饱和致密砂岩中动态剪切模量硬化的实验研究与理论模拟

Gassmann理论认为岩石的剪切模量在饱和流体前后保持不变,这一认识被广泛应用于高孔高渗常规储层中.然而,致密砂岩等非常规储层通常具有低孔、低渗以及孔隙结构复杂等岩石物理特征,因此Gassmann流体替换理论在此类储层的适用性尚不明确.针对这一问题,本文在1～60 MPa有效压力内分别测量了干燥与饱水致密砂岩样品的超...

Experimental study and theoretical simulation of dynamic shear modulus hardening in saturated tight sandstone

According to Gassmann theory, the shear modulus of rock remains unchanged before and after saturation, which is widely used in conventional reservoirs with high porosity and high permeability. However, tight sandstone and other unconventional reservoirs usually have low porosity, low permeability and complex pore structure, therefore the applicability of Gassmann fluid substitution theory in such reservoirs is not clear. In order to solve this problem, ultrasonic P-wave and S-wave velocities of dry and saturated tight sandstone samples were measured under the effective pressure of 1~60 MPa, and the variation characteristics of shear modulus of sandstones before and after water saturation were analyzed. The results show that the shear modulus of tight sandstone may weaken or harden before and after water saturation. Through the quantitative analysis of the pore aspect ratio of tight sandstone samples, it is found that samples of micro fractures mainly distributed between particles and within particles, with wide distribution range of soft pore aspect ratio and high content of soft pore are more likely to show the characteristics of shear hardening. The general squirt flow model can be used to simulate the shear hardening phenomenon, but the influence of micro crack closure on the input parameters should be considered. Shear hardening is mainly due to the dispersion of squirt flow excited by high frequency elastic wave. Considering the frequency dispersion may also occur in seismic and logging frequency bands, it is necessary to realize that the shear modulus may change during fluid replacement in unconventional reservoirs such as tight sandstone. Blind using of the Gassmann theory with assumption that shear modulus does not change in fluid replacement may cause large prediction error.