DENSITY-SENSITIVITY STUDY ABOUT NONLINEAR MULTIPARAMETER PRESTACK INVERSION BASED ON THE EXACT ZOEPPRITZ EQUATION

摘要

Density information is of critical in identifying the presence and estimating the saturation of hydrocarbon in the reservoir. However, density is difficult to estimate based on conventional prestack inversion approach. In addition, prestack inversion is a nonlinear and ill-posed problem. In order to alleviate the ill-posedness and obtain reliable density information, we propose a nonlinear multiparameter prestack inversion method by constructing the edge-preserving regularized objective function based on anisotropic Markov random fields; we make an attempt to directly use the exact Zeoppritz equation and employ fast simulated annealing algorithm to solve the nonlinear optimization problem. Numerical analysis indicates that density parameter contributes relatively greater on altering the reflectivity magnitude at small incidence angle. In 2D synthetic test, we employ different gathers with specified angle range to test the inverted results and analyze the sensitivity of density. The synthetic results demonstrate that we can obtain reliable density result with small incidence angle by the proposed inversion method. In addition, density results can be improved by scaling the regularization term of density. The inverted density from the field data reveals detailed structural information and shows good agreement with the logging curves; the satisfactory density inverted results from small angle gather validates the conclusion from the numerical results.