Research On The Method Of High-precision Time-lapse AVO Inversion For Elastic Parameters And Reservoir Parameters

Time-lapse seismic technology is a method for the reservoir monitoring.It is defined as two or more seismic surveys that are acquired across the same area at different times during hydrocarbon production.From the variations in the seismic signature,the changing situation at the reservoir can be predicted.It can be used to perform the dynamic monitoring and management for the reservoir.In theory,the differences in seismic imaging results or inversion results between the baseline survey and monitor survey describe the dynamic changes of the reservoir fluids,such as the fluid saturation change and effective pressure change.The differences do not include the static properties of rocks underground,such as the structure and lithology.The elasticparameter changes and reservoir-parameter changes caused by the oil and gas exploitation can be obtained through this method.It can be used to monitor the boundary changes of oil and gas and the movement of injected fluids,guide the rational arrangement of the development well location,and improve the ultimate oil-gas recovery factor.Time-lapse AVO(Amplitude variation with offset)inversion is widely used in the time-lapse seismic analysis.It is based on the law of seismic reflection amplitude versus offset,and can be used to estimate the changes in the elastic parameters by using the pre-stack common midpoint gathers in the baseline and monitor surveys.Time-lapse AVO inversion usually includes two steps: 1)estimating the changes in the elastic parameters,and 2)estimation of reservoir-parameter changes using a rock-physical model.Time-lapse AVO inversion is simple and intuitive in theory,and can be used to estimate the lithologically related parameters.It is easily implemented and has a wide and effective application.However,there are still some disadvantages in the traditional time-lapse AVO inversion method.First,the inversion is usually based on the linear approximate expression of the Zoeppritz equations.There are certain constraints and some errors when using the approximate expression.Secondly,The real-life reservoir is a two-phase medium that consists of solid and fluid components.However,the reservoir needs to be treated as a single-phase solid in the traditional time-lapse AVO inversion based on the Zoeppritz equations.This will cause some errors.Thirdly,estimating reservoir-parameter changes in a two-step manner may introduce cumulative errors.Aiming at the above problems,the thesis improves and developes the existing time-lapse AVO inversion method,and proposes two time-lapse joint AVO inversion methods that are based on the exact Zoeppritz equations and the theory of two-phase media respectively.The thesis applies the method of time-lapse AVO inversion to estimate the time-lapse changes in the elastic parameters and reservoir parameters accurately.The research contents include four aspects.The first is the research on the time-shift correction.The time shift means the difference in the two-way traveltime of the seismic wave between the baseline survey and monitor survey in the reservoir.It is necessary to perform the time-shift correction before the time-lapse AVO inversion,and the correction effect is directly related to the accuracy of the inversion.The second is the research on the seismic wavelet extraction.The seismic wavelet is the basis of the inversion.The accuracy of the wavelet is directly related to the accuracy of the inversion.The third is the research on the time-lapse joint AVO inversion based on the exact Zoeppritz equations.This method can be used to estimate the time-lapse changes in the elastic parameters,namely P-wave velocity,S-wave velocity,density,and their changes.The fourth is the research on the time-lapse joint AVO inversion based on the theory of twophase media.This method can be used to estimate the time-lapse changes in the reservoir parameters,namely porosity,clay content,water saturation,effective pressure,and their changes.The main research achievements of the thesis include four aspects.1)The thesis developes the method of the time-shift correction using the least-square method based on the Gauss-Newton algorithm.The thesis applies the optimization inversion method to estimate the time shifts.First,the objective function with respect to the time shifts is constructed based on the principle of least squares.Then,the Gauss-Newton algorithm is used to estimate the time shifts.Using the time shifts,the time-shift correction is carried out by using the interpolation method.2)The thesis developes the method of the seismic wavelet extraction based on the particle swarm optimization algorithm.The thesis applies the deterministic method for the seismic wavelet extraction.The seismic wavelet extraction is regarded as an inversion problem.The objective function with respect to the seismic wavelet is constructed based on the principle of least squares.Then,the particle swarm optimization algorithm is used to estimate the seismic wavelet.3)The thesis proposes the method of the time-lapse joint AVO inversion based on the exact Zoeppritz equations.First,the equation for the time-lapse AVO forward modeling is obtained by using the exact Zoeppritz equations and the convolutional model.Secondly,the objective function with respect to the elastic parameters is constructed based on the principle of least squares,and the low-frequency constraint term and smoothing term are also added to the objective function to improve the inversion results.Thirdly,the first-order Taylor expansion formula is used to linearize the objective function,and the equation for estimating the elastic parameters can be derived.Finally,based on the equation for estimating the elastic parameters and combined with the idea of time-lapse joint AVO inversion,the equation of time-lapse joint AVO inversion can be derived.It can be used to estimate the elastic parameters and their time-lapse changes simultaneously,namely P-wave velocity,S-wave velocity,density,and their changes.4)The thesis proposes the method of the time-lapse joint AVO inversion based on the theory of two-phase media.First,the equation for the time-lapse AVO forward modeling is obtained by using a reflection-coefficient equation in two-phase media,a rock-physical model,and the convolutional model.Secondly,the Bayesian theory is used to construct the objective function with respect to the reservoir parameters.The reservoir parameters follow a four-variable Cauchy prior distribution,and the modelconstraint term is also added to the objective function to compensate the low-frequency information.Thirdly,the Gauss-Newton algorithm is used to obtain the equation for estimating the reservoir parameters.Finally,based on the equation for estimating the reservoir parameters and combined with the idea of time-lapse joint AVO inversion,the equation of time-lapse joint AVO inversion can be derived.It can be used to estimate the reservoir parameters and their time-lapse changes simultaneously,namely porosity,clay content,water saturation,effective pressure,and their changes.The innovation points of the thesis include two aspects.1)The thesis proposes the method of the time-lapse joint AVO inversion based on the exact Zoeppritz equations.It overcomes the limitations of the traditional method based on the linear approximate expression of the Zoeppritz equations.In addition,the nonlinear problem is linearized in the inversion.The calculation cost is small and it is easily implemented.2)The thesis proposes the method of the time-lapse joint AVO inversion based on the theory of twophase media.The reflection-coefficient equation in two-phase media is used in the inversion.The properties of two-phase media are considered in the reservoir,and the theory is more accurate.In addition,the reservoir parameters and their changes are estimated directly using the time-lapse seismic records.It can avoid the introduction of intermediate parameters and eliminate cumulative errors.