Reverse-time migration velocity modeling for tunnel prediction

摘要

Reflection seismology method is the one of the most commonly used methods in tunnel geological prediction, the velocity modeling and the migration imaging are core parts in its process. We propose migration velocity model-building, which is based on reverse-time migration, using the residual curvature analysis and layer stripping strategy to build the velocity model. We estimate the initial velocity model by direct wave and divide the model into several layers. In each layer we select the right amount of control points of velocity. Taking common imaging point gathers we update the velocity by event leveling criteria and the error relationship between the depth of imaging and the formation parameters. The convergent solution are gotten by 2-3 times iterative calculation. In the complex structure and the acute velocity regions, we use the smooth function between adjacent layers and increase the control point to improve the accuracy of the velocity analysis. Its applications in tunnel prediction have good results. Numerical examples and field data show that reasonable velocity model and high quality images have been obtained ahead of the tunnel face by reverse-time migration velocity modeling.