摘要
常规AVO反演主要针对各向同性介质进行,含有裂隙的地层实际上是各向异性的。地震反演问题是典型的不适定问题,各向异性参数的加入,使得地震反演问题的非线性程度显著增加,比传统反演难度加大。SA-PSO算法将粒子群算法与模拟退火法进行混合优化,具有全局寻优能力更强、克服早熟现象、不易陷入局部极小值等优点。将含有垂向裂隙的介质近似为HTI介质,基于HTI介质近似反射系数方程,建立关于各向异性参数的目标函数,通过SA-PSO算法实现HTI介质各向异性参数反演,进一步,利用流体因子、裂隙弱度与各向异性参数的关联,对裂隙相关参数及流体因子进行预测。建立含有HTI型煤层的理论模型,对原始数据同时使用SA-PSO算法与PSO算法进行反演,对算法的优越性进行测试,SA-PSO算法反演精度更高,反演过程更稳定;对多层理论模型数据分别加上信噪比为5,10,20的高斯随机噪声,对反演方法的抗噪性进行测试,反演结果误差较小,反演过程能快速收敛,反演抗噪性较好;根据HTI型含水裂隙中各向异性参数的特点,对Marmousi2模型的含水砂岩部分进行改造,引入各向异性参数,通过反演获取速度、密度及各向异性参数剖面,并实现流体因子预测,反演及预测结果与理论模型吻合,论证了方法的有效性。
Conventional AVO inversion is mainly carried out for isotropic media,but the strata with fractures are actually anisotropic.Seismic inversion is a typical ill-posed problem.With the addition of anisotropic parameters,the nonlinear degree of seismic inversion is increased significantly,which makes it more difficult than traditional inversion.The SA-PSO algorithm is a hybrid optimization of particle swarm optimization and simulated annealing algorithm,which has the advantages of stronger global optimization ability,overcoming precocity phenomenon and not easy to fall into local minimum value.In this paper,the medium containing vertical cracks is approximated as HTI medium.Based on the HTI medium reflection coefficient equation,the objective function of anisotropic parameters is established.Anisotropic parameter inversion is realized by SA-PSO algorithm.Furthermore,the fracture related parameters and fluid factors are predicted based on the correlation between the fluid factor,fracture weakness and anisotropy parameters.The theoretical model of coal seam containing HTI type is established,and the original data are inverted by SA-PSO algorithm and PSO algorithm at the same time.The superiority of the algorithm is tested.The inversion accuracy of SA-PSO algorithm is higher and the inversion process is more stable.By adding Gaussian random noise with SNR of 5,10,20 to the data of multi-layer theoretical model,the anti-noise performance of the inversion method is tested.The inversion results show that the error of the inversion results is small,the inversion process can converge quickly,and the anti-noise performance of the inversion is good.According to the characteristics of anisotropic parameters in the HTI type water-bearing fracture,the water-bearing sandstone part of Marmousi2 model is modified,and anisotropic parameters are introduced.The velocity,density and anisotropic parameter profiles are obtained through inversion,and the fluid factor prediction is realized.The inversion and prediction results are consistent with the theoretical model,which demonstrates the effectiveness of the method.
作者
李勤
王玮
王瀚林
LI Qin;WANG Wei;WANG Hanlin(School of Geology and Environment,Xi’an University of Science and Technology,Xi’an 710054,China)
出处
《煤炭学报》
EI
CAS
CSCD
北大核心
2021年第6期1925-1935,共11页
Journal of China Coal Society
基金
国家自然科学基金资助项目(41674135)。
