Tomography Velocity Inversion Based On Common Imaging Gathers

With the higher requirement in the field of Petroleum Exploration and Production, seismic data processing technology has been gradually reveals too many flaws of its own in the conventional oil industry, for example, results after migration usually can not let all migrations with data from different domains yield the same subsurface image, and incorrect reflect interface. In the progress of seismic data processing, seismic waves used almost don't penetrate the target layer or there are difference velocity value between the target layer and the one upper. Due to those reasons talked above, there are lots fake imaging points. At the same time, because most methods used to migration can not resolve the coupling relation between velocity and depths. Each of those above can cause discontinuous in the events and abnormality of amplitude. Faced to these flows of their own methods, we need to try more chances, as the content of this thesis named Tomographic velocity analysis. Tomographic velocity analysis done based on the results of migration can make velocity update better and events become continuous better.Tomography introduced to seismic data progress in the 80th last century has been widely used in the field of statics correction and deeply velocity updating. The basic theories have been sorted to two kinds, wave theoretical tomography and ray tomography. Comparing the two methods, the first one has some disadvantages, but the later one has developed well. Ray tomography consists of forward and inversion. The most difficult problem of ray tomography is how to pick up accurate travel time. To solve this problem, this paper doesn't use gathers in the conventional tomographic velocity analysis but for common imaging point gathers, uses all the advantage of migrated dataset, and improves method of picking up travel time of real seismic dataset. Traverse velocity after migration to the initial velocity model used to tomography. Ray trace on the initial velocity model, and the angle of rays must be the same as the angle on the common imaging gathers. Then traverse depth information on the common imaging gathers to time information. Prepare all the data used to inversion, choose nonlinear inversion which can seek all possible result. As a result, tomographic velocity analysis used in this paper make a better result.