Research On The Evaluation Of Geological Hazard Susceptibility In Big Data Environment

Geological disasters are natural disasters which seriously affect economic construction,social development and the safety of people's lives and property,therefore,it is of great significance to establish an evaluation model of geological disasters susceptibility and a warning and forecasting system of geological disasters for protecting the safety of people's property and maintaining social stability.However,there are so many factors that affect geological disasters and interact with the other different factors,making it more difficult for early warning and forecasting of geological disasters.With the development of mapping,remote sensing,sensors and other technologies,the geological disaster prevention and control department has accumulated data on geological disasters from different sources and categories.It is an problem that needs to be solved urgently by geological disaster prevention and control department to develop an evaluation model of geological disasters susceptibility that integrates multi-source spatial-temporal data with the help of the theoretical methods such as geological disaster modeling,spatial analysis and spatio-temporal data mining,and to establish a flexible,fast and elastic geological disaster susceptibility evaluation system based on big data technology.Thus,this paper studies the theories and methods for evaluating the susceptibility of geological disasters in the context of big data.The main work includes:(1)The related research and progress of geological disaster modeling and information technology at home and abroad are systematically summarized.It points out the main problems and key challenges existing in current research,so as to clarify the main research contents and research ideas of the paper.(2)Aiming at the problem of fast retrieval of geological disaster data in big data environment,it defines and describe the storage format of geological disaster data in big data environment and it has developed an improved spatiotemporal coding method for geological disaster based on Geohash.On this basis,a new spatio-temporal index method for geological disaster data is developed by extending the Rowkey of Hbase.(3)Aiming at the problem of accurate evaluation in the geological disaster-prone areas,a comprehensive evaluation model of geological disaster susceptibility which integrates multisource spatio-temporal data was developed.Firstly,based on the distribution characteristics of historical geological disaster data,a spatio-temporal density clustering algorithm for geological disaster data is developed.Furthermore,the method of spatial-temporal clustering and convex hull construction is used to realize the automatic detection of historical geological hazard-prone spatial areas.Finally,considering the different factors that affect the geological disasters,a comprehensive evaluation model of geological disaster susceptibility based on Association Rules Mining was developed,and the feasibility of the model mentioned above was verified by experiments.(4)Aiming at the calculation efficiency problem existing in geological disaster susceptibility evaluation in big data environment,it can solve the parallelization problem of key algorithms of geological disaster susceptibility evaluation model.Aiming at the spatial overlay analysis algorithm in the model,MapReduce is used to solve the key problems in the polygon overlay calculation,such as spatial retrieval,overlay matching load balancing,transboundary polygon processing and so on,which ensure the computational performance of the geological disaster susceptibility evaluation algorithm in the big data environment.(5)Considering the actual needs of geological disaster warning and forecasting,it designed the system architecture of geological disasters evaluation system as well as its main functions,and established a lightweight,efficient and extensible big data management and decision support system for geological disasters,which can realize a prototype system for evaluating the susceptibility of geological disasters with the help of Lily,Hbase and other big data technologies.(6)It summarized the research achievements and main innovations of this paper,and prospected the work that needs to be further researched in this paper.