intelligent seismic depth velocity modeling and imaging
In the establishment and speed filling work, some manual operations are replaced by intelligent means. In the process of model subdivision and speed iteration, the speed is corrected and subdivided by machine learning means, and the speed field with better details can be obtained.
Machine learning can better integrate a variety of geological information restriction, better describe the horizontal and vertical changes of velocity and the interface of velocity mutation, thus helps the micro amplitude structure become accurate image.
true surface velocity modeling and imaging
The conventional floating surface could make the rugged surface smoothly, and so that the ray path is approximately in vertical direction near the surface. Under complex surface conditions, there is no approximate ray path, which can be imaged more accurately.
After reprocessing, the imaging was significantly improved
extracting and suppressing the strong reflection feature based on the theory of long and short cycle theory
The core lies in the long and short cycle decomposition technology of stratified model. Through the seismic response analysis and top bottom interface interpretation of shielding layer and background layer, the stability reflection coefficient analysis, spectrum analysis, principal component analysis of the shielding layer and time window selection of shielding layer decomposition, the principal component and short cycle component sections are finally generated.
The top surface of Ordovician system is unconformable as a strong wave impedance interface, the internal reflection of upper and lower carbonate strata is weak, so it is difficult to identify the strike slip fault and target stratum. After processing, the stratum of one room set stratum can be tracked, and the imaging of strike slip fault and fracture cave system is clear.