Research On Frequency-dependent Seismic Inversion Prediction Method Of Fluid Mobility Of Hydrocarbon Reservoirs

Reservoir prediction based on seismic information has always been one of the research hotspots in the field of geophysical oil and gas exploration.The reasons are the particularity of the research object,the complexity of the underground burial conditions and the multi-solution of the corresponding geophysical mathematical features.However,the accuracy of oil and gas prediction is not high.It is imperative to adopt an effective reservoir prediction method.The mechanism of significant attenuation and dissipation of seismic waves caused by fluid flow in heterogeneous porous media provides a possible opportunity to estimate fluid mobility from seismic data.Fluid mobility is defined as the ratio of the permeability of the reservoir to the viscosity coefficient of the fluid.It comprehensively reflects the interaction of permeability and pore fluid in reservoir rock skeleton.The current calculations on fluid mobility are mostly based on time-frequency decomposition methods,but the resolution is difficult to meet practical application requirements.Furthermore the petrophysical significance of the above method based on the fluid mobility attribute is not clear.In this paper,the low-frequency asymptotic analysis theory of Biot's porous media model was discussed in detail.Based on this theory,the petrophysical model is designed to analyze the petrophysical parameters affecting the frequency-dependent reflection coefficient.After that,the sensitivity of fluid mobility was evaluated.Then the target functional of fluid mobility frequency-dependent inversion under Bayesian theory framework is derived.The direct inversion calculation method of fluid mobility is proposed.The feasibility of the method is tested by one-dimensional model and actual data.Although the stability of conventional sparse inversion in time domain is good,the formation resolution is limited.Frequency domain inversion benefits from the selection of favorable frequency bands,and the resolution is improved.However the inversion results are vulnerable to noise.The basis pursuit inversion method has obvious features for inversion stratigraphic boundary and has better resolution.Therefore,this paper combines the time-frequency domain inversion method and the basis pursuit algorithm to develop a time-frequency joint domain based on basis pursuit inversion method.Through the model and actual data processing,it proves that the method improves the accuracy of P-wave impedance inversion and reduces the multi-solution.Moreover,we use the basis pursuit algorithm to perform frequency-dependent inversion of fluid mobility.The actual data test proves the feasibility of the method in oil and gas reservoir prediction.The instantaneous phase of seismic signals contains a wealth of geological information,but it has not been fully utilized in actual production.In order to utilize the phase information reasonably,the frequency-dependent inversion of fluid mobility is combined with the phase information for the detection of hydrocarbon reservoirs.The actual data processing results indicate that the method reduces the uncertainty of seismic data interpretation with higher applicability,which provides a new idea for oil and gas detection.