Seismic AVO Inversion Method Application With Compressed Sensing Thoery

With the deepening of oil and gas exploration process and the increasing difficulty of exploration and development,oil and gas exploration is gradually changing from conventional oil and gas reservoirs to unconventional oil and gas reservoirs.As one of the most important geophysical methods of oil and gas exploration,seismic exploration plays as more essential part.From post-stack impedance inversion to pre-stack elastic parameter inversion,physical parameter inversion and anisotropic parameter inversion,the research method of seismic inversion become more diversified.Based on pre-stack seismic data,AVO inversion technology can excavate physical information and lithologic information,fluid information and anisotropic information from which angle gather and azimuth gather.It plays an important role in reservoir description,oil dessert prediction and fracture prediction.In addition,AVO inversion technology also benefited from the continuous development of signal acquisition and processing technology.In other words,some new opportunities have emerged that can not be ignored directions is the application of compressed sensing theory in seismic inversion.This paper based on the sparse theory and Compressed Sensing models is introduced then seismic signal processing method based on compressed sensing is studied.The feasibility of the Compressed Sensing model in the field of seismic digital signal processing is verified,and a multi-resolution Compressed Sensing model is selected after seismic data processing analysis.Then,we discussed the accuracy of different Zoeppritz linear approximation equations,and k-u-? equation with better matrix condition performance was selected.On the basis of the compressed sensing model,a Daubechies wavelet field AVO inversion method is studied.The convex set projection algorithm(POCS)is used to solve the inverse problem,and the inversion efficiency is improved by this sparse constraint iterative projection.Analysis a log model test results,we find that the sparse domain seismic AVO inversion can maintain good transverse continuity with high resolution,and density parameters can be more stably inverted than conventional inversion algorithms.In addition,the 3D wide-azimuth seismic exploration are increasingly popularized,and the prediction of the fractured reservoir is becoming more and more important.Therefore,an azimuthal anisotropic AVO inversion based on Compressed Sensing model is further developed for the wide azimuth seismic data.We derived E-?-?N-?T approximate equation form Ruger reflection coefficient equation.Using this equations an azimuthal anisotropic AVO inversion method that could Simultaneous inverting Young's modulus,Poisson's ratio and crack weakness parameters is developed for HTI medium.And an acceptable inversion resolution has been obtained after model tests.Finally,the seismic data inversion method based on Compressed Sensing model is applied to the 3D wide azimuth seismic area of the Ordos Basin.The application example shows that the seismic AVO inversion based on Compressed Sensing model is robust for the case with poor seismic data quality.To a certain extent,the correctness and feasibility of the proposed method are verified.