Research And Implementation Of Mine Microseismic Event Location And Cluster Analysis System

As a major coal mining country,China mines 35 percent of the world’s coal resources every year,but the number of casualties caused by coal power disasters is as high as more than 50 percent of the global casualties.In order to ensure the safety of coal mine production,the microseismic monitoring system is more and more widely used to monitor the energy change in the mine area in real time.With the deepening of research and the popularization of application,the situation of safe production is developing year by year.Using microseismic monitoring system for real-time seismic data,automatically pick up to micro seismic source parameters,such as inversion analysis formation mechanism research,geological structure,the discriminant of microseismic events positioning is the core part of the microseismic monitoring system and the research focus,due to the complexity of environmental monitoring,restricting factors and correlation,the existing localization algorithm accuracy remains to be improved;At the same time,due to the different formation mechanisms of microseismic events,the microseismic signals received by the monitoring system will also show different characteristics,so the clustering analysis algorithm can be used to classify the automatically recognized microseismic signals of the system,providing a theoretical basis for the next step of disaster warning,recovery and production.Through analysis,this paper has studied and realized the mine microseismic event positioning and clustering analysis system,and adopted the B/S architecture to realize four functional modules,namely data preprocessing module,microseismic event positioning module,microseismic event clustering module and result visualization module,which have been applied in some mining areas in Liaoning and Shanxi.The data preprocessing module realizes the functions of data reading,energy ratio method picking up and microseismic event interception and storage.The positioning module of microseismic events includes the traditional positioning module and the positioning correction module.In the traditional positioning module,three positioning and main event positioning are realized,among which three positioning can be used for coarse positioning of microseismic events when the number of triggers of monitoring stations is greater than or equal to 3.Main event positioning can be used for relative positioning of events when the number of triggers at the monitoring station is greater than or equal to 4 and the main event is clear.In order to further improve the positioning accuracy,this system has realized the orientation correction module,the module is proposed and implemented an improved algorithm of particle swarm orientation correction based on LCM,the algorithm in the traditional algorithm based on the results of positioning,using Logistic chaotic mapping uniform initial particle swarm is mapped to the solution space,and then using the dynamic inertia weight of the particle swarm algorithm,iterative search for the optimal solution,finally realizes the positioning results of optimization correction.Microseismic events clustering module is proposed and implemented the improvement of K-means based on DTW clustering algorithm,this algorithm fully considering the time sequence characteristics of seismic signal,using DTW distance as the similarity measure between the signal,to eliminate the wave drift error,and improve the K-means algorithm of update strategy,adopts combined with DTW characteristics of the DBA to update method,update procedure of iterative optimization clustering center.The results visualization module realizes the visualization of the positioning results and clustering results of microseismic events.Experiments show that the improved particle swarm positioning correction algorithm can achieve the purpose of positioning correction and improve the positioning accuracy of the system.The improved K-means clustering algorithm can make full use of the waveform characteristics of microseismic signals to achieve relatively high accuracy of microseismic event clustering.