The Researches On Improving The Precision Of The Multi-component Seismic Velocity Analysis

Multi-component velocity analysis is an important part in the Multi-component seismic data prospecting, and how to improve the precision of P-wave and S-wave velocities is a key problem.Because of the ray asymmetry of the downgoing and upgoing waves, the velocity analysis based on the CMP gather cannot be adapted to that of the converted wave (CCP). Main difficulties are from special characteristics of the CCP bin and CCP time-distance relation. For one thing, even if we assume that the reflection layer is homogeneous and horizontal, the position of the common converting reflection point will be changed with the offset, the depth of reflection layer and ratio of the S to P wave velocity. For another, the time-distance relation of the converted wave is not hyperbolic. If using hyperbolic time-distance relation to make out velocity analysis, we cannot get the correct value of S-wave velocity, especially with the shallow-middle events or large offsets. On the other hand, if we analyze the velocities using all traces traditionally, the resolution of velocity spectra will decrease greatly when one wishes to distinguish between neighboring primary events from reflectors with conflicting dip, or to identify primaries in the presence of multiples.Based on the previous work, by analyzing, researching and concluding the problems above, this thesis has finished the following researches. First, analyzed the CCP traveling characteristics, factors and computation formulae, the method of CCP gathering bin has been realized. Secondly, the traditional velocity analysis is made out with CCP time-distance relation and the formulae of CCP computing. Thirdly, based on the traditional velocity analysis, a so-called selective correlation velocity analysis, which can improve the resolution of velocity spectra greatly, has been systematically studied and accomplished. At last, three methods above were checked and compared with theoretical and practical data. Conclusively, it is shown that the selective correlation velocity analysis is a highly accurate method of the high resolution.