岩石物理驱动的储层裂缝参数与物性参数概率地震反演方法

长波长假设条件下，各向同性背景地层中发育一组平行排列的垂直裂缝可等效为具有水平对称轴的横向各向同性（HTI）介质.基于不同观测方位的岩石地震响应特征变化，宽方位地震数据不仅可实现裂缝岩石弹性参数与各向异性参数的预测，同时也蕴含着丰富的孔隙度等储层物性参数信息.本文结合实际地震资料提出了贝叶斯框架下岩石物理驱动的储层裂缝参数与物性参数概率地震联合反演方法，首先基于AVAZ反演裂缝岩石的弹性参数与各向异性参数，并在此基础上通过统计岩石物理模型表征孔隙度、裂缝密度等各向异性介质储层参数与裂缝岩石参数的相互关联，并采用马尔科夫链蒙特卡洛（MCMC）抽样方法进行大量样本的随机模拟，使用期望最大化（EM）算法估计后验条件概率分布，最终寻找最大后验条件概率对应的孔隙度、裂缝密度等HTI裂缝介质储层参数即为反演结果.测井及实际地震数据处理表明，该方法能够稳定合理地从方位地震资料中获取裂缝岩石弹性参数与各向异性参数，并提供了一种较为可靠的孔隙度、裂缝密度等裂缝介质储层参数概率地震反演方法.

Probabilistic seismic inversion for reservoir fracture and petrophysical parameters driven by rock-physics models

A single set of vertically aligned fractures embedded in a purely isotropic background medium may be considered to be a long-wavelength effective transversely isotropic medium with a horizontal symmetry axis (HTI). The wide-azimuth seismic data can be used to invert the elastic and anisotropic parameters on the subsurface by observing variation in subsurface seismic response along different azimuths, which also contains the abundant information of reservoir property parameters such as porosity. In this paper, a method of probabilistic seismic joint inversion for reservoir fracture and petrophysical parameters driven by rock-physics models is proposed. Firstly, the elastic and anisotropic parameters of fractured rocks are derived based on the amplitude variation with angles of incidence and azimuth (AVAZ). Then a statistical rock-physics model is built to characterize the interrelationship between the reservoir petrophysical parameters, such as porosity and fracture density, and the fractured parameters, and then we create a large number of samples of stochastic simulation using the Markov chain Monte Carlo (MCMC) method. The expectation maximization (EM) algorithm is used to estimate the a posteriori probabilistic density function (PDF), and finally the inversion results of the anisotropic reservoir parameters are found to be the place that the maximum a posteriori PDF is. Tests on well log data and seismic data validate that the method proposed in this paper can be used for stably and reasonably deriving the elastic and anisotropic parameters of fractured rocks, and provides a reliable method of probabilistic seismic inversion for reservoir fracture and petrophysical parameters, such as porosity and fracture density.