US 12,468,057 B2
Detection of hydrocarbon presence in subsurface from seismic images using relational learning
Kuang-Hung Liu, Basking Ridge, NJ (US); Huseyin Denli, Basking Ridge, NJ (US); Mary Johns, Houston, TX (US); and Jacquelyn Daves, Houston, TX (US)
Assigned to ExxonMobil Technology and Engineering Company, Spring, TX (US)
Appl. No. 18/027,266
Filed by ExxonMobil Technology and Engineering Company, Spring, TX (US)
PCT Filed Sep. 13, 2021, PCT No. PCT/US2021/071435
§ 371(c)(1), (2) Date Mar. 20, 2023,
PCT Pub. No. WO2022/061329, PCT Pub. Date Mar. 24, 2022.
Claims priority of provisional application 62/706,942, filed on Sep. 18, 2020.
Prior Publication US 2023/0375735 A1, Nov. 23, 2023
Int. Cl. G01V 1/30 (2006.01); G01V 1/28 (2006.01)
CPC G01V 1/301 (2013.01) [G01V 1/282 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for detecting geological elements or fluid in a subsurface from seismic images, the method comprising;
accessing different sets of seismic data;
performing unsupervised machine learning on the different sets of seismic data in order to learn one or more relational properties between the different sets of seismic data;
interpreting the one or more relational properties using reference data points to identify one or both of the geological elements or the fluid; and
performing one or more actions in the subsurface based on the identified one or both of the geological elements or the fluid;
wherein the unsupervised machine learning comprises variational machine learning (VRL).