Research And Realize Of Automatic Pick Superimposed Velocity Based On Density Peak And Three Dimensional Imaging

With the continuous expansion of the scale of tunnel construction,the geological and hydrological conditions in tunnel construction tend to be more complicated.Traditional speed analysis methods have many manual interactions and cannot efficiently and accurately pick the superimposed speed.Based on the average amplitude energy and correlation coefficient velocity analysis criteria,this paper conducts research on the automatic picking algorithm of the superimposed velocity in seismic wave data velocity analysis and the three-dimensional imaging of seismic wave data.Among the commonly used superposition velocity picking algorithms,the grid search method and the simulated annealing algorithm can obtain the local maximum value for the superposition velocity inversion.However,for a large amount of seismic wave data,the search range is relatively large,which requires a lot of time,memory and computing resources.At the same time,if there is an error in the calculated superimposition velocity,the position of the abnormal body cannot be accurately located during the three-dimensional imaging.At present,when the three-dimensional imaging tunnel geological advance prediction system selects the seismic wave superimposition velocity,it uses manual velocity scanning analysis method,and the selected superimposition velocity is not unique,low efficiency,and low accuracy.Aiming at the problem of inaccurate selection of seismic wave velocity and high time cost,the algorithm is based on the density peak clustering to automatically extract the superimposed velocity,and the PCL library and ray projection algorithm are used to realize the three-dimensional visualization of seismic wave velocity data.The main research work of this paper is as follows:1.Automatically extract the seismic wave superimposed velocity based on the density peak clustering algorithm.First,use the density peak clustering algorithm to group the seismic wave velocity spectrum;second,use the least square method to linearly fit the amplitude values of the grouped different superimposed velocity signals;then,use the similar function to perform the amplitude value of the fitted velocity.Optimization,to achieve the purpose of the computer automatically selecting the superimposed velocity in the seismic wave superimposed velocity spectrum;finally,the model data and multiple sets of measured data in the tunnel construction are used to verify and analyze the effectiveness of the algorithm in this paper.The results show that the algorithm in this paper achieves good results in terms of prediction accuracy,algorithm stability,and operation speed.2.Use PCL library to realize the three-dimensional imaging of seismic wave superimposed velocity data.Firstly,analyze the binary storage format of seismic wave three-dimensional data;secondly,use PCL library to generate three-dimensional imaging map,and then modify the three-dimensional imaging map through color mapping algorithm and ray projection algorithm;finally realize the three-dimensional imaging of model data and measured data of Dajianshan tunnel.The imaging results intuitively reflect the location of the bad geological bodies in the tunnel,help the interpretation of seismic wave data,provide technical support for tunnel construction,and ensure the smooth completion of the project.