Design And Implementation Of Karst Landslide Monitoring And Early Warning Cloud Platform Based On Hadoop

Karst landslide is a phenomenon of geological hazards widely distributed in the mountainous areas of southwest my country.It is not only restricted by internal geological structure,but also affected by many external factors such as precipitation and mining.In order to effectively assess disaster risks and effectively reduce post-disaster losses,setting up multiple sensors for all-weather real-time monitoring and pre-warning in typical areas where landslides are possible in karst mountainous areas is a realistic approach to comprehensive prevention and control of karst landslide disasters.With the continuous upgrading of sensor technology,the monitoring work centered on informatization is more complete in space,more accurate in time,and more practical in effect.However,the following problems of difficult data integration,such as large amount of monitoring data,multiple types and mixed sources,have become increasingly prominent,which makes the complexity of the transmission,management and analysis of the original monitoring data continuously increase.Therefore,there is an urgent need for a more efficient information integration technology to realize the transmission,management and analysis of various types of landslide monitoring data.Analyze,in order to more effectively explore and make full use of the potential value of hard-won monitoring data.Starting from the study of theories and methods of cloud computing and big data,the thesis explored in depth the container orchestration technology of cloud platforms,the HDFS distributed file system of big data platforms and the Map Reduce parallel programming framework components,etc.,aiming at "early identification and early identification of large landslides in karst mountainous areas" Based on the requirements of “monitoring and early warning”,the karst landslide monitoring and early warning cloud platform is built based on the distributed system infrastructure Hadoop,which realizes the online real-time visualization and automatic processing of karst landslide monitoring data,and on this basis,the karst landslide in the target area is dynamic Monitoring and forecasting and early warning.The main research work and results of this paper are as follows:1.Multi-source data fusion storage and processing based on Hadoop.Starting from the source of multi-source monitoring data,this thesis designed a classified storage scheme for landslide monitoring multi-source data with the support of big data technologies such as HDFS and Map Reduce,and used the Sqoop component to transmit the result data of the big data platform in real time,and achieved the results.The high-speed retrieval of data satisfies the requirements of efficient query,real-time visualization and diversified statistical analysis of cloud platform data.2.Design and implementation of demand-oriented cloud platform.According to the different needs of the members of the project,the paper planned the overall design of the system development,and completed the development of the karst landslide monitoring and early warning cloud platform on the basis of the detailed design.The cloud platform includes practical functions such as real-time statistical analysis of monitoring data,spatial data visualization,three-dimensional stratum model building,and warning information abnormality notification,which lays the foundation for karst landslide warning analysis.3.Deployment of cloud platform for karst landslide monitoring and early warning.At present,the karst landslide monitoring and early warning cloud platform has been deployed in the Chang'an University campus network environment.It has access to more than 30 sensor real-time data in karst landslide demonstration areas such as Faer Town,Guizhou Province,and has completed multiple sources based on the Chang'an University big data platform.Data integration work.Tests have shown that the system functions are relatively complete,which initially meets the actual needs of landslide monitoring and early warning in the karst landslide demonstration area in Guizhou Province.