Study on selection and direct inversion method of brittleness index for shale reservoir

• Published in: Acta geophysica 2023-04, Vol.71 (2), p.757-768
• Main Authors: Qin, Ziyu, Wen, Xiaotao, Pan, Shulin, Chen, Jing-yi, Gou, Qiyong
• Format: Article
• Language: English
• Subjects: Algorithms; Boreholes; Brittleness; Earth and Environmental Science; Earth Sciences; Geophysics/Geodesy; Geotechnical Engineering & Applied Earth Sciences; Organic matter; Porosity; Research Article - Applied Geophysics; Structural Geology
• ISSN: 1895-7455; 1895-6572; 1895-7455
• DOI: 10.1007/s11600-022-00923-3

The brittleness index can be indirectly converted by elastic parameters which are obtain by pre-stack amplitude variation with offset inversion and extended to the whole work area. However, indirect conversion will bring cumulative errors. In order to improve the accuracy of obtaining the brittleness index, the exact Zoeppritz equation including different brittleness indices is derived. Before inversion, we analyzed the characteristics of the brittleness index under the changes of brittle minerals, porosity and organic matter content through rock physics model, and selected the brittleness index most suitable for the work area. Based on the Bayesian framework, we introduce the Limited-memory Broyden–Fletcher–Goldfarb–Shanno (L-BFGS) method to invert theoretical and actual data. Theoretical data inversion results demonstrate this method can achieve good results in both PP- and PS-wave joint inversion as well as only PP-wave inversion. To further verify the effectiveness of the algorithm, the brittleness index of actual data is directly inverted by using the studied algorithm and process. The inversion results of the borehole-side trace are in good agreement with the brittleness index calculated by log data. Both theoretical and practical data prove the feasibility of our proposed method.