Probabilistic reservoir-properties estimation for anisotropic shales using statistical rock physics and seismic data

Zhang, Bing (Jilin University) | Liu, Cai (Jilin University) | Guo, Zhiqi (Jilin University) | Lu, Neng (Jilin University) | Liu, Xiwu (Sinopec and National Key Laboratory of Corporation of Shale Oil/Gas Enrichment Mechanism and Effective Development)


A stochastic inversion method of reservoir properties for anisotropic shales is proposed by combing rock physics model and Bayesian estimation. Quantitative relations between elastic parameter such as P- and S-wave impedances and reservoir properties including fracture and porosity are investigated using the statistical rock physics model. During the modeling, the error between the rock physics model and reservoirs, as well as noises in the seismic data are considered. For the process of estimating reservoir properties from elastic parameters, Bayesian inversion based on statistical rock physics model is applicable to the uncertainty problem by computing the posterior probability distribution (PDF) of the reservoir properties. Based on rock physics modeling and given prior knowledge of the reservoir, reservoir properties are obtained by the maximum a posteriori (MAP) criterion and associated uncertainty analysis. In the stochastic inversion, the SA-PSO algorithm which combines the simulated annealing method and the particle swarm optimization method shows its advantages in accuracy and efficiency. The estimated reservoir properties can be used for better characterizations of the sweet spots in shale reservoirs.

This paper has been withdrawn from the Technical Program and will not be presented at the 87th SEG Annual Meeting.