公开(公告)号：US20250131284A1

公开(公告)日：2025-04-24

申请号：US18916824

申请日：2024-10-16

Applicant: Schlumberger Technology Corporation

Inventor： Valerian Guillot ,  Arvind Sharma ,  Mohamed Khalil Nafi ,  Jing Wu ,  Alexey Ruzhnikov

IPC: G06N3/096 ,  G06N3/0455 ,  G06N3/0475

Abstract: A method may include receiving a description of an event occurring at a wellsite; extracting a failure mode from the description using a fine-tuned large language model (LLM); identifying a matching failure mode from historical data processed using the fine-tuned LLM, where the matching failure mode is associated with one or more remedial actions that successfully resolved the matching failure mode; and outputting the one or more remedial actions for implementation at the wellsite.

公开(公告)号：US20250164656A1

公开(公告)日：2025-05-22

申请号：US18832534

申请日：2023-06-23

Applicant: Schlumberger Technology Corporation

Inventor： Adam Niven Shumaker ,  Arvind Sharma

IPC: G01V1/28

Abstract: A method includes receiving field seismic data that represents a subsurface, identifying features in the field seismic data using a machine learning model that was trained using at least one first synthetic seismic data set that includes one or more features and one or more labels of the features, and at least one second synthetic seismic data set, the first and second synthetic seismic data sets both generated based on a geological model. Noise is injected into the second synthetic seismic data based on the geological model. The method also includes generating a model of the subsurface based at least in part on the features that were identified in the field seismic data using the machine learning model.