基于贝叶斯理论的孔隙流体模量叠前AVA反演

Prestack AVA inversion for pore fluid modulus based on the Bayesian theory

孔隙流体模量是一种极其敏感的流体指示因子,对该参数进行地震尺度的估算在含油气储层识别中具有重要的意义。首先在孔隙弹性介质理论的指导下,推导了包含孔隙流体模量的Zoeppritz方程的线性近似方程;然后在贝叶斯理论框架下,假设先验分布服从四元柯西分布、似然函数服从高斯分布,建立了包含正则化约束的叠前AVA反演方法。模型试算结果表明,利用该方法估算孔隙流体模量,能够提高流体识别的精度;实际资料的试应用取得了较好效果,证明了该方法的有效性和实用性。

Pore fluid modulus is an extremely sensitive fluid indicator.It is significant to estimate pore fluid modulus for reservoir fluid discrimination in seismic scale.Based on the poroelastic theory,a Zoeppritz approximation equation containing the pore fluid modulus is derived.Then in the theoretical framework of Bayesian theory,we assume that the prior distribution can be modelled via the four-variable Cauchy distribution and the likelihood function follows the Gaussian distribution.The objective function of prestack AVA inversion is constructed containing the regularized constraint.A synthetic data test shows that the estimation of pore fluid modulus with this method improves the accuracy of fluid identification,and the application results of actual seismic data demonstrates its effectiveness and practicability.