S-wave Petrophysical Modeling And Bayesian Pre-stack Stochastic Inversion

Seismic inversion is significant to achieve high-precision reservoir prediction and increase oilfield production.The inversion objective is estimating the elastic and physical parameters that can reflect the characteristics of underground structure and lithology from the surface observed data,and finally realizing the evaluation of underground oil and gas properties.With the rapid development of current exploration,the structural oil and gas reservoirs in most domestic oilfields have gradually dried up.Therefore,in recent years,the core target of seismic exploration has turned to lithologic oil and gas reservoirs.At this time,the post-stack inversion technology is usually ineffective.Therefore,it is very important to study prestack inversion.Most pre-stack inversion methods only rely on a single seismic information,and the inversion results have strong multi-solution and low resolution,while the pre-stack inversion methods under the statistical framework of Bayesian pre-stack inversion can integrate seismic,logging,geological and other multi-source information to achieve high-precision and high-resolution inversion.The core of this thesis is the Bayesian pre-stack inversion method.In order to further improve the resolution of inversion results,Cauchy distribution is selected as the prior probability distribution of the Bayesian inversion method,and the theoretical model and actual data are tested respectively to verify the effectiveness of the inversion method.However,the above inversion methods are still not enough to meet the description of thin interbeds in domestic oilfields.Therefore,this thesis also studies geostatistical inversion and carries out pre-stack geostatistical random inversion case analysis based on oilfield examples.Compared with post-stack inversion,the biggest advantage of pre-stack inversion is that it can provide shear wave velocity,which is of great significance for identifying oil and gas bearing property of reservoir and dividing lithofacies.Because the initial model of prestack seismic inversion is mostly derived from logging data,the S-wave velocity information in logging data is very important for prestack seismic inversion.At present,there is a general lack of S-wave information in domestic oilfield logging data,so S-wave prediction is very important.Traditional prediction methods are either based on empirical relationship,lack of universality,or based on complex rock physical model,which is inconvenient to use.This thesis studies a shear wave prediction method based on consolidation index.This method is simple and convenient,which is enough to solve the above problems and provide accurate shear wave information for seismic inversion.