The Study On Seismic Scattering Waves In Random Heterogeneous Media

In complex seismic exploration fields,the subsurface media are rich in small-scale heterogeneities,such as carbonate cavity-fractured reservoirs,Fluvial sand channels and permafrost regions with gas hydrates.These small-scale heterogeneities are randomly distributed and too small to be resolved by conventional seismic imaging.Coda waves are generated when the seismic waves propagate through the random media and multiple scattering occurs.The phases of Coda waves are random,so Coda waves are usually regarded as noises,which are removed or ignored in the conventional seismic exploration fields.However,the researches demonstrate that the energy(envelopes)of the scattered waves depend on the stochastic properties of the random heterogeneous media.Thus,the stochastic parameters of small-scale heterogeneities,such as characteristic length,representing the scales of small heterogeneities,and fluctuation strength,that characterizes the strength of velocity fluctuation,can be estimated from the envelopes of seismic scattering data.The work is essential and worthy,because it's useful to better understand and characterize the subsurface media.In addition,it's necessary to assess the oil or gas total volume in the heterogeneous reservoir.This paper focuses on the modeling of seismic scattering waves in random heterogeneous media and the estimation of stochastic parameters of the random media.Firstly,the random model theory is briefly introduced,which is the characterization of the small-scale heterogeneous media.The characteristics of the Gaussian?Exponential and von Karman random model often used in geology are discussed.Then,the radiative transfer theory is studied,which simulates the seismic scattering energy in the random medium.The acoustic radiative transfer equation is given,and the Monte-Carlo simulation to solve the equation is provided.In to make the modeling method more practical,the interface boundary in the multiple-layer media is derived,and the corresponding strategy in Monte-Carlo simulation is solved.To validate the Monte-Carlo radiative transfer method,the solutions of the acoustic Monte-Carlo radiative transfer method and these calculated with finite difference method are compared for several designed random models.The results show that the solutions computed with these two method match well.Meanwhile,the Monte-Carlo radiative transfer simulation costs less time than the finite difference simulation.Thus,it is a good option to apply Monte-Carlo radiative transfer simulation to model the energy transport of seismic scattering waves in random heterogeneous media.Finally,the data of seismic scattering waves are applied to extract the stochastic parameters in random media-characteristic length and fluctuation strength of the random media.In this paper,the grid scan technique based on Monte-Carlo radiative transfer method is used to invert the stochastic parameters.The seismic crosswell modeling test is used to investigate and study the inversion method.The objective function is modified to properly balance the energy effect of direct waves and coda waves in the inversion process.Revision of this objective function makes the inversion result more accurate and more stable.Furthermore,a physical experiment is carried out to obtain the transmitted seismic data for the random heterogeneous sample,and the estimation of the fluctuation strength and the characteristic length of the artificial random media are achieved.The estimated results of the heterogeneity scale and strength match well with true values,indicating that the inversion for stochastic parameters based on MRCT method is feasible and valid.