Model Parameterization and P-wave AVA Direct Inversion for Young's Impedance
Model Parameterization and P-wave AVA Direct Inversion for Young's Impedance
复制标题
杨氏阻抗的模型参数化和 P 波 AVA 直接反演
DOI：
10.1007/s00024-017-1529-7
复制
发表时间：
2017
影响因子：
2
通讯作者：
Yin Xingyao
中科院分区：
文献类型：
--
作者：
Zong Zhaoyun;Yin Xingyao
AVA反演是弹性参数估计的重要工具，可以指导岩性预测和油气藏“甜点”识别。杨氏模量与密度的乘积（本研究中称为杨氏阻抗）被认为是非常规油气藏的有效岩性和脆性指标。密度很难从地震数据预测，这使得传统方法中杨氏阻抗的估计不准确。在这项研究中，一个实用的地震AVA反演方法，只有P波叠前地震数据估计的杨氏阻抗，以避免密度带来的不确定性。首先，在线性化的纵波和横波模量近似反射率方程的基础上，根据纵波模量、横波模量、杨氏模量与泊松比之间的关系，推导了杨氏阻抗近似反射率方程。将该方程与精确Zoeppritz方程和线性化的纵波近似反射率方程在纵、横波速度和密度方面进行了比较，结果表明，当入射角在临界角以内时，该方程具有足够的精度，可用于AVA反演。参数敏感性分析表明，杨氏阻抗和密度之间的高度相关性，使杨氏阻抗的估计困难。因此，在实际的AVA反演中使用去相关方案，利用贝叶斯推断来仅利用叠前P波地震数据估计杨氏阻抗。仿真算例表明，该方法能够在中等噪声条件下稳定地预测杨氏阻抗，现场数据算例验证了该方法在杨氏阻抗估计和“甜蜜点”评价中的有效性。
AVA inversion is an important tool for elastic parameters estimation to guide the lithology prediction and “sweet spot” identification of hydrocarbon reservoirs. The product of the Young’s modulus and density (named as Young’s impedance in this study) is known as an effective lithology and brittleness indicator of unconventional hydrocarbon reservoirs. Density is difficult to predict from seismic data, which renders the estimation of the Young’s impedance inaccurate in conventional approaches. In this study, a pragmatic seismic AVA inversion approach with only P-wave pre-stack seismic data is proposed to estimate the Young’s impedance to avoid the uncertainty brought by density. First, based on the linearized P-wave approximate reflectivity equation in terms of P-wave and S-wave moduli, the P-wave approximate reflectivity equation in terms of the Young’s impedance is derived according to the relationship between P-wave modulus, S-wave modulus, Young’s modulus and Poisson ratio. This equation is further compared to the exact Zoeppritz equation and the linearized P-wave approximate reflectivity equation in terms of P- and S-wave velocities and density, which illustrates that this equation is accurate enough to be used for AVA inversion when the incident angle is within the critical angle. Parameter sensitivity analysis illustrates that the high correlation between the Young’s impedance and density render the estimation of the Young’s impedance difficult. Therefore, a de-correlation scheme is used in the pragmatic AVA inversion with Bayesian inference to estimate Young’s impedance only with pre-stack P-wave seismic data. Synthetic examples demonstrate that the proposed approach is able to predict the Young’s impedance stably even with moderate noise and the field data examples verify the effectiveness of the proposed approach in Young’s impedance estimation and “sweet spots” evaluation.