Stochastic seismic inversion can integrate diverse datasets to estimate the spatial distribution of subsurface elastic properties. High-resolution stochastic inversion results are significant in the development stage of an oil field. Nevertheless, existing statistical inversion approaches are commonly restricted by the heavy calculation burden. To address this issue, we propose a strategy to accelerate stochastic inversion. Based on a Bayesian linearized inversion theory, we propose a feasible and efficient stochastic inversion. Using the proposed method, we can not only obtain as good stochastic inversion results as the conventional stochastic inversion methods, but also avoid the heavy calculation burdens of forward simulation and computation the inverse of a complex kernel matrix. We test this method by a section of field data, and compare it with the conventional stochastic inversion method. The test result illustrates the effectiveness of this method.
Presentation Date: Monday, October 12, 2020
Session Start Time: 1:50 PM
Presentation Time: 2:15 PM
Location: Poster Station 9
Presentation Type: Poster
