Research And Application On Local Structure Based Seismic Geometric Attributes

In recent years, people demand more and more energy, especially petroleum. According to the experts estimate that proven oil resources will be facing depletion in the next30to50years. Thus, oil and gas prospecting is a very important goal of petroleum exploration field. Typically, faults and fractures can become a container of oil and gas resources. Therefore, identifying faults and fractures is particularly important for oil and gas exploration, and seismic geometric attributes just can meet this requirement.This paper aims at improving the resolution of3D seismic attributes, and highlighting the fault and fracture and other geologic structure. This paper mainly studied and discussed on the structure tensor of three-dimensional signal.Firstly this paper introduces the concept and classification of the seismic geometric attributes, and details the latest research progress at home and abroad.Secondly we study the dip and azimuth, coherence, curvature attribute and lateral changes in amplitude, and make a theoretical analysis to fix the window size of geometric attributes extraction of the seismic data. We apply it to real seismic data and obtain good effects.Again, we mainly study the coherence attribute of the anisotropic gradient structure tensor. In order to improve the spatial resolution of seismic attributes, in this paper, an anisotropic gradient structure tensor based on anisotropic window function is proposed.Finally, aiming at the problem of the gradient structure tensor not sensitive to even symmetries, an energy structure tensor is introduced. The spatial localization of the energy structure tensor is more accurate, and highlights the subtle fractures that the gradient structure tensor can not detect. In order to improve the resolution of seismic attributes, we use a Gaussian smoothing window function to smooth the Response of the quadrature filter. Meanwhile, the multi-scale radial kernel functions in the energy structure tensor prominent seismic attribute information on the different scales.