Research On Technologies Of Earthquake Disaster Monitoring And Early Warning Based On Cloud-Edge Collaboration

The prediction of the earthquake is a world problem that cannot be solved in the short term.In recent decades,the Earthquake Early Warning System has developed rapidly at home and abroad,and has made great contributions to the cause of disaster prevention and mitigation.However,in areas with complex geological environment and poor communication conditions,the data transmission delay is large,which poses new challenges to earthquake disaster monitoring and early warning(EDMEW).In this paper,based on the cloud-edge collaborative technology framework,the edge information perception,waveform feature extraction,and cloud model service are realized.On this basis,a prototype system is developed.Through case analysis and simulation,the feasibility and timeliness of EDMEW based on cloud-edge collaboration are analyzed,which provides technical support for the development of EDMEW system based on cloud-edge collaboration.The main work of this paper is as follows:1.Based on the concepts of cloud computing and edge computing,focusing on the problems of data transmission delay and high computational pressure in the earthquake early warning system based on traditional cloud center processing framework,a technical scheme and system architecture of EDMEW based on cloud-edge collaborative framework are proposed.2.Edge nodes perform information perception and feature extraction.Specifically,the edge device collects ground motion data in real time,and uses the internal computing power to realize the edge processing and feature parameter extraction of ground motion signals,reducing the upload of massive data.In the aspect of edge waveform feature extraction,a new method of ground motion P-wave extraction is proposed:a seismic event detection method based on P-wave initial power(IPP)algorithm,and the selection principle of threshold was given.Through simulation and example test,this method is more accurate than the existing STA/LTA and STP/LTP methods,which can greatly reduce the noise interference and reduce the false alarm rate of event extraction.3.Based on cloud computing,the cloud model algorithm and service are studied,and distributed to the seismic monitoring equipment at the edge to enhance the robustness of edge data processing.In epicenter location,the earthquake phase association algorithm based on Bayesian Gaussian mixture model is used to locate the epicenter.The performance of the algorithm and the impact of the number of stations on the accuracy were verified by the test of 5 seismic events in Changning.In terms of magnitude estimation,based on signal-to-noise ratio and STP/LTP method,a synchronous magnitude estimation method for P-wave phase detection in earthquake early warning system(Psnr)is proposed.The timeliness,high precision,stability and high robustness of the Psnr method are verified,which greatly improves the efficiency of earthquake early warning.In terms of intensity rapid report,the regional intensity distribution map is generated based on station intensity and Kriging interpolation algorithm to provide earthquake damage in time.4.A prototype system of EDMEW is developed based on cloud-edge collaboration technology framework.By example analysis and simulation,the feasibility and timeliness of EDMEW technology based on cloud-edge coordination are verified.