Navegando por Assunto "Velocity analysis"

Agora exibindo 1 - 2 de 2

Acesso aberto (Open Access)
Estimativa de parâmetros em meios VTI usando aproximações de sobretempo não hiperbólicas
(Universidade Federal do Pará, 2015-09-30) PEREIRA, Rubenvaldo Monteiro; CRUZ, João Carlos Ribeiro; http://lattes.cnpq.br/8498743497664023
Transversely isotropic (TI) media is a more realistic model for processing seismic data, for example, fractured media with preferred fracture direction, or composite by periodic thin layers. In particular, TI media with vertical symmetry axis (VTI) are widely used as models for P-wave propagation in shales, abundant rock in hydrocarbon reservoirs. However, the P-wave propagation in homogeneous media VTI have as their main characteristics, depend on four parameters of rigidity and also to possess: complicated algebraically phase velocity equation, difficult group velocity equation to explain and moveout equation nonhyperbolic. Therefore, several authors have presented parameterization and obtained approximations to these equations depending on three parameters only. Among these, the moveout approximations have been widely used in inverse methods to estimate lithological parameters in homogeneous media VTI. Such methods have generally been successful in estimated stacking velocity vn and the anellipticity parameter η, since these are the only ones required for generating initial models for the steps of seismic processing in the time domain. One of the most used methods for estimating parameters is the basedsemblance velocity analysis, though, because this method is limited to sections with small offset-depth ratio, adaptations for anisotropic media, considering nonhyperbolic moveout approximatios are required. In this paper, based on anelliptical approximation shifted hyperbola, anelliptical rational approximations are presented for: phase velocity, group velocity and moveout nonhyperbolic in homogeneous VTI horizontally layered media. The validity of these approximations is made by calculating their relative errors by comparing with other known approximations in the literature. To semblance-based velocity analysis is performed to measure the accuracy of the rational moveout approximations to estimate parameters in VTI media. The results demonstrate the great potential of rational approximations in inverse problems. In order to adapt to VTI media, we modify two coherence measurements by semblance which are sensitive to amplitude and phase variations. The accuracy and robustness of the adapted coherence measurements are validated by estimation of in anisotropic parameters in VTI media.
Acesso aberto (Open Access)
Processamento de dados sísmicos reais da região amazônica
(Universidade Federal do Pará, 2006-05-17) GOMES, Anderson Batista; LEITE, Lourenildo Williame Barbosa; http://lattes.cnpq.br/8588738536047617
The treatment of seismic data is divided basically in three parts: preprocessing and processing an imaging. In the present thesis we discuss the stages of preprocessing and two important methods of processing directed to the simulation of zero offset (ZO) sections from multiple coverage data. Conventional (NMO/DMO) processing, and the Common Reflection Surface (CRS) processing have been applied to seismic data from some seismic lines of group 204 of the data set of the Tacutu Graben (Brazil). We used the CWP/SU System to carried out the stages of preprocessing and the stage of conventional (NMO/DMO) processing. The CRS processing was carried out with WIT/CRS System. The stages of preprocessing consisted basically of three parts: organization of the geometry; zeroing and muting of noisy traces; and filtering in the temporal frequency (f filter) and velocity filter (f-k filter). Deconvolution was carried out, however due to results that did not bring any information; the results were not of further use. Besides, the elevation static correction was not applied because the topography is very smooth (elevation variation less than 20 m) in Tacutu plateau. The quality of the results of NMO/DMO processing was strongly biased due to the dependence of the method on a velocity model, that in this case it was accurate enough. We also found difficulties with the velocity analysis (VA) due to great amount of noise present in the data. As a consequence, the normal moveout correction (NMO) and migration did not generate better results. Based on the estimated attributes of the CRS stack method, a smooth macrovelocity model was obtained using reflection tomographic inversion. Using this macro-model, pre- and post- stack depth migration were carried out. Also, the CRS attributes are used in the method residual static correction, and the results demonstrate a better resolution of the stacked section. The sections resulting from stack and migration have been interpreted aiming at the delineation of structures. From the visual details of the panels, we have interpreted thinning, a main faulted anticline and discontinuity, and plays of horsts and grabens, and rollovers were traced. On the other hand, the basement could not easily be traced.