.Based on the comprehensive analysis of more than 21000 sets of monitoring equipment and more than 126 million measured curves of monitoring data,the paper summarizes and classifies the types of monitoring curves,constructs the reliability evaluation system of monitoring equipment,studies the early warning model of landslide process and the key technologies of its implementation,and constructs a set of hybrid architecture(B/S architecture,C/S architecture,App).The real-time monitoring and early warning system of landslide realizes the integrated management of the whole process of Landslide Evolution,which combines geology with geography,space data and attribute data.The “Process-based Early Warning” of dynamic tracking in the whole process of landslide is realized by computer technology,which effectively improves the precision of landslide early warning.The main achievements are as follows:(1)Constructed the “process-based early warning” model of landslide and its automatic solution algorithm: Combined the three parameters of deformation rate,rate increment and improved tangent angle,constructed a “process-based early warning” model.Starting with the monitoring data of landslide deformation,the monitoring curve types are divided,and the automatic recognition theory and computer technology of the evolution stage of landslide deformation are studied to realize the dynamic tracking and early warning of the whole process of landslide.(2)Established the reliability evaluation system of monitoring equipment and the early warning mechanism of multi equipment linkage: Through the dynamic evaluation of the reliability of monitoring equipment,combined with the linkage early warning mechanism,the reliability of the early warning conclusion is evaluated,so as to improve the success rate of monitoring early warning.The computer technology is used to automatically identify the deformation evolution process of landslides,realize automatic and real-time “process-based early warning”,and provide theoretical and technical support for the business and automatic operation of the early warning model.(3)Constructed the automatic processing method of monitoring data: This paper studies the preprocessing method of the real monitoring data,and provides the related algorithm for the computer to automatically process the monitoring data.By setting up the monitoring data filter and using the Laida Criterion to realize the initial filtering and gross error processing of abnormal data,combined with the data characteristics,the moving average method and the least square method are used to fit the data respectively to identify the deformation trend of the data.Based on the characteristics of monitoring data curve,the corresponding data processing methods are automatically selected to provide more accurate data for the subsequent early warning model calculation and improve the early warning accuracy.(4)Built a real-time and efficient monitoring data integration and sharing unified management platform: Combined with Internet of Things,Message Queuing(MQ),Load Balancing and other technologies,the monitoring data coding system is established,and a set of real-time monitoring data transmission and integration scheme based on MQTT protocol is proposed to realize the integration of multi-source heterogeneous monitoring data terminals and the integrated management of monitoring data collection,transmission and collection,so as to provide real-time data guarantee for monitoring and early warning.(5)Constructed a set of strategy-based real-time early warning technology of landslide deformation process: From the aspects of model calculation,early warning message release,the theoretical model of landslide early warning is combined with practical application.We developed the Early Warning Level Solver,and the general calculation framework of early warning model based on policy.Finally,realize the goal of real-time process-based early warning.(6)Built an integrated data management platform for the whole process of landslide deformation evolution: Based on the multi-dimensional observation technology of “Sky-Space-Ground” landslide,Web GL technology is used to cross platform 3D Digital Earth,providing an intuitive and real 3D reality roaming platform,realizing the integrated management and sharing of massive basic data,real-time monitoring data and video,and also providing a powerful and data rich 3D Display Platform for the real-time monitoring and early warning system.The integrated data management system and comprehensive information model of landslide in the whole process of Landslide Evolution can provide data support for expert early warning decision-making of landslide.(7)Developed a real-time monitoring and early warning system of landslide based on hybrid architecture: Based on the above research results,a hybrid architecture system(B/S,C/S,App)is studied,and a real-time monitoring and early warning system for landslide is developed based on Micro Services.Each architecture closely cooperates to give full play to the advantages of each architecture.Establish a unified data integrated display platform to provide technical solutions for process based early warning model.Realize landslide monitoring and early warning business operations,and provide basic data support and early warning information services for landslide emergency rescue.